Underground coal machine working condition monitoring system and method based on running state

Abstract

This invention discloses an underground coal mining machinery operating condition monitoring system and method based on operational status, specifically relating to the field of coal mining machinery operating condition monitoring technology, and is used to solve the problem of poor identification of damage caused by abnormal operating conditions. By sampling, aligning, and normalizing the state signals of underground coal mining machinery, a high-precision time-series tensor input basis is constructed. A feature vector sequence is constructed by combining dynamic response gradient, disturbance coupling, and abnormal activity index. A gated cyclic structure model is used to realize dynamic prediction of coal mining machinery operating conditions, construct the mapping relationship between operating conditions and unit time output loss, energy consumption increase, and scheduling delay, quantify the economic impact of future operating conditions, generate an economic loss trend sequence, identify high loss risk windows, and then mine the state propagation chain and synchronous mutation probability based on the scheduling dependency graph to screen intervention points and formulate multi-path intervention strategies, evaluate their implementation costs and response effects, and improve the intelligent perception, prediction, and scheduling intervention capabilities of coal mining machinery in complex underground environments.

Description

Technical Field

[0001] This invention relates to the field of coal mining machinery condition monitoring technology, and more specifically, to an underground coal mining machinery condition monitoring system and method based on operating status. Background Technology

[0002] Coal mining machinery, also known as coal mining machines, is the core equipment for underground coal mining operations. It is mainly used for cutting, loading, and transporting coal seams. As coal mines develop towards intelligent and deep-well operations, the underground working environment is becoming increasingly complex. Coal mining machinery operates under conditions of high humidity, high temperature, high dust, and strong impact loads, often facing unstable factors such as severe vibration, hydraulic fluctuations, and sudden load changes. These complex working conditions can easily lead to equipment wear and efficiency reduction, and even induce safety accidents.

[0003] The shortcomings of existing technologies are as follows: the monitoring of underground coal mining machinery operating conditions mainly relies on static single-point signals for anomaly detection, which cannot reflect the evolution trend of the state in the time dimension, resulting in the inability to provide early warning of potential faults. Moreover, it is mostly based on the physical state of the equipment and lacks a systematic assessment of the economic consequences of changes in operating conditions, such as production capacity loss, energy consumption fluctuations and scheduling delays. This makes it difficult to provide a cost-driven basis for intervention strategies. At the same time, the lack of scheduling behavior based on state propagation relationships causes the intervention measures to be disconnected from the actual sources of impact, which can easily lead to resource allocation imbalance and decreased scheduling efficiency, thus affecting the efficiency of underground coal mining machinery. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an underground coal mining machine condition monitoring system and method based on operating status, so as to solve the problem of poor identification of damage caused by abnormal operating conditions in the above-mentioned background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] The method for monitoring the operating conditions of underground coal mining machinery based on its operational status includes the following steps:

[0007] The operating status signals of underground coal mining machines are collected, and sampling alignment and nonlinear normalization mapping under a unified time reference are performed to construct a structured time series tensor.

[0008] Based on the extraction of dynamic response gradient factor, disturbance coupling factor and abnormal activity factor from time series tensor, a state feature vector of time window is constructed, and the state evolution trend is modeled according to the gated loop structure to identify the working condition category of coal mining machine in future time period.

[0009] Based on the operating condition category and the historical response experience of the equipment, establish the correspondence between the operating condition category and the unit time output loss, energy consumption increase, and scheduling delay factor, generate a set of economic impact factors and bind them to the prediction time node.

[0010] The total economic loss value for each time window is calculated based on three types of economic factors, and the economic loss trend sequence is summarized to identify high-risk windows and mark potential intervention areas in combination with the working condition category.

[0011] A scheduling dependency graph is constructed based on the economic loss trend sequence and high-risk window to determine the synchronous mutation probability nodes, screen out intervention points, generate intervention strategy paths, and evaluate the intervention strategy paths.

[0012] The results of intervention strategies, status trend charts, and economic losses are visualized, and the strategy recommendations are deployed and executed in the form of task orders.

[0013] In a preferred embodiment, the operating status signal of the underground coal mining machine is acquired, and sampling alignment and nonlinear normalization mapping under a unified time reference are performed to construct a structured time series tensor. The specific process is as follows:

[0014] The sensor group includes a three-phase current acquisition unit installed at the output end of the cutting motor, a pressure transmitter and temperature sensor arranged in the front and rear chambers of the hydraulic propulsion cylinder, a triaxial vibration sensor installed on the drum structure, and a speed encoder integrated into the propulsion mechanism.

[0015] The acquisition frequency of all sensor channels is uniformly set to the period synchronized with the clock signal of the coal mining machine control host, and time interpolation is used for completion and synchronization correction.

[0016] A nonlinear amplitude warping strategy with interval compression is used to map and transform all sensor channels;

[0017] After amplitude normalization, a state tensor sequence is constructed using a sliding window of fixed time length. All sensor channels are arranged in chronological order to form a state matrix. The rows of the matrix represent the channel numbers, the columns represent the time sequence positions, and the matrix is ​​the time series tensor corresponding to the sliding window.

[0018] In a preferred embodiment, the dynamic response gradient factor, perturbation coupling factor, and abnormal activity factor are extracted based on the time-series tensor, and the specific process is as follows:

[0019] The continuous signal data of the selected channel is scanned, and the numerical change between all two adjacent data points within the entire time window is calculated. The absolute change between each pair of adjacent data points is recorded as an instantaneous gradient value. Within the time window, the maximum value among all instantaneous gradient values ​​is selected as the dynamic response gradient factor of the channel.

[0020] Compare whether the two channels exhibit the same trend at each sampling time, count the number of data point pairs with consistent trends throughout the entire window period, and use the proportion as the perturbation coupling factor for the channel pairs.

[0021] Within the time window, the change in value between two adjacent sampling points is calculated one by one to determine whether it exceeds the disturbance threshold set for the channel. All data points that meet the condition that the change exceeds the disturbance threshold are regarded as a high-frequency disturbance event. The total number of high-frequency disturbance events within the time window is counted and compared with the total number of sampling points within the time window as the abnormal activity factor.

[0022] The dynamic response gradient factor is used to measure the maximum change intensity of a single channel within a time window, representing the degree of abrupt change in the operating state of the component corresponding to the channel;

[0023] The perturbation coupling factor is used to measure the degree of synchronization between different channels;

[0024] The abnormal activity factor is used to measure the proportion of each channel in the high-frequency disturbance region within a time window.

[0025] In a preferred embodiment, a state feature vector for a time window is constructed, and the state evolution trend is modeled based on a gated loop structure to identify the operating condition category of the coal mining machine in the future time period. The specific process is as follows:

[0026] Within each time window, the dynamic response gradient factor, disturbance coupling factor and abnormal activity factor features are calculated separately in each channel or channel pair and then uniformly spliced ​​together to form the state feature vector of the time window.

[0027] After constructing the feature vectors for all time windows, they are combined in chronological order to form a continuous feature sequence, which serves as the input for the dynamic behavior trajectory during the operation of the coal mining machine. The working condition classification model is then used for classification and prediction.

[0028] The working condition classification model uses a sequence modeling algorithm based on a gated loop structure. The model input is a sequence of state feature vectors for several consecutive time windows, and the output is the prediction result of the working condition category of the coal mining machine in the future time period.

[0029] The operating condition categories include normal operation, light load operation, heavy load operation, high frequency vibration, hydraulic fluctuation, and shutdown warning.

[0030] In a preferred embodiment, based on the operating condition category and the equipment's historical response experience, a correspondence is established between the operating condition category and the unit time output loss, energy consumption increase, and scheduling delay factor. This generates a set of economic impact factors and binds them to the prediction time node. The specific process is as follows:

[0031] Historical equipment response experience includes historical equipment data analysis and expert knowledge annotation. Three types of economic factors include unit time output loss factor, unit time energy consumption increase factor, and unit time scheduling delay factor.

[0032] Under normal operating conditions, the standard output value of the coal mining machine per unit time is recorded as the benchmark capacity. Historical data is extracted for the target state, and all window time periods under the corresponding working condition category are selected. The actual average output of the coal mining machine in all time periods is calculated. The average output under the target state is subtracted from the benchmark capacity to obtain the output loss value. The output loss value is converted to the benchmark capacity by ratio to obtain the output loss factor per unit time.

[0033] The total energy consumption per unit time of the coal mining machine under normal conditions is statistically analyzed and used as a benchmark value, including the sum of the power consumption of the cutting motor, the traction system, and the hydraulic system. For the target working condition, the time window data corresponding to the historical conditions is extracted, and the average values ​​of the current, pressure, and temperature channels per unit time are calculated and converted into total energy consumption. The average energy consumption per unit time under the target condition is obtained. The difference between the average energy consumption under the target condition and the total energy consumption under normal conditions is compared, and the difference is converted into a ratio with the benchmark value. The ratio is used as the energy consumption increase factor.

[0034] During the normal operation cycle of the coal mining machine, the average advance distance per unit time, the number of cutting tasks or the production cycle are statistically analyzed and used as a reference for the standard scheduling time series. For the target working condition, historical working condition records are extracted, and the actual task pause time, delay time or task interruption duration caused by the target working condition are statistically analyzed. The average cumulative delay duration per unit time is calculated and used as the unit time scheduling delay factor for the target working condition.

[0035] The predicted operating conditions are bound together, and within each time window, the operating condition category corresponding to the time window and the values ​​of the three types of economic factors obtained by mapping are recorded to generate a sequence of economic impact factors.

[0036] In a preferred embodiment, the total economic loss value for each time window is calculated based on three types of economic factors, and the results are summarized to form an economic loss trend sequence. High-risk windows are identified, and potential intervention areas are marked in conjunction with the operating condition category. The specific process is as follows:

[0037] For each time window, based on the three types of economic factors bound to it, losses are calculated from three dimensions: capacity loss, electricity cost, and dispatch cost, and the sum is taken as the total economic loss value of the time window.

[0038] Using the window number as the time axis, the economic loss values ​​of all time windows are summarized in sequence to construct a complete time series structure, which serves as the economic loss trend series.

[0039] The economic loss values ​​for each time window are arranged in the order of collection, forming a data sequence with time as the horizontal axis and economic loss as the vertical axis.

[0040] Risk thresholds are set based on the loss trend sequence. High quantiles of the economic loss trend sequence are selected as high-risk criteria. The entire economic loss trend sequence is traversed, and window numbers of all economic loss values ​​higher than the risk threshold are extracted, identified as high-risk windows, and marked as potential intervention areas.

[0041] In a preferred embodiment, for each time window, losses are calculated from three dimensions—capacity loss, electricity cost, and dispatch cost—based on the three types of economic factors associated with it, and the sum is taken as the total economic loss value for the time window. The specific process is as follows:

[0042] Based on the unit time output loss factor and normal coal mining quota, calculate the theoretical output value loss caused by the decrease in production capacity during the time window.

[0043] Calculate the additional energy cost generated within the time window based on the energy consumption increase factor per unit time and the electricity price per kilowatt-hour of equipment operation.

[0044] Based on the unit time scheduling delay factor and the scheduling manpower cost benchmark, calculate the manpower and equipment idle cost caused by scheduling delay within the time window;

[0045] The three results are added together to form the total economic loss value within the time window.

[0046] In a preferred embodiment, a scheduling dependency graph is constructed based on the economic loss trend sequence and the high-risk window to determine synchronous mutation probability nodes, screen intervention points, and generate intervention strategy paths. The specific process is as follows:

[0047] Define the set of nodes in the scheduling dependency graph, where each node corresponds to an independent device unit. The set of edges in the scheduling dependency graph is used to represent the scheduling dependency or operational linkage between two nodes, and the direction of the edge indicates the direction of influence.

[0048] The weights on the edges are set to the degree of state correlation, which are calculated based on the probability of synchronous mutation of the two channels in the high-risk window in historical data.

[0049] For each window marked as high risk, analyze its corresponding state feature vector and working condition category sequence to determine the main working condition types that cause economic losses. Using the rise of the economic loss factor as the signal source, expand outward in the graph to track the associated scheduling dependency graph nodes.

[0050] Identify the node group with continuous abnormal responses, and determine the intervention point based on the state propagation priority score and state impact index;

[0051] After identifying the points that can be intervened, an intervention strategy path is generated based on the operation control rules, equipment adjustment parameters, and historical intervention effect records.

[0052] In a preferred embodiment, the intervention strategy path is evaluated, and the specific process is as follows:

[0053] The implementation cost is the combined cost of labor, material costs, and equipment downtime losses required to complete all actions in the intervention strategy path;

[0054] Response latency is the average time interval between the issuance of a strategy and the actual feedback of its effect.

[0055] Historical success rate is the percentage of intervention strategies that have successfully reduced economic losses to below a set threshold in historical operating conditions.

[0056] The underground coal mining machinery condition monitoring system based on operational status is used to implement the aforementioned underground coal mining machinery condition monitoring method based on operational status, including:

[0057] The coal mining machine data acquisition module collects the operating status signals of the underground coal mining machine and performs sampling alignment and nonlinear normalization mapping under a unified time reference to construct a structured time series tensor.

[0058] The working condition category identification and acquisition module extracts dynamic response gradient factors, disturbance coupling factors and abnormal activity factors based on time series tensors, constructs state feature vectors for time windows, and models the state evolution trend according to the gated loop structure to identify the working condition category of the coal mining machine in the future time period.

[0059] The economic impact module establishes a correspondence between operating condition categories and unit time output loss, energy consumption increase, and scheduling delay factor based on operating condition categories and historical equipment response experience, generates a set of economic impact factors, and binds them to the prediction time nodes.

[0060] The risk identification module calculates the total economic loss value for each time window based on three types of economic factors, summarizes them to form an economic loss trend sequence, identifies high-risk windows, and marks potential intervention areas in conjunction with the working condition category.

[0061] The path intervention module constructs a scheduling dependency graph based on the economic loss trend sequence and high-risk windows, determines the synchronous mutation probability nodes, screens intervention points, generates intervention strategy paths, and evaluates the intervention strategy paths.

[0062] The coal mining equipment adjustment module visualizes the results of intervention strategies, status trend charts, and economic losses, and deploys and executes strategy recommendations in the form of task work orders.

[0063] The technical effects and advantages of this invention are as follows:

[0064] This invention constructs a high-precision time-series tensor input foundation by sampling, aligning, and normalizing key state signals such as cutting motor current, voltage, vibration, hydraulic oil pressure, and drum temperature under a unified time reference. It constructs a feature vector sequence by combining indicators such as dynamic response gradient, disturbance coupling, and abnormal activity, and uses a gated cyclic structure model to dynamically predict the coal mining machine's operating conditions, improving the forward-looking identification capability of abnormal operating states. By constructing a mapping relationship between operating conditions and unit-time output loss, energy consumption increase, and scheduling delay, it quantifies the economic impact of future operating states, generates an economic loss trend sequence, effectively identifies high-loss risk windows, and further mines state propagation chains and synchronous mutation probabilities based on scheduling dependency graphs to accurately screen intervention points and formulate multi-path intervention strategies, evaluating their implementation costs and response effects to ensure the feasibility and timeliness of scheduling schemes. Finally, it combines a visualization platform to achieve intuitive display and coordinated deployment of monitoring results and strategies, significantly improving the intelligent perception, prediction, and scheduling intervention capabilities of the coal mining machine system in complex underground environments. Attached Figure Description

[0065] Figure 1 This is a flowchart of the underground coal mining machine condition monitoring method based on the operating status of the present invention.

[0066] Figure 2 This is a schematic diagram of the underground coal mining machinery condition monitoring system based on the operating status of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Example 1: As Figure 1 As shown, the method for monitoring the operating condition of underground coal mining machinery based on its operational status includes the following steps:

[0069] The operating status signals of underground coal mining machines are collected, and sampling alignment and nonlinear normalization mapping under a unified time reference are performed to construct a structured time series tensor.

[0070] Based on the extraction of dynamic response gradient factor, disturbance coupling factor and abnormal activity factor from time series tensor, a state feature vector of time window is constructed, and the state evolution trend is modeled according to the gated loop structure to identify the working condition category of coal mining machine in future time period.

[0071] Based on the operating condition category and the historical response experience of the equipment, establish the correspondence between the operating condition category and the unit time output loss, energy consumption increase, and scheduling delay factor, generate a set of economic impact factors and bind them to the prediction time node.

[0072] The total economic loss value for each time window is calculated based on three types of economic factors, and the economic loss trend sequence is summarized to identify high-risk windows and mark potential intervention areas in combination with the working condition category.

[0073] A scheduling dependency graph is constructed based on the economic loss trend sequence and high-risk window to determine the synchronous mutation probability nodes, screen out intervention points, generate intervention strategy paths, and evaluate the intervention strategy paths.

[0074] The results of intervention strategies, status trend charts, and economic losses are visualized, and the strategy recommendations are deployed and executed in the form of task orders.

[0075] Step 1 involves acquiring underground coal mining machine operation data and constructing a time-series window tensor. This includes collecting multi-channel operating status parameters of the underground coal mining machine's cutting motor, hydraulic propulsion cylinder, drum vibration unit, and the entire machine's motion system. The acquired signals are then time-synchronized and aligned, and their nonlinear amplitudes are normalized to construct a structured time-window tensor. The specific process is as follows:

[0076] A sensor group is formed by deploying various industrial sensors on key components of the coal mining machine to collect multi-channel data on the machine's operating status at a fixed sampling period. The sensor group includes, but is not limited to: a three-phase current acquisition unit installed at the output of the cutting motor to collect cutting current fluctuation parameters; pressure transmitters and temperature sensors deployed in the front and rear chambers of the hydraulic propulsion cylinder to acquire propulsion oil pressure and return oil temperature respectively; a triaxial vibration sensor installed on the drum structure to acquire mechanical impact acceleration values ​​during cutting; and a speed encoder integrated into the propulsion mechanism to collect information on the coal mining machine's forward speed and displacement changes. The acquisition frequency of all sensor channels is uniformly set to a 50-millisecond cycle synchronized with the coal mining machine control host clock signal, meaning one data point is acquired for each channel every 50 milliseconds, with the initial acquisition time aligned with the rising edge of the control host clock signal.

[0077] Using the unified clock signal provided by the coal mining machine's main control system as the reference time source, timestamps are added to all channel-acquired data, and time alignment is performed. The alignment method is as follows: the timestamps collected from different channels are uniformly mapped to the sampling time point of the main control system's clock signal to ensure the timing consistency of data from each channel at the same time point. In the case of timestamp offset, time interpolation is used for completion and synchronization correction.

[0078] The specific method of time interpolation is as follows: If a channel fails to complete effective acquisition at the target time point, the numerical change trend of the two effective data points before and after is used on that channel, and linear extrapolation is performed according to the sampling period ratio to interpolate the fitted value at the target time. For example, in the propulsion oil pressure channel, if the oil pressure data at a certain target time is missing, the pressure value at that point can be fitted using the acquired data within 100 milliseconds before and after, ensuring that there are no gaps in all channels under the same time axis.

[0079] Even after time alignment, the data still have differences in physical dimensions, such as current expressed in amperes, oil pressure in megapascals, vibration in meters per second squared, temperature in degrees Celsius, and velocity in meters per second.

[0080] To eliminate the influence of physical scale and improve the comparability of multi-channel data, a nonlinear amplitude normalization strategy with interval compression is adopted to map and transform all channels. The normalization rule is as follows: based on the steady-state operating range of each channel under standard operating conditions, a center value and a sensitive interval are set, and a nonlinear S-shaped mapping function is constructed to map the original value of each channel to the interval (0, 1). The mapping process retains higher resolution for edge values ​​and maintains symmetrical buffering for the median interval. For example, in the cutting current channel, if the standard operating range is 400 to 800 amperes, the center value is set to 600 amperes, and the boundary value is mapped to a two-sided value close to 0 and 1 after compression.

[0081] After amplitude normalization, a state tensor sequence is constructed using a sliding window of fixed time length. The sliding window length is set to 5 seconds, and the sliding step size is set to 1 second. Each window contains 100 consecutive sampling points (with a sampling period of 50 milliseconds). Within each window, all channels are arranged in chronological order to form a state matrix. The rows of the matrix represent the channel numbers, and the columns represent the time series positions. The matrix is ​​the temporal tensor corresponding to that window. For example, in the nth window, if there are 10 acquisition channels, the window tensor will be 10 rows and 100 columns, representing the complete dynamic change process of each channel within the 5-second time period.

[0082] The periodic acquisition period and sliding window length are set according to the dynamic response characteristics of the coal mining machine's operating conditions. The acquisition period (50 milliseconds) covers the high-frequency vibration change response during typical coal and rock cutting, while the sliding window length (5 seconds) covers the typical propulsion rhythm and hydraulic stabilization time constant range, which can fully capture the data changes of the coal mining machine from low load to high load and from stable cutting to sudden impact.

[0083] Each constructed time-series tensor will serve as the basic input unit for subsequent state recognition modules, arranged sequentially in chronological order to form a multi-channel time-series tensor sequence during operation. All time-series tensor data are cached in the local storage area of ​​the control system in the form of a three-dimensional array, where the first dimension is the time window number, the second dimension is the channel number, and the third dimension is the time point sequence within the window. For example, the storage structure is: window index, channel number, time point number. All tensors are arranged in ascending order of acquisition time, supporting subsequent fast access by time index.

[0084] It should be noted that the calibration of the sensor group is performed in a combination of daily and periodic methods according to the downhole operating conditions. For example, the cutting current sensor is compared with a reference by the control cabinet outputting a fixed current before mining each day, the hydraulic sensor is calibrated on-site at full range every week, and the vibration sensor is tested monthly on a standard vibration table for amplitude and frequency response verification. If the sensor detection error exceeds the set tolerance (e.g., 2%), the system will automatically issue an alarm and switch to the reference channel or execute the interpolation redundancy compensation mechanism.

[0085] Step 2: Extract feature pattern information of the coal mining machine operation process from the constructed multi-channel time-series tensor sequence, and identify the operating condition category in future time periods. The specific process is as follows:

[0086] In each time series tensor generated in step 1, structural features are extracted and dynamic behavior is characterized for the data change trajectories of all acquisition channels. A multi-dimensional state feature vector of the coal mining machine is constructed within the window time period. This feature vector is used to describe the operating state of the coal mining machine within the time window, such as load response, disturbance coupling, vibration fluctuation and structural coordination, providing a highly timely and interpretable feature basis for subsequent working condition judgment.

[0087] The first type is the dynamic response gradient factor, which measures the maximum change intensity of a single channel within a time window, indicating the degree of abrupt change in the operating state of the component corresponding to that channel. For example, in the cutting current channel, if a significant load increase occurs within a certain 5-second window, the current value fluctuation of that channel within the window will increase significantly, and the corresponding dynamic response gradient factor value will also increase, indicating the possible phenomena such as abrupt changes in coal seam hardness or increased cutting resistance;

[0088] The dynamic response gradient factor is calculated as follows: the continuous signal data of the selected channel is scanned, the numerical change between all two adjacent data points within the entire time window is calculated, and the absolute change between each pair of adjacent data points is recorded as an instantaneous gradient value. Within the time window, the maximum value among all instantaneous gradient values ​​is selected as the dynamic response gradient factor of the channel. If the maximum change value exceeds the preset dynamic fluctuation threshold, it indicates that the channel has experienced sudden load, energy shock, or fault precursor behavior within the current window period.

[0089] For example, if a current value is collected every 50 milliseconds within a 5-second window in a cut-off current channel, then the change between adjacent values ​​is calculated among 100 consecutive current values, resulting in 99 gradient values. The one with the largest change is taken as the gradient factor of the channel under that window.

[0090] The second category is the disturbance coupling factor, which measures the degree of synchronous change between different channels and reflects the structural coordination and load transmission behavior between equipment. Multiple channel pairs with physical coupling relationships are selected, such as propulsion hydraulic pressure and forward speed, cutting current and drum vibration, etc. The direction and rate of change of these channels within the window are compared, and the consistency of their disturbance changes is calculated. If multiple coupled channels show an upward or downward trend at the same time, it indicates that their behavior has structural coupling response characteristics; otherwise, it indicates that the coupling is loose or there is a local anomaly.

[0091] A specific method for calculating the disturbance coupling factor is as follows: Select any two channels that are physically linked, such as propulsion hydraulic pressure and coal mining machine forward speed, or cutting current and drum vibration; within this time window, determine the trend of the sampled value sequences of the two channels, that is, determine whether the direction of change of each pair of adjacent data points is upward, downward or unchanged, compare whether the two channels show the same trend at each sampling moment (e.g., rising or falling at the same time), count the number of data point pairs with consistent trends throughout the entire window period, and use their proportion as the disturbance coupling factor of the channel pair. If this proportion is higher than the set coupling threshold (e.g., more than 70%), then the channel pair is considered to have significant disturbance coupling characteristics within this time window.

[0092] For example, if a total of 100 data points are collected for the propulsion hydraulic pressure within a 5-second time window, and the same number of data points are collected for the forward speed, and a trend comparison reveals that the two channels change in the same direction at 78 moments, then the disturbance coupling factor of this channel pair is 78%, which can be judged as a strongly coupled response state.

[0093] The third category is the abnormal activity factor, which is used to measure the proportion of each channel in the high-frequency disturbance region within the window, reflecting the degree of instability of the equipment during operation. By setting the disturbance threshold of each channel, the frequency of signal changes exceeding the threshold within a unit of time can be determined, thereby identifying working condition characteristics such as enhanced cutting vibration, frequent hydraulic fluctuations, and gear meshing impact. When the abnormal activity factor maintains a high value in some channels for a long time, it may indicate potential problems such as component aging or poor lubrication.

[0094] A specific calculation method for the abnormal activity factor is as follows: First, a disturbance threshold is set for each channel. This disturbance threshold is obtained empirically based on the rate of change or fluctuation amplitude of the channel under normal operating conditions. Within a time window, the change in value between two adjacent sampling points is calculated one by one to determine whether it exceeds the disturbance threshold set for the channel. All data points that meet the condition that the change exceeds the disturbance threshold are regarded as a high-frequency disturbance event. The total number of high-frequency disturbance events within the time window is counted and proportionally converted with the total number of sampling points within the window. This proportional value is used as the final result of the abnormal activity factor. The higher the value, the more unstable the channel is within the window period.

[0095] For example, if the roller vibration channel collects 100 data points in a window, and each change amplitude exceeding a certain acceleration threshold is defined as a high-frequency disturbance, and a total of 24 high-amplitude vibration events occur, then the abnormal activity factor of the roller vibration channel in this window is 24%.

[0096] Within each time window, the three types of features mentioned above are calculated in each channel or channel pair and then concatenated to form the state feature vector of that window. The length of the feature vector depends on the number of channels, the number of channel pairs, and the feature type. For example, for 10 channels and 15 channel pairs, the feature vector may contain more than 100 dimensions of information.

[0097] After constructing the feature vectors for all time windows, they are combined in chronological order to form a continuous feature sequence. This sequence is the dynamic behavior trajectory input during the operation of the coal mining machine, which is then entered into the working condition classification model for classification and prediction.

[0098] The operating condition classification model employs a sequence modeling algorithm based on a gated loop structure, enabling it to model the state evolution trend over long time series. The model input is a sequence of state feature vectors for several consecutive time windows, and the output is a prediction of the coal mining machine's operating condition category for a future time period. The operating condition categories are divided into several typical states, including but not limited to normal operation, light load operation, heavy load operation, high-frequency vibration, hydraulic fluctuation, and shutdown warning. Each operating condition category corresponds to key state characteristics in the actual operating behavior of the coal mining machine.

[0099] Taking a specific case as an example, if the cutting current in multiple consecutive windows shows a significant upward trend, and the activity of the vibration channel increases significantly while the oil pressure stability decreases, the model may output a label of heavy-load operation or high impact risk, indicating that the coal mining machine is currently in a complex working condition and attention should be paid to the risk of equipment heat load and structural loss.

[0100] During model training, supervised learning is conducted using a set of working condition samples composed of historical coal mining machine operation data and on-site annotation records to ensure that the model can identify the state change patterns in real production scenarios. After the model training is completed, it is deployed in underground or ground edge servers to achieve real-time inference. Each time a new sliding window data collection cycle is completed, the working condition classification results are automatically updated and a corresponding prediction label sequence is generated.

[0101] The generation cycle of predicted labels is consistent with the time window step size to ensure that the model output keeps pace with data updates.

[0102] Step 3: Based on the identified coal mining machine operating condition categories, construct the correspondence between operating condition status and economic influencing factors, and generate a factor set for economic loss assessment. The specific implementation is as follows:

[0103] In step 2, each time window has been categorized into a specific coal mining machine operating condition category, which identifies the current operating status of the coal mining machine. Each operating condition corresponds to different levels of energy consumption intensity, production capacity fluctuation, and operation and maintenance scheduling intervention costs. These status labels need to be further mapped into measurable economic impact indicators.

[0104] Based on historical equipment response experience (historical equipment data analysis and expert knowledge annotation), a one-to-one correspondence table is established between operating condition categories and their corresponding three main economic factors. The three economic factors are: unit-time output loss factor, unit-time energy consumption increase factor, and unit-time scheduling delay factor, specifically defined as follows:

[0105] The first category is the unit time output loss factor, which is used to measure the degree of output reduction under a certain working condition compared with the ideal working condition (i.e. normal operation). The value of this factor is obtained by statistical regression based on historical production records. For example, if the coal mining machine advance speed decreases significantly under heavy load operation and the coal mining volume per unit time is only 70% of the normal value, then the output loss factor under this condition is set to 30%, which means that the output capacity will be lost every hour, equivalent to 30% of the normal operation.

[0106] The unit-time output loss factor is generated as follows: Under normal operating conditions, the standard output value of the coal mining machine per unit time (e.g., per hour) is statistically analyzed and recorded as the benchmark capacity. Historical data is extracted for the target operating condition (e.g., heavy-load operation), and all window time periods under the corresponding operating condition category are selected. The actual average output of the coal mining machine within these time periods is calculated. Then, the average output under the target condition is subtracted from the normal output to obtain the unit-time output loss of that condition relative to the benchmark. The output loss value is then converted into a ratio with the benchmark capacity to obtain the unit-time output loss factor for that condition. The result is usually expressed as a percentage. If there are significant factors such as a decrease in propulsion speed, drum jamming, or hydraulic instability under that condition, a correction coefficient can be introduced in conjunction with characteristic indicators for manual optimization compensation. For example, if the coal mining machine's hourly coal extraction rate is 500 tons under normal conditions, but the actual average hourly coal extraction rate is only 350 tons under heavy-load operation, then the output loss is 150 tons, and the corresponding unit-time output loss factor is 30%.

[0107] The second category is the unit time energy consumption increase factor, which is used to reflect the increase in energy consumption of the electric and hydraulic systems of the coal mining machine under specific working conditions compared to the normal state. For example, under abnormal vibration conditions, the motor operation is unstable due to component resonance or cutting impact, and energy consumption usually increases significantly. If the current and hydraulic power under this state are increased by an average of 20 percent compared to the normal state, then the corresponding energy consumption increase factor is set to 20 percent.

[0108] The generation method for the unit time energy consumption increase factor is as follows: First, the total unit time energy consumption of the coal mining machine under normal conditions is statistically analyzed and used as a baseline value, including the sum of the power consumption of the cutting motor, traction system, and hydraulic system. For the target operating condition, window data corresponding to the historical conditions is extracted, and the average values ​​of energy consumption-related channels such as current, pressure, and temperature are calculated for each unit time. Then, using the electrical and hydraulic characteristic parameters of the equipment, the above state values ​​are converted into total energy consumption, resulting in the average energy consumption per unit time under the target condition. The difference between the energy consumption under the target condition and the normal baseline energy consumption is compared, and the ratio of this difference to the baseline value is converted into the energy consumption increase factor for that state, expressing the additional burden on energy consumption under that state. For example, if the unit time energy consumption of the coal mining machine is 200 kWh under normal conditions, but increases to 240 kWh under abnormal vibration conditions, the energy consumption increase is 40 kWh, and the corresponding unit time energy consumption increase factor is 20%.

[0109] The third category is the unit-time scheduling delay factor, used to assess the potential time-related impact of this state on work rhythm, task progress, and human resource scheduling. This factor is derived by statistically analyzing the duration of the working condition and the time of subsequent task misalignment. For example, in a shutdown warning state, if it typically triggers 5 to 10 minutes of maintenance downtime or fault handling time, the corresponding scheduling delay factor can be set to cause an average task delay of 8 minutes per hour.

[0110] The unit-time scheduling delay factor is generated as follows: During the normal operation cycle of the coal mining machine, the average advance distance, cutting task quantity, or production cycle time per unit time are statistically analyzed and used as a reference for the standard scheduling time series. For the target working condition, historical working condition records are extracted, and it is statistically analyzed whether the continuous occurrence of this condition causes the advance to stop, maintenance to be inserted, or remote intervention to be performed. The actual task pause time, delay time, or task interruption duration caused by this condition is analyzed, and the average cumulative delay duration within a unit time (e.g., 1 hour) is calculated. This time is used as the unit-time scheduling delay factor for this condition. The result is in minutes or seconds and reflects the time offset impact of the scheduling system caused by the abnormal state. If the state triggers a system-level shutdown warning or linkage protection strategy, a delay amplification weight can be introduced to make the delay factor reflect the actual decision cost. For example, if the coal mining machine needs to be suspended and remote confirmation and manual maintenance operations are performed under the shutdown warning state, resulting in an average pause time of 10 minutes each time, and this state occurs once per hour on average, then the unit-time scheduling delay factor is 10 minutes.

[0111] To ensure the stability and traceability of the mapping relationship, the setting of all factor values ​​must be comprehensively confirmed in conjunction with the following three types of information sources:

[0112] The on-site operation data comes from historical operating condition sequences and corresponding capacity and power consumption records during actual coal mining operations;

[0113] The maintenance system records data from the equipment management system regarding fault response time, processing cycle, and repair time.

[0114] Expert experience and knowledge, which are correction values ​​for manual operation experience given by coal mine operation and maintenance experts, dispatch supervisors and others based on long-term operational experience;

[0115] After the above factors are constructed, they are bound to the working condition categories predicted in step 2. Within each time window, the working condition category corresponding to the time window and the three types of economic factors obtained by mapping are recorded to generate a set sequence of economic impact factors.

[0116] The structure of the economic impact factor set sequence is a three-field dataset arranged by time window number. Each record contains the operating condition category to which the window belongs, the output loss factor, the energy consumption increase factor, and the scheduling delay factor. For example, for a window numbered W13, if its operating condition category is heavy load operation, then the economic factor set corresponding to this window may be [heavy load operation, 30%, 20%, 5 minutes].

[0117] Step 4: Based on the constructed set of economic influencing factors, establish a loss estimation function to quantify the cumulative economic loss of the coal mining machine over a future time period, and identify high-risk economic zones during operation by combining time distribution characteristics. The specific steps are as follows:

[0118] For each time window, based on the three types of economic factors associated with it, losses are estimated from three dimensions: capacity loss, electricity cost, and dispatch cost. The sum of these estimates is taken as the economic loss value for that window. The specific implementation is as follows:

[0119] Based on the unit time output loss factor and normal coal mining quota, calculate the theoretical output value loss caused by the decline in production capacity in this window, and express the result in monetary units; based on the unit time energy consumption increase factor and the electricity price per kilowatt-hour of equipment operation, calculate the additional energy cost generated in this window; based on the unit time dispatch delay factor and dispatch manpower cost benchmark, calculate the manpower and equipment idle cost caused by dispatch delay in this window, and add the above three results to form the total economic loss value in this window;

[0120] For example, if a window is operating under heavy load, its output loss factor is 30%, its energy consumption increase factor is 15%, and its scheduling delay is 5 minutes per hour. Combining the price per ton of coal, the cost of electricity per kilowatt-hour, and the unit price of labor hours, the total loss corresponding to this window can be calculated as a specific monetary amount (such as the loss amount per hour).

[0121] Using the window number as the time axis, the economic loss values ​​of all time windows are summarized sequentially to construct a complete time series structure, which serves as the economic loss trend series. The specific implementation is as follows:

[0122] The economic loss values ​​for each time window are arranged in the order of collection, forming a data sequence with time as the horizontal axis and economic loss as the vertical axis.

[0123] The sequence is smoothed to eliminate random interference from single-point spikes, making it more trend-stable.

[0124] The economic loss curve is generated by connecting the economic loss values ​​over time, showing the economic burden level of the coal mining machine at each stage of the entire operating cycle.

[0125] Analyzing this trend curve can further identify high-loss sections, stable sections, and inflection points.

[0126] Risk thresholds are set based on the loss trend sequence. Typically, the high quantile of the economic loss trend sequence (e.g., 85%) is selected as the high-risk criterion. The entire economic loss trend sequence is traversed, and all window numbers with economic loss values ​​higher than the threshold are extracted and identified as high-risk windows. All high-risk windows are timestamped and jointly recorded with the corresponding working condition category and state feature vector.

[0127] For example, during coal mining on a workday, if the economic losses between 9:00 and 9:45 AM are consistently higher than the set baseline threshold, and are accompanied by frequent occurrences of high vibration and propulsion lag, the system will mark that time period as a potential intervention area.

[0128] Step 5: Based on the generated economic loss trend sequence and the identified high-risk windows, construct a scheduling dependency graph of the operating status of multiple components, and generate a candidate set of scheduling intervention strategies by combining the operating condition category and historical response behavior. Output adjustable task paths and maintenance window clusters. The specific implementation is as follows:

[0129] The scheduling dependency graph is used to represent the dependency structure between various operating subsystems of the coal mining machine in terms of task execution and operating condition influence. Its construction is based on the coal mining machine structure topology, task execution order and operating condition influence flow direction;

[0130] Define the set of nodes in the scheduling dependency graph, including key components and system modules of the coal mining machine, such as cutting motor, drum assembly, propulsion cylinder, hydraulic pump station, traction system, cooling module, etc., with each node corresponding to an independent equipment unit;

[0131] Define the set of edges in a graph to represent scheduling dependencies or operational linkages between two nodes. The direction of the edge indicates the direction of influence. For example, current fluctuations in the cutting motor can affect the load of the hydraulic system, and changes in hydraulic oil temperature can be fed back to the cooling system.

[0132] The weights on the edges are set to the degree of state relevance, calculated based on the synchronous mutation probability of the two channels in the high-risk window in historical data, generating a multi-level linkage structure graph and forming a complete scheduling dependency graph to guide the chain design of intervention strategies.

[0133] After constructing the scheduling dependency graph, it is necessary to further track the state change paths in high-risk windows, identify the propagation chains that lead to increased economic losses, and determine the control points that can be intervened.

[0134] For each window marked as high-risk, analyze its corresponding state feature vector and operating condition category sequence to determine the main operating condition types that cause economic losses;

[0135] In the scheduling dependency graph, the rise of the economic loss factor is used as the signal source, and the graph is expanded outward to track the scheduling dependency graph nodes associated with it.

[0136] The path analysis algorithm is used to determine which paths contain groups of nodes with consecutive abnormal responses, i.e., the existence of state propagation chains.

[0137] Nodes with higher state propagation priority scores and larger state impact indices in the state propagation chain are selected as intervention points to determine whether they have adjustment capabilities or maintenance possibilities.

[0138] All key points are summarized to form a set of actionable intervention entry points.

[0139] The specific implementation of obtaining the synchronous mutation probability is as follows:

[0140] First, a set of time periods identified as high-economic-loss windows in historical records is selected and denoted as the high-risk window set. These windows correspond to periods in the past operation of the coal mining equipment system where abnormal economic indicators such as decreased production capacity, increased energy consumption, or scheduling delays have occurred. They have strong state disturbance characteristics and can therefore serve as key observation samples for identifying state linkage relationships.

[0141] For any two channel parameters to be analyzed (such as cutting motor current and drum temperature), perform the following steps:

[0142] State variation labeling is performed by preprocessing the time series signal of each channel. First, the direction of parameter value change within each high-risk window is calculated, i.e., the local change sign of each window. If the parameter shows a continuous upward or downward trend within the current window, it is labeled as variation; if the change amplitude is small or the trend is not obvious, it is labeled as stable.

[0143] To determine if two channels mutate simultaneously within each high-risk window, if both channels are marked as mutated within the same window, then the channel pair is considered to be a synchronous mutation event within that window.

[0144] Iterate through all high-risk windows, count the number of times the two channels synchronize and mutate across all windows, and divide this number by the total number of high-risk windows to obtain the synchronization mutation probability value of the channel pair.

[0145] The higher the synchronous mutation probability as a boundary weight, the stronger the linkage between the two state channels in the abnormal state. It is suitable to be used as a candidate target for coordinated intervention in subsequent scheduling intervention. For example, if the synchronous mutation probability of hydraulic propulsion pressure and vibration acceleration is significantly higher than that of other channel pairs, then adjusting hydraulic parameters should be given priority as an indirect control method when vibration is abnormal.

[0146] After identifying the points that can be intervened, the system automatically generates multiple feasible intervention strategy paths based on the operation control rules, equipment adjustment parameters and historical intervention effect records, for the scheduling platform to make decisions and deploy.

[0147] The intervention strategy is generated as follows:

[0148] For each critical intervention point, review the equipment control protocol to determine whether it supports functions such as automatic adjustment, running speed limiting, parameter limiting, or temperature control switching.

[0149] Call the historical execution library to extract intervention actions that were successfully executed under similar conditions and their feedback effects, such as "limiting the hydraulic pump differential pressure by 5% can alleviate the rise in oil temperature";

[0150] Based on the current working condition trend prediction results, various intervention actions are combined to form different strategy paths, and a candidate strategy set is constructed.

[0151] Each strategy path includes the target to be adjusted, the timing of execution, the duration, the expected effect, and the rollback conditions to ensure controllability and rollback mechanism;

[0152] Evaluate the implementation costs, response latency, and historical success rates of all candidate strategies, and mark the preferred path.

[0153] Nodes with higher state propagation priority scores and larger state impact indices in the state propagation chain are selected as intervention points to determine their adjustability or maintainability. This requires ranking and screening based on the following two quantitative indicators:

[0154] The state propagation priority score is used to measure the leading role of a node on a state propagation path, and is defined as follows:

[0155] If a node is at the beginning or middle of multiple propagation paths, and its state mutation occurs more frequently than other nodes, then its state propagation priority score is higher. The state propagation priority score is calculated by: counting the percentage of times the node appears as the first mutation node or intermediate transmission node in all observed propagation chains.

[0156] The state impact index is used to measure the contribution of a node's mutation to the final economic loss, and is defined as follows:

[0157] In all historical high-risk windows, calculate the average cumulative loss caused by the mutation of the node, normalize and sort this value among all nodes, and use it as the impact score.

[0158] The intervention criticality score is constructed by weighting the state propagation priority score and the state impact index. For example, the priority weight is set to 0.6 and the impact weight is set to 0.4.

[0159] If the criticality score of the intervention exceeds a preset threshold (e.g., 0.65), and the node is currently in an abnormal state, it is identified as a critical node with priority intervention value.

[0160] After the intervention strategy paths are generated, each path needs to be quantitatively evaluated in the following three aspects:

[0161] The implementation cost is defined as the combined cost of labor, materials, and equipment downtime losses required to complete all actions in the intervention strategy path. For example, if the path includes replacing a hydraulic pump, check the average downtime and cost in its historical maintenance records. The response delay is defined as the average time interval from the issuance of the strategy to the feedback of the actual effect. The historical success rate is the ratio of the economic losses successfully reduced to below a set threshold in similar historical working conditions. Records of the strategy path that have been executed can be extracted from the database, and the success rate can be counted as a percentage of the total number of attempts.

[0162] For example, in a high-risk window, if the hydraulic system malfunctions and causes increased vibration, three intervention options can be output: one is to actively reduce the propulsion rate by 5%, the second is to switch the cooling mode to high-speed operation, and the third is to lower the pre-adjusted pressure limit by 0.3 MPa, along with the estimated effect improvement ratio and scheduling impact duration.

[0163] In addition to short-cycle interventions, for structural problems or component degradation trends, the system needs to combine task planning and maintenance plans, reserve maintenance windows, and output the optimal intervention path. The implementation method is as follows:

[0164] Summarize information on consecutive high-risk windows and their corresponding equipment modules to determine if there is a degradation trend;

[0165] If the performance is determined to be in the medium to long term, the maintenance window priority is planned and matched with the subsequent work task cycle.

[0166] Insert adjustment paths into the current scheduling plan, such as switching to backup advance paths, adjusting the advance sequence of working faces, or compressing the work cycle;

[0167] Generate maintenance candidate window time periods and task alternative paths, and submit them to the scheduling center for review and execution.

[0168] For example, if there are 100 high-risk windows in history, and in 36 of these windows, both the cutting motor current and the drum temperature show a significant increase or decrease (i.e., a consistent trend of disturbance), then their synchronous variation probability is 36%. This value can be used to assign edge connection weights between the two, indicating the degree to which they have a correlated disturbance trend in a high-risk state of the system.

[0169] Step 6: Based on the operating condition classification results, economic loss trend sequence, and scheduling intervention strategy set generated in the preceding steps, construct a multi-level evaluation index map and realize the structured and visual expression of the evaluation results. Through the scheduling platform interface, implement the task-oriented deployment of intervention strategies and link them with the ground system. The specific implementation is as follows:

[0170] The operating conditions are expanded along a time axis, and economic loss curves are generated by combining the economic loss calculation results. Anomaly nodes corresponding to each scheduling-dependent unit (such as the cutting motor, drum, and hydraulic propulsion system) are marked on the curves. Based on this, a multi-dimensional assessment map is constructed according to the distribution density, growth threshold, and trend inflection points of the loss values. This includes, but is not limited to: a dynamic operating condition risk level map (mapping operating condition categories to risk levels), an economic loss intensity heat map (showing the economic pressure range corresponding to each channel per unit time), and a scheduling intervention feasibility grid map (indicating the adjustment or maintenance strategies that can be triggered under the current operating condition and their expected success). All assessment maps are aligned with a unified time reference and displayed in a layered format, facilitating cross-observation and comparative analysis of state evolution, risk outbreaks, and the timing of strategy implementation.

[0171] Furthermore, by combining the key nodes marked in the visualization map with the preset strategy execution interface, the intervention strategy is transformed into an executable task package through the scheduling platform interface and distributed to the well management system. Each task package includes an identification number for the corresponding working condition, suggested actions (such as frequency conversion load reduction, hydraulic pressure limiting, shutdown inspection, etc.), expected control cycle, and operation feedback path. After receiving the task package, the system automatically deploys the task and transmits scheduling instructions based on the equipment connection status and operation permissions. During the deployment process, the platform provides real-time feedback on the strategy execution status and embeds the feedback signal into the status monitoring system, updating the map display results. This forms a closed-loop response process from status identification, economic assessment, strategy recommendation to scheduling execution, realizing the linkage between surface and downhole control and the cyclical driving of assessment information.

[0172] It should be noted that the thresholds involved in the embodiments are determined and adjusted according to specific scenarios and needs.

[0173] This invention constructs a high-precision time-series tensor input foundation by sampling, aligning, and normalizing key state signals such as cutting motor current, voltage, vibration, hydraulic oil pressure, and drum temperature under a unified time reference. It constructs feature vector sequences by combining indicators such as dynamic response gradient, disturbance coupling, and abnormal activity, and uses a gated cyclic structure model to dynamically predict the coal mining machine's operating conditions, improving the forward-looking identification capability of abnormal operating states. By constructing a mapping relationship between operating conditions and unit-time output loss, energy consumption increase, and scheduling delay, it quantifies the economic impact of future operating states, generates an economic loss trend sequence, and effectively identifies high-loss risk windows. Furthermore, based on scheduling dependency graph mining, it mines state propagation chains and synchronization mutation probabilities to accurately screen intervention points and formulate multi-path intervention strategies, evaluating their implementation costs and response effects to ensure the feasibility and timeliness of scheduling schemes. Finally, it combines a visualization platform to achieve intuitive display and coordinated deployment of monitoring results and strategies, significantly improving the intelligent perception, prediction, and scheduling intervention capabilities of the coal mining machine system in complex underground environments.

[0174] Example 2: Underground coal mining machinery condition monitoring system based on operating status, such as Figure 2 As shown, it specifically includes:

[0175] The coal mining machine data acquisition module collects the operating status signals of the underground coal mining machine and performs sampling alignment and nonlinear normalization mapping under a unified time reference to construct a structured time series tensor.

[0176] The working condition category identification and acquisition module extracts dynamic response gradient factors, disturbance coupling factors and abnormal activity factors based on time series tensors, constructs state feature vectors for time windows, and models the state evolution trend according to the gated loop structure to identify the working condition category of the coal mining machine in the future time period.

[0177] The economic impact module establishes a correspondence between operating condition categories and unit time output loss, energy consumption increase, and scheduling delay factor based on operating condition categories and historical equipment response experience, generates a set of economic impact factors, and binds them to the prediction time nodes.

[0178] The risk identification module calculates the total economic loss value for each time window based on three types of economic factors, summarizes them to form an economic loss trend sequence, identifies high-risk windows, and marks potential intervention areas in conjunction with the working condition category.

[0179] The path intervention module constructs a scheduling dependency graph based on the economic loss trend sequence and high-risk windows, determines the synchronous mutation probability nodes, screens intervention points, generates intervention strategy paths, and evaluates the intervention strategy paths.

[0180] The coal mining equipment adjustment module visualizes the results of intervention strategies, status trend charts, and economic losses, and deploys and executes strategy recommendations in the form of task work orders.

[0181] The above formulas are all dimensionless calculations. Dimensionless calculations can be performed using various methods such as standardization, which will not be elaborated here. The formulas are derived from software simulations based on a large amount of collected data, and the preset parameters in the formulas can be set by those skilled in the art according to the actual situation.

[0182] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, ATA hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state ATA hard disk.

[0183] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0184] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0185] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0186] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0187] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0188] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for monitoring the operating conditions of underground coal mining machinery based on its operational status, characterized in that, Includes the following steps: The operating status signals of underground coal mining machines are collected, and sampling alignment and nonlinear normalization mapping under a unified time reference are performed to construct a structured time series tensor. Based on the extraction of dynamic response gradient factor, disturbance coupling factor and abnormal activity factor from time series tensor, a state feature vector of time window is constructed, and the state evolution trend is modeled according to the gated loop structure to identify the working condition category of coal mining machine in future time period. Based on the operating condition category and the historical response experience of the equipment, establish the correspondence between the operating condition category and the unit time output loss, energy consumption increase, and scheduling delay factor, generate a set of economic impact factors and bind them to the prediction time node. The total economic loss value for each time window is calculated based on three types of economic factors, and the economic loss trend sequence is summarized to identify high-risk windows and mark potential intervention areas in combination with the working condition category. A scheduling dependency graph is constructed based on the economic loss trend sequence and high-risk window to determine the synchronous mutation probability nodes, screen out intervention points, generate intervention strategy paths, and evaluate the intervention strategy paths. The results of intervention strategies, status trend charts, and economic losses are visualized, and the strategy recommendations are deployed and executed in the form of task work orders. A scheduling dependency graph is constructed based on the economic loss trend sequence and high-risk windows. Synchronous mutation probability nodes are determined, intervention points are selected, and intervention strategy paths are generated. The specific process is as follows: Define the set of nodes in the scheduling dependency graph, where each node corresponds to an independent device unit. The set of edges in the scheduling dependency graph is used to represent the scheduling dependency or operational linkage between two nodes, and the direction of the edge indicates the direction of influence. The weights on the edges are set to the degree of state correlation, which are calculated based on the probability of synchronous mutation of the two channels in the high-risk window in historical data. For each window marked as high risk, analyze its corresponding state feature vector and working condition category sequence to determine the main working condition types that cause economic losses. Using the rise of the economic loss factor as the signal source, expand outward in the graph to track the associated scheduling dependency graph nodes. Identify the node group with continuous abnormal responses, and determine the intervention point based on the state propagation priority score and state impact index; After identifying the points that can be intervened, an intervention strategy path is generated based on the operation control rules, equipment adjustment parameters, and historical intervention effect records.

2. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 1, characterized in that: The operating status signals of underground coal mining machines are collected, and sampling alignment and nonlinear normalization mapping under a unified time reference are performed to construct a structured time series tensor. The specific process is as follows: The sensor group includes a three-phase current acquisition unit installed at the output end of the cutting motor, a pressure transmitter and temperature sensor arranged in the front and rear chambers of the hydraulic propulsion cylinder, a triaxial vibration sensor installed on the drum structure, and a speed encoder integrated into the propulsion mechanism. The acquisition frequency of all sensor channels is uniformly set to the period synchronized with the clock signal of the coal mining machine control host, and time interpolation is used for completion and synchronization correction. A nonlinear amplitude warping strategy with interval compression is used to map and transform all sensor channels; After amplitude normalization, a state tensor sequence is constructed using a sliding window of fixed time length. All sensor channels are arranged in chronological order to form a state matrix. The rows of the matrix represent the channel numbers, the columns represent the time sequence positions, and the matrix is ​​the time series tensor corresponding to the sliding window.

3. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 2, characterized in that: The dynamic response gradient factor, perturbation coupling factor, and abnormal activity factor are extracted based on time-series tensors. The specific process is as follows: The continuous signal data of the selected channel is scanned, and the numerical change between all two adjacent data points within the entire time window is calculated. The absolute change between each pair of adjacent data points is recorded as an instantaneous gradient value. Within the time window, the maximum value among all instantaneous gradient values ​​is selected as the dynamic response gradient factor of the channel. Compare whether the two channels exhibit the same trend at each sampling time, count the number of data point pairs with consistent trends throughout the entire window period, and use the proportion as the perturbation coupling factor for the channel pairs. Within the time window, the change in value between two adjacent sampling points is calculated one by one to determine whether it exceeds the disturbance threshold set for the channel. All data points that meet the condition that the change exceeds the disturbance threshold are regarded as a high-frequency disturbance event. The total number of high-frequency disturbance events within the time window is counted and compared with the total number of sampling points within the time window as the abnormal activity factor. The dynamic response gradient factor is used to measure the maximum change intensity of a single channel within a time window, representing the degree of abrupt change in the operating state of the component corresponding to the channel; The perturbation coupling factor is used to measure the degree of synchronization between different channels; The abnormal activity factor is used to measure the proportion of each channel in the high-frequency disturbance region within a time window.

4. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 3, characterized in that: The state feature vector of the time window is constructed, and the state evolution trend is modeled according to the gated loop structure to identify the working condition category of the coal mining machine in the future time period. The specific process is as follows: Within each time window, the dynamic response gradient factor, disturbance coupling factor and abnormal activity factor features are calculated separately in each channel or channel pair and then uniformly spliced ​​together to form the state feature vector of the time window. After constructing the feature vectors for all time windows, they are combined in chronological order to form a continuous feature sequence, which serves as the input for the dynamic behavior trajectory during the operation of the coal mining machine. The working condition classification model is then used for classification and prediction. The working condition classification model uses a sequence modeling algorithm based on a gated loop structure. The model input is a sequence of state feature vectors for several consecutive time windows, and the output is the prediction result of the working condition category of the coal mining machine in the future time period. The operating condition categories include normal operation, light load operation, heavy load operation, high frequency vibration, hydraulic fluctuation, and shutdown warning.

5. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 4, characterized in that: Based on operating condition categories and historical equipment response experience, a correspondence is established between operating condition categories and unit time output loss, energy consumption increase, and scheduling delay factor. An economic impact factor set is then generated and bound to the prediction time node. The specific process is as follows: Historical equipment response experience includes historical equipment data analysis and expert knowledge annotation. Three types of economic factors include unit time output loss factor, unit time energy consumption increase factor, and unit time scheduling delay factor. Under normal operating conditions, the standard output value of the coal mining machine per unit time is recorded as the benchmark capacity. Historical data is extracted for the target state, and all window time periods under the corresponding working condition category are selected. The actual average output of the coal mining machine in all time periods is calculated. The average output under the target state is subtracted from the benchmark capacity to obtain the output loss value. The output loss value is converted to the benchmark output by ratio to obtain the output loss factor per unit time. The total energy consumption per unit time of the coal mining machine under normal conditions is statistically analyzed and used as a benchmark value, including the sum of the power consumption of the cutting motor, the traction system, and the hydraulic system. For the target working condition, the time window data corresponding to the historical conditions is extracted, and the average values ​​of the current, pressure, and temperature channels per unit time are calculated and converted into total energy consumption. The average energy consumption per unit time under the target condition is obtained. The difference between the average energy consumption under the target condition and the total energy consumption under normal conditions is compared, and the difference is converted into a ratio with the benchmark value. The ratio is used as the energy consumption increase factor. During the normal operation cycle of the coal mining machine, the average advance distance per unit time, the number of cutting tasks or the production cycle are statistically analyzed and used as a reference for the standard scheduling time series. For the target working condition, historical working condition records are extracted, and the actual task pause time, delay time or task interruption duration caused by the target working condition are statistically analyzed. The average cumulative delay duration per unit time is calculated and used as the unit time scheduling delay factor for the target working condition. The predicted operating conditions are bound together, and within each time window, the operating condition category corresponding to the time window and the values ​​of the three types of economic factors obtained by mapping are recorded to generate a sequence of economic impact factors.

6. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 5, characterized in that: The total economic losses for each time window are calculated based on three types of economic factors, and the results are summarized to form an economic loss trend sequence. High-risk windows are identified, and potential intervention areas are marked in conjunction with the operating condition category. The specific process is as follows: For each time window, based on the three types of economic factors bound to it, losses are calculated from three dimensions: capacity loss, electricity cost, and dispatch cost, and the sum is taken as the economic loss value of the time window. Using the window number as the time axis, the economic loss values ​​of all time windows are summarized in sequence to construct a complete time series structure, which serves as the economic loss trend series. The economic loss values ​​for each time window are arranged in the order of collection, forming a data sequence with time as the horizontal axis and economic loss as the vertical axis. Risk thresholds are set based on the loss trend sequence. High quantiles of the economic loss trend sequence are selected as high-risk criteria. The entire economic loss trend sequence is traversed, and window numbers of all economic loss values ​​higher than the risk threshold are extracted, identified as high-risk windows, and marked as potential intervention areas.

7. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 6, characterized in that: For each time window, based on the three types of economic factors involved, losses are calculated from three dimensions: capacity loss, electricity cost, and dispatch cost. The sum of these values ​​is then used as the economic loss value for the time window. The specific process is as follows: Based on the unit time output loss factor and normal coal mining quota, calculate the theoretical output value loss caused by the decrease in production capacity during the time window. Calculate the additional energy cost generated within the time window based on the energy consumption increase factor per unit time and the electricity price per kilowatt-hour of equipment operation. Based on the unit time scheduling delay factor and the scheduling manpower cost benchmark, calculate the implicit manpower and equipment idle cost caused by scheduling delay within the time window; The three results are added together to form the total economic loss value within the time window.

8. The method for monitoring the operating condition of underground coal mining machinery based on its operating status according to claim 7, characterized in that: The intervention strategy pathways were then evaluated, and the specific process is as follows: The implementation cost is the combined cost of labor, material costs, and equipment downtime losses required to complete all actions in the intervention strategy path; Response latency is the average time interval between the issuance of a strategy and the actual feedback of its effect. Historical success rate is the percentage of intervention strategies that have successfully reduced economic losses to below a set threshold in historical operating conditions.

9. A monitoring system for the operating conditions of underground coal mining machinery based on operational status, used to implement the monitoring method for the operating conditions of underground coal mining machinery based on operational status as described in any one of claims 1-8, characterized in that, include: The coal mining machine data acquisition module collects the operating status signals of the underground coal mining machine and performs sampling alignment and nonlinear normalization mapping under a unified time reference to construct a structured time series tensor. The working condition category identification and acquisition module extracts dynamic response gradient factors, disturbance coupling factors and abnormal activity factors based on time series tensors, constructs state feature vectors for time windows, and models the state evolution trend according to the gated loop structure to identify the working condition category of the coal mining machine in the future time period. The economic impact module establishes a correspondence between operating condition categories and unit time output loss, energy consumption increase, and scheduling delay factor based on operating condition categories and historical equipment response experience, generates a set of economic impact factors, and binds them to the prediction time nodes. The risk identification module calculates the total economic loss value for each time window based on three types of economic factors, summarizes them to form an economic loss trend sequence, identifies high-risk windows, and marks potential intervention areas in conjunction with the working condition category. The path intervention module constructs a scheduling dependency graph based on the economic loss trend sequence and high-risk windows, determines the synchronous mutation probability nodes, screens intervention points, generates intervention strategy paths, and evaluates the intervention strategy paths. The coal mining equipment adjustment module visualizes the results of intervention strategies, status trend charts, and economic losses, and deploys and executes strategy recommendations in the form of task work orders.

Images