An online detection method and system for coal mine scraper operation

Abstract

The application provides a coal mine scraper operation online detection method and system, and relates to the technical field of coal mine scraper monitoring.The method comprises the following steps: obtaining a running video of a coal mine scraper under a first running parameter; determining coal mine transportation information of each transportation section by using a transportation information determination model based on the running video of the coal mine scraper under the first running parameter; constructing a transportation atlas according to the head-to-tail order of the transportation sections; determining a chain tensioning degree reasonable interval of each transportation section of the coal mine scraper by processing the transportation atlas based on a graph neural network; determining scraper adjustment conveying speed and chain adjustment tensioning degree based on the chain tensioning degree reasonable interval of each transportation section; and adjusting the coal mine scraper based on the scraper adjustment conveying speed and the chain adjustment tensioning degree.The method can dynamically optimize the chain tensioning degree and the conveying speed according to the actual situation of each transportation section of the scraper to ensure stable and efficient operation.

Description

Technical Field

[0001] This invention relates to the field of coal mine scraper conveyor monitoring technology, specifically to an online detection method and system for coal mine scraper conveyor operation. Background Technology

[0002] In coal mining, scraper conveyors serve as crucial conveying equipment in underground longwall faces and roadways, and their stable operation directly impacts coal production efficiency and operational safety. Due to the complex and harsh underground conditions, scraper conveyors operate under high loads and high dust levels, making their transport process susceptible to factors such as coal seam conditions, coal particle size, and mining pace, resulting in significant dynamic changes in transport status. In traditional scraper conveyor operation and management, chain tension and conveying speed adjustments rely heavily on manual experience or simple preset parameters, making it difficult to adapt to fluctuations in the transport load of each section in real time. When chain tension does not match the actual load, it can easily lead to chain skipping, chain jamming, and chain breakage. Furthermore, an imbalance between conveying speed and coal quantity can cause coal accumulation and blockage or low transport efficiency. In addition, the operating status of each transport section of the scraper conveyor is interconnected; if an anomaly in one section is not promptly reported to other sections, it can trigger a chain reaction, exacerbating equipment wear and even interrupting production. With the development of intelligent coal mining, higher demands are placed on the precise perception and dynamic control of scraper conveyor operating status.

[0003] Therefore, how to accurately and dynamically optimize the chain tension and conveying speed according to the actual conditions of each transport section of the scraper conveyor to ensure its stable and efficient operation is an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem this invention addresses is how to accurately and dynamically optimize the chain tension and conveying speed based on the actual conditions of each transport section of the scraper conveyor to ensure its stable and efficient operation.

[0005] According to a first aspect, the present invention provides an online detection method for the operation of a coal mine scraper conveyor, comprising: acquiring a video of the operation of the coal mine scraper conveyor under first operating parameters, the first operating parameters of the coal mine scraper conveyor including a first conveying speed and a first chain tension; determining coal mine transportation information for each transportation segment based on the video of the operation of the coal mine scraper conveyor under the first operating parameters using a transportation information determination model; constructing a transportation map according to the order of the transportation segments from beginning to end, the transportation map including multiple transportation segment nodes and multiple edges between the multiple transportation segment nodes, the node features of each transportation segment node including the coal mine transportation information of the transportation segment, and the edges between different transportation segment nodes representing the difference in coal mine load; processing the transportation map based on a graph neural network to determine a reasonable range of chain tension for each transportation segment of the coal mine scraper conveyor; determining the adjustment of the scraper conveyor conveying speed and the adjustment of the chain tension based on the reasonable range of chain tension for each transportation segment; and adjusting the coal mine scraper conveyor based on the adjustment of the scraper conveyor conveying speed and the adjustment of the chain tension.

[0006] In one possible implementation, determining the scraper conveyor's adjustable conveying speed and chain tension based on the reasonable range of chain tension for each transport segment includes: determining a reasonable range of conveying speed for each transport segment based on the reasonable range of chain tension and coal mine transport information for each transport segment; generating multiple second operating parameters for the coal mine scraper conveyor based on the reasonable range of chain tension and conveying speed for each transport segment; determining operating information for each second operating parameter based on the multiple second operating parameters and coal mine transport information for each transport segment; and determining the scraper conveyor's adjustable conveying speed and chain tension based on the operating information of each second operating parameter.

[0007] In one possible implementation, the transportation information determination model is a recurrent neural network.

[0008] In one possible implementation, the operational information of each second operational parameter includes overall operational stability, transport segment load balance, chain life loss, and transport efficiency.

[0009] According to a second aspect, the present invention provides an online monitoring system for the operation of a coal mine scraper conveyor, comprising: an acquisition module for acquiring a video of the operation of the coal mine scraper conveyor under first operating parameters, the first operating parameters of the coal mine scraper conveyor including a first conveying speed of the scraper conveyor and a first chain tension; a transportation information determination module for determining coal mine transportation information for each transportation segment based on the video of the operation of the coal mine scraper conveyor under the first operating parameters using a transportation information determination model; and a construction module for constructing a transportation map according to the order of transportation segments from beginning to end, the transportation map including multiple transportation segment nodes and multiple transportation segment sections. The system consists of multiple edges between points, with each transport segment node's node characteristics including coal mine transport information for that segment, and edges between nodes in different transport segments representing differences in coal mine load. A tension analysis module is used to process the transport map based on a graph neural network to determine the reasonable range of chain tension for each transport segment of the coal mine scraper conveyor. An operating parameter determination module is used to determine the scraper conveyor's adjustment speed and chain tension based on the reasonable range of chain tension for each transport segment. An adjustment module is used to adjust the coal mine scraper conveyor based on the adjusted conveyor speed and chain tension.

[0010] In one possible implementation, the operating parameter determination module is further configured to: determine a reasonable range of conveying speed for each transport segment based on a reasonable range of chain tension for each transport segment and coal mine transport information for each transport segment; generate multiple second operating parameters for the coal mine scraper conveyor based on the reasonable range of chain tension for each transport segment and the reasonable range of conveying speed for each transport segment; determine operating information for each second operating parameter based on the multiple second operating parameters of the coal mine scraper conveyor and coal mine transport information for each transport segment; and determine the adjustment of conveying speed and chain tension for the scraper conveyor based on the operating information of each second operating parameter.

[0011] In one possible implementation, the transportation information determination model is a recurrent neural network.

[0012] In one possible implementation, the operational information of each second operational parameter includes overall operational stability, transport segment load balance, chain life loss, and transport efficiency.

[0013] According to a third aspect, embodiments of the present invention provide an electronic device, comprising: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method described above, the method comprising: acquiring an operating video of a coal mine scraper conveyor under first operating parameters, the first operating parameters of the coal mine scraper conveyor including a first conveying speed of the scraper conveyor and a first chain tension; determining coal mine transportation information for each transportation segment using a transportation information determination model based on the operating video of the coal mine scraper conveyor under the first operating parameters; and proceeding from the beginning to the end of the transportation segment... The transportation graph is constructed by sorting the tails. The transportation graph includes multiple transportation segment nodes and multiple edges between the nodes. The node features of each transportation segment node include the coal mine transportation information of the transportation segment, and the edges between different transportation segment nodes represent the difference in coal mine load. The transportation graph is processed based on a graph neural network to determine the reasonable range of chain tension for each transportation segment of the coal mine scraper conveyor. Based on the reasonable range of chain tension for each transportation segment, the conveyor speed and chain tension of the scraper conveyor are adjusted. The coal mine scraper conveyor is then adjusted based on the conveyor speed and chain tension.

[0014] According to the fourth aspect, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the aforementioned online detection method for coal mine scraper conveyors. The method includes: acquiring a video of the coal mine scraper conveyor operating under first operating parameters, the first operating parameters of the coal mine scraper conveyor including a first conveying speed of the scraper conveyor and a first chain tension; determining coal mine transportation information for each transportation segment using a transportation information determination model based on the video of the coal mine scraper conveyor operating under the first operating parameters; and constructing a transportation information model according to the order of the transportation segments from beginning to end. The transportation graph includes multiple transportation segment nodes and multiple edges between these nodes. Each transportation segment node's features include coal mine transportation information, and the edges between different transportation segment nodes represent the differences in coal mine load. A graph neural network is used to process the transportation graph to determine a reasonable range for the chain tension of each transportation segment of the coal mine scraper conveyor. Based on this reasonable range, the conveyor speed and chain tension of the scraper conveyor are adjusted. Finally, the coal mine scraper conveyor is adjusted based on these adjustment parameters.

[0015] This invention provides an online detection method and system for coal mine scraper conveyors. The method includes acquiring a video recording of the coal mine scraper conveyor operating under first operating parameters, including a first conveying speed and a first chain tension. Based on the video recording of the scraper conveyor operating under the first operating parameters, a transportation information determination model is used to determine the coal mine transportation information for each transportation segment. A transportation map is constructed according to the order of the transportation segments from beginning to end. The transportation map includes multiple transportation segment nodes and multiple edges between the multiple transportation segment nodes. The node characteristics of each transportation segment node include... The method involves processing the coal mine transportation information for each transportation segment, where the edges between nodes in different transportation segments represent the differences in coal mine load. A graph neural network is used to process the transportation map to determine the reasonable range of chain tension for each transportation segment of the coal mine scraper conveyor. Based on this reasonable range, the method determines the adjustment of the scraper conveyor's conveying speed and chain tension. The scraper conveyor is then adjusted based on these adjustment speeds and chain tensions. This method can accurately and dynamically optimize the chain tension and conveying speed according to the actual conditions of each transportation segment of the scraper conveyor to ensure its stable and efficient operation. Attached Figure Description

[0016] Figure 1 A schematic flowchart illustrating an online detection method for the operation of a coal mine scraper conveyor provided in an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram of a coal mine scraper conveyor according to an embodiment of the present invention;

[0018] Figure 3 This is a schematic diagram of a process for determining the adjustment of conveying speed and chain tension of a scraper conveyor, provided by an embodiment of the present invention.

[0019] Figure 4 This is a schematic diagram of an online monitoring system for the operation of a coal mine scraper conveyor, provided as an embodiment of the present invention. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0021] In this embodiment of the invention, the following are provided: Figure 1 The method shown is an online detection method for the operation of a coal mine scraper conveyor, which includes steps S1 to S6:

[0022] Step S1: Obtain the operating video of the coal mine scraper conveyor under the first operating parameters, which include the first conveying speed of the scraper conveyor and the first tension of the chain.

[0023] A scraper conveyor is a continuous conveying device that can be used in underground coal mine longwall faces and roadways. Figure 2 This is a schematic diagram of a coal mine scraper conveyor according to an embodiment of the present invention. The coal mine scraper conveyor uses a motor to drive a sprocket to rotate, which in turn drives a scraper chain that passes around the sprocket to circulate in the central trough. The scrapers on the scraper chain can be used to transport materials such as coal from the working face to a designated location.

[0024] The initial operating parameters of a coal mine scraper conveyor are the set of operating parameters set during the initial operation phase. These initial operating parameters include the initial conveying speed of the scraper conveyor and the initial chain tension.

[0025] The first conveying speed of the scraper conveyor is the speed parameter of the scraper conveyor when conveying coal under the first operating parameters in a coal mine.

[0026] The first chain tension is the tension parameter of the chain of a coal mine scraper conveyor under the first operating parameters.

[0027] Tension is a standardized quantitative value in the range of 0-1, directly corresponding to the physical tension of the chain. The closer the value is to 1, the higher the chain tension. Specifically, the closer the angle between the line connecting two adjacent sprockets and the chain is to 180 degrees, the more horizontal and taut the chain appears, with no obvious sag, and the sprockets and links meshing tightly with uniform gaps. There is also no shaking or vibration during operation. Conversely, the closer the value is to 0, the higher the chain slack. This is characterized by a smaller angle between the line connecting two adjacent sprockets and the lowest point of the chain's sag, resulting in a U-shaped sag, larger and fluctuating gaps between links at the meshing points, and a tendency to exhibit wavy swaying during operation.

[0028] For example, the Φ26×92 high-strength circular link chain for mining, suitable for medium and large scraper conveyors with a span of 100 meters, has a reasonable preset tension range of 0.4-0.6, corresponding to an angle between the line connecting the two sprockets and the chain of 160-170 degrees. This also results in moderate chain sag, avoiding both stress concentration due to excessive tension and poor meshing caused by slack. If the tension is 0.8, the angle between the line connecting the two sprockets and the chain will approach 180 degrees, leading to excessive chain rigidity and potentially accelerating sprocket tooth wear and motor load. If the tension is as low as 0.15, the angle between the line connecting the two sprockets and the chain will be less than 130 degrees, resulting in severe chain sag and a risk of friction with the central groove, tooth skipping, or chain derailment.

[0029] The video footage of the coal mine scraper conveyor operating under the first set of operating parameters is captured by video surveillance equipment installed near the scraper conveyor. This video footage can be used to record dynamic information such as the scraper conveyor's operating status and the coal transportation process, as well as to reflect the actual working condition of the scraper conveyor under the current parameters.

[0030] Step S2: Based on the operating video of the coal mine scraper conveyor under the first operating parameters, the coal mine transportation information for each transportation segment is determined using the transportation information determination model.

[0031] The transportation information determination model is a recurrent neural network. The input of the transportation information determination model is the operating video of the coal mine scraper conveyor under the first operating parameters, and the output of the transportation information determination model is the coal mine transportation information for each transportation segment.

[0032] Recurrent Neural Networks (RNNs) are neural network models that excel at processing time-series data and can be used to analyze dynamic processes that change over time. RNNs retain historical input information and capture temporal dependencies in the data through recurrent connections in hidden layers. When processing video sequences, RNNs can use information from the previous frame to aid in understanding the current frame.

[0033] The transport section refers to the physical section that corresponds one-to-one with the central trough of the coal mine scraper conveyor. Each transport section serves as the basic unit for transporting coal by the scraper conveyor, and the scope of the transport section strictly corresponds to the spatial area of ​​the central trough of a single section.

[0034] The coal transportation information for each transport segment is extracted from the output of the transportation information determination model to accurately describe the transportation load and status of each segment. This information includes the instantaneous coal quantity, coal distribution density, and the contact state between the scraper and the coal in that segment.

[0035] The operating video of the coal mine scraper conveyor under the first operating parameters contains the continuous operating status of each transportation section over time, such as the dynamic processes of coal loading, moving, and unloading. This temporal information directly reflects the actual transportation situation of each transportation section. A recurrent neural network can analyze the operating video frame by frame and extract the visual features of each transportation section in each frame, such as the pixel proportion of the coal area and the scraper position. Then, through recurrent connections, the feature information from previous frames is passed to the current frame to analyze the changing patterns of features over time, such as the increase or decrease trend of coal quantity in a certain transportation section and the synchronicity between scraper movement and coal transportation. By combining the physical division of transportation sections, the model can associate the temporal features with the corresponding transportation sections, thereby calculating data such as the coal quantity and transportation speed of each transportation section at different times.

[0036] Step S3: Construct a transportation graph according to the order of transportation segments from beginning to end. The transportation graph includes multiple transportation segment nodes and multiple edges between multiple transportation segment nodes. The node characteristics of each transportation segment node include the coal mine transportation information of the transportation segment, and the edges between different transportation segment nodes are the coal mine load difference.

[0037] The transportation map is constructed according to the sequence of the coal mine scraper conveyor transportation sections, and can intuitively present the transportation status of each transportation section and the load differences between them.

[0038] The transportation map consists of multiple transportation segment nodes and edges between the nodes. Each transportation segment node is characterized by the corresponding coal mine transportation information, and the edges between nodes represent the difference in coal mine load between adjacent transportation segments.

[0039] Constructing transportation maps can provide structured data for analyzing the correlation between transportation segments.

[0040] Step S4: Based on the graph neural network, process the transport map to determine the reasonable range of chain tension for each transport section of the coal mine scraper conveyor.

[0041] Graph Neural Networks (GNNs) are neural network models capable of processing graph data. Through message passing mechanisms, GNNs enable nodes to aggregate features from their neighbors and edge information, updating their own representations. GNNs can capture topological relationships and node associations within a graph. The input to the GNN is the transportation graph, and the output is the reasonable range of chain tension for each transportation segment of a coal mine scraper conveyor.

[0042] The reasonable range of chain tension for each transport section of a coal mine scraper conveyor is determined by processing the transport map using a graph neural network for each transport section.

[0043] The reasonable range of chain tension for each transport section of a coal mine scraper conveyor is a reasonable threshold that ensures the chain in that transport section is neither too loose nor too tight, and can prevent chain slippage and breakage.

[0044] By constructing a transport map, the physical connections and load transfer relationships between each transport segment can be clearly reflected. The chain tension of a single transport segment is directly affected by load changes in adjacent transport segments. Using coal mine transport information for each segment as node features, and the load difference between adjacent segments as edge features, allows for more comprehensive utilization of the scraper conveyor's operational data. This helps the model better understand the load status and mutual influence patterns of each transport segment, improving the accuracy of determining the reasonable range for chain tension.

[0045] Graph Neural Networks (GNNs) can transform the node and edge features of a transportation graph into initial vector representations. Through the message passing mechanism of GNNs, each node can aggregate the features of its neighboring nodes and the load difference information of its edges. For example, a node in a transportation segment might receive transportation information and load differences from the preceding segment and analyze the impact of the preceding segment on itself. By iteratively updating node representations through multi-layer networks and fusing global topology information, GNNs can uncover implicit relationships between transportation segments, such as the cascading effect of a sudden load change in one segment on the tension of subsequent segments. Based on the updated node features, GNNs can learn the mapping relationship between transportation load, adjacent load differences, and chain tension. For example, transportation segments with high loads and large load differences require narrower tension intervals to ensure stability. Ultimately, GNNs can output a reasonable chain tension interval for each transportation segment, ensuring that the interval adapts to its transportation status and the impact of adjacent segments.

[0046] Step S5: Determine the scraper conveyor speed and chain tension adjustment based on the reasonable range of chain tension for each transport segment.

[0047] In some embodiments, Figure 3 This is a schematic flowchart illustrating the process of determining the conveying speed and chain tension of a scraper conveyor according to an embodiment of the present invention. The process of determining the conveying speed and chain tension includes steps S51-S54:

[0048] Step S51: Determine the reasonable range of conveying speed for each transport segment based on the reasonable range of chain tension for each transport segment and the coal mine transport information for each transport segment.

[0049] In some embodiments, a speed determination model can be used to determine the reasonable range of conveying speed for each transport segment. The speed determination model is a Transformer model. The inputs of the speed determination model are the reasonable range of chain tension for each transport segment and the coal mine transport information for each transport segment. The output of the speed determination model is the reasonable range of conveying speed for each transport segment.

[0050] The Transformer model consists of an encoder and a decoder. The encoder learns representations from the input data, including a self-attention mechanism and a feed-forward network. The decoder, in addition to the encoder, introduces a multi-head attention mechanism, which decodes the encoder output and generates the target sequence.

[0051] The reasonable range of transport speed for each transport segment is determined by a speed determination model and is the range of transport speeds for each transport segment.

[0052] The reasonable range of chain tension for each transport segment defines the safe operating range of the chain. Coal mine transportation information provides data such as the amount of coal and transportation demand for that segment. This information provides constraints for the model to determine the conveying speed range; too high a speed may cause the chain tension to exceed the upper limit, while too low a speed will fail to meet the transportation demand.

[0053] The Transformer model's encoder encodes the reasonable range of chain tension and coal mine transportation information for each transport segment. Through a self-attention mechanism, the model can focus on the correlation between tension and transport volume within the same transport segment and speed coordination between different transport segments, such as the sensitivity of high coal volumes to the upper limit of tension, and the impact of preceding segment speeds on subsequent segment transport volumes. The decoder, based on the encoded global features and the positive correlation between speed and tension, generates candidate speed ranges for each transport segment and verifies these ranges to ensure that the chain tension does not exceed its reasonable range at these speeds and meets transport volume requirements.

[0054] In some embodiments, determining the reasonable range of conveying speed for each transport segment based on the reasonable range of chain tension for each transport segment and the coal mine transport information for each transport segment includes steps S61 to S63:

[0055] Step S61: Based on the coal mine transportation information of each transportation segment, determine the transportation pressure level of each transportation segment, the transportation stability requirement of each transportation segment, and the scraper running resistance coefficient of each transportation segment.

[0056] In some embodiments, deep neural networks may be used to determine the transport pressure level of each transport segment, the transport smoothness requirement of each transport segment, and the scraper running resistance coefficient of each transport segment.

[0057] The transportation pressure level of each transportation segment is determined by a deep neural network, reflecting the classification of the coal load pressure borne by the transportation segment.

[0058] The transport stability requirement for each transport segment is determined by a deep neural network to measure the degree of stable operation that segment needs to maintain. For example, transport segments with uneven coal distribution and frequent scraper contact have higher stability requirements to avoid coal spillage or accelerated equipment wear due to operational fluctuations.

[0059] The scraper running resistance coefficient for each transport section is a coefficient output by a deep neural network that reflects the magnitude of the resistance experienced by the scraper during operation in that transport section.

[0060] Deep neural networks can learn the intrinsic relationship between coal mine transportation information and output parameters for each transportation segment through a multi-layered network structure. For example, the model can automatically capture the corresponding patterns between an increase in instantaneous coal volume and an increase in transportation pressure level, and between uneven coal distribution density and increased transportation stability requirements. Moreover, without the need for manually preset mapping rules, the coal mine transportation information for each transportation segment can be transformed into structured corresponding transportation pressure levels, transportation stability requirements, and scraper running resistance coefficients.

[0061] Step S62: Based on the transportation pressure level of each transportation segment, the transportation stability requirement of each transportation segment, the scraper running resistance coefficient of each transportation segment, and the reasonable range of chain tension of each transportation segment, determine the speed safety boundary of each transportation segment, the speed efficiency boundary of each transportation segment, multiple sets of simulated speed schemes, and the feasibility score of each set of simulated schemes.

[0062] In some embodiments, deep neural networks can be used to determine the speed safety boundary for each transport segment, the speed efficiency boundary for each transport segment, multiple sets of simulated speed schemes, and the feasibility score for each set of simulated schemes.

[0063] The speed safety boundary for each transport segment is determined by a deep neural network in combination with parameters such as the reasonable range of chain tension and transport pressure level of the transport segment.

[0064] The speed safety boundary ensures that when the scraper conveyor is running at a speed not exceeding this limit, the chain tension will not exceed the safety range, thus effectively avoiding safety hazards such as chain breakage due to excessive tightness or slippage due to excessive looseness caused by excessive speed.

[0065] The speed efficiency boundary for each transport segment is a lower speed limit determined by a deep neural network based on factors such as transport pressure level and transport stability requirements.

[0066] The speed efficiency boundary ensures that the scraper conveyor can meet the coal transportation demand of the transportation section when operating at this speed and above, thereby avoiding coal accumulation due to excessively low speed and ensuring that the transportation efficiency is not lower than the necessary level.

[0067] Multiple simulated speed schemes are generated by a deep neural network within the speed safety boundary and speed efficiency boundary, taking into account parameters such as the transport pressure level of the transport section and the scraper running resistance coefficient, as well as the corresponding chain tension adjustment trends. Each scheme simulates the operating state of the scraper conveyor at different speeds.

[0068] The feasibility score for each simulation scheme is derived from a score obtained by quantitatively evaluating multiple simulated speed schemes from dimensions such as safety, efficiency, and stability, based on the output of a deep neural network. The higher the score, the more feasible the scheme is in actual operation of the current transportation segment.

[0069] There are multi-dimensional constraints among the transportation pressure level, transportation stability requirements, scraper running resistance coefficient, and reasonable chain tension range for each transportation segment. The higher the transportation pressure level, the stricter the restriction on the speed safety boundary. The reasonable chain tension range directly defines the safe speed range, while the scraper running resistance coefficient affects the setting of the speed efficiency boundary. Deep neural networks, through hierarchical feature fusion, can learn the inherent patterns of these constraints. For example, they can capture the matching pattern between the transportation pressure level and the speed safety boundary under a specific tension range from historical data, as well as the correspondence between the resistance coefficient and the efficiency boundary. This allows for the generation of speed safety boundaries and speed efficiency boundaries that meet all constraints, as well as multiple simulated speed schemes within the boundaries and their corresponding feasibility scores.

[0070] Step S63: Determine the reasonable range of transport speed for each transport segment based on the speed safety boundary of each transport segment, the speed efficiency boundary of each transport segment, the multiple sets of simulated speed schemes, and the feasibility score of each set of simulated schemes.

[0071] In some embodiments, deep neural networks can be used to determine a reasonable range of transport speeds for each transport segment.

[0072] The speed safety boundary, speed efficiency boundary, multiple simulated speed schemes, and feasibility scores for each simulated scheme together constitute complete information for determining a reasonable range of transport speeds for each transport segment. These information elements have complementary boundaries that provide range constraints, while the schemes and scores provide practical adaptability references. Deep neural networks, through training, can learn how to integrate this information, such as learning from historical data how to select speed ranges covered by high-scoring schemes within the safety and efficiency boundaries, thereby accurately outputting a reasonable range of transport speeds that balances constraints and feasibility.

[0073] Step S52: Generate multiple second operating parameters for the coal mine scraper conveyor based on the reasonable range of chain tension and the reasonable range of conveying speed for each transport segment.

[0074] In some embodiments, a second parameter generation model can be used to generate multiple second operating parameters of the coal mine scraper conveyor. The second parameter generation model is a deep neural network model. The input of the second parameter generation model is the reasonable range of chain tension and the reasonable range of conveying speed of each transport segment. The output of the second parameter generation model is multiple second operating parameters of the coal mine scraper conveyor.

[0075] Deep neural network models include deep neural networks (DNNs). Deep neural networks consist of multiple hidden layers. Through non-linear activation functions, deep neural networks can extract and transform input features layer by layer and learn complex non-linear relationships.

[0076] The secondary operating parameters of the coal mine scraper conveyor are generated through a secondary parameter generation model, which includes a set of alternative parameters for the combination of scraper conveyor speed and chain tension. Each secondary operating parameter must satisfy a reasonable range for chain tension and conveyor speed across all transport segments. These multiple secondary operating parameters reflect different adjustment strategies, such as emphasizing efficiency or stability.

[0077] Deep neural networks can transform the tension and speed ranges of each transport segment into high-dimensional feature vectors. The first hidden layer can extract basic features such as the boundary values ​​and ranges of the segments. Subsequent layers then learn the relationships between segments through nonlinear transformations, such as the matching relationship between a tension segment and the speed range of an adjacent segment. Based on the learned relationships, the model can generate multiple combinations of transport speed and chain tension that consider global coordination within a reasonable range, such as adjusting tension while increasing overall speed to maintain balance, and verify the feasibility of the generated combinations.

[0078] Step S53: Determine the operating information of each second operating parameter based on the multiple second operating parameters of the coal mine scraper conveyor and the coal transportation information of each transportation section.

[0079] In some embodiments, an information determination model can be used to determine the operating information of each second operating parameter. The information determination model is a Transformer model, and its inputs are multiple second operating parameters of the coal mine scraper conveyor and coal transportation information for each transport segment. The output of the information determination model is the operating information of each second operating parameter.

[0080] The operational information for each second operational parameter is specific information output by the information determination model used to evaluate the operational performance of the combination of each second operational parameter. The operational information for each second operational parameter includes overall operational stability, transport segment load balance, chain life loss, and transport efficiency.

[0081] The operational information obtained from the second operating parameter can reflect the merits and demerits of the parameter combination from different dimensions.

[0082] Several secondary operating parameters of a coal mine scraper conveyor are fundamental data describing the equipment's operating status, directly reflecting the real-time performance of each core component. For example, motor power reflects power output, and chain temperature reflects friction intensity. Coal transportation information for each transport segment provides the context for parameter operation, clarifying the external driving factors for parameter changes. For instance, a sudden surge in coal volume in a transport segment may lead to an increase in motor power. The changes in the values ​​of the secondary operating parameters, combined with transportation information, can help determine whether the changes are reasonable.

[0083] The Transformer model, through self-attention and multi-head attention mechanisms, can simultaneously capture the intrinsic relationships between multiple secondary operating parameters of a coal mine scraper conveyor, as well as the relationships between these secondary operating parameters and coal transportation information for each transport segment. The Transformer model's encoder transforms discrete parameter values ​​and transportation information into high-dimensional feature vectors, then learns different dimensional correlation patterns in parallel through multi-head attention, such as the relationship between short-term parameter fluctuations and instantaneous transport volume, and the relationship between long-term trends and transport segment characteristics. The decoder, based on these fused features, generates multi-dimensional operating information including state features and change trends. The Transformer's ability to process long-sequence data effectively captures the dynamic changes of parameters in different transport segments, ensuring that the output operating information reflects both real-time status and scenario adaptability.

[0084] Step S54: Determine the scraper conveyor speed and chain tension adjustment based on the operating information of each second operating parameter.

[0085] In some embodiments, a tension determination model can be used to determine the scraper conveyor speed adjustment and the chain tension adjustment. The tension determination model is a deep neural network model. The input to the tension determination model is the operating information of each of the second operating parameters, and the output of the tension determination model is the scraper conveyor speed adjustment and the chain tension adjustment.

[0086] The scraper conveyor adjusts the conveying speed and chain tension by analyzing the operating information of each secondary operating parameter through a tension determination model, and then selecting the optimal parameter combination.

[0087] Adjusting the conveying speed of the scraper conveyor is the final determined speed parameter for conveying coal.

[0088] Chain tension adjustment is the final parameter determining the chain's tightness.

[0089] Adjusting the conveying speed and chain tension of the scraper conveyor can balance the operational stability, efficiency, and chain life of the coal mine scraper conveyor.

[0090] The operational information for each second operating parameter comprehensively evaluates the performance of each parameter combination from four dimensions: overall operational stability, load balance of the transport segment, chain lifespan loss, and conveying efficiency. This provides a quantitative basis for the model to select the optimal parameters. The differences in this information enable the model to identify the parameter combination with the best overall performance. Deep neural networks have the ability to process multi-dimensional, non-linearly correlated data and can extract features from this multi-dimensional information layer by layer through multiple hidden layers, capturing the complex relationships between various indicators, such as the potential trade-off between high conveying efficiency and high chain loss. Through non-linear transformations, the model can learn the weights of each indicator under different scenarios, such as prioritizing stability in heavy-load scenarios and focusing on improving efficiency in light-load scenarios. Based on this learned correlation and weights, the model can quantitatively evaluate the overall performance of each second operating parameter and ultimately select the parameter combination that achieves the optimal balance between stability, load balance, loss, and efficiency, thereby determining the adjustment of the scraper conveyor's conveying speed and chain tension.

[0091] Step S6: Adjust the coal mine scraper conveyor based on the scraper conveyor's adjustable conveying speed and the chain's adjustable tension.

[0092] Once the scraper conveyor speed and chain tension are determined, the coal mine scraper conveyor is adjusted and controlled based on these parameters.

[0093] Based on the same inventive concept Figure 4 This is a schematic diagram of an online monitoring system for the operation of a coal mine scraper conveyor, provided by an embodiment of the present invention. The online monitoring system for the operation of the coal mine scraper conveyor includes:

[0094] The acquisition module 61 is used to acquire the operating video of the coal mine scraper conveyor under the first operating parameters, which include the first conveying speed of the scraper conveyor and the first tension of the chain.

[0095] The transportation information determination module 62 is used to determine the coal transportation information of each transportation segment based on the operating video of the coal mine scraper conveyor under the first operating parameters using the transportation information determination model;

[0096] Module 63 is used to construct a transportation graph according to the order of transportation segments from beginning to end. The transportation graph includes multiple transportation segment nodes and multiple edges between multiple transportation segment nodes. The node characteristics of each transportation segment node include the coal mine transportation information of the transportation segment, and the edges between different transportation segment nodes are the coal mine load difference.

[0097] The tension analysis module 64 is used to process the transport map based on a graph neural network to determine the reasonable range of chain tension for each transport section of the coal mine scraper conveyor.

[0098] The operating parameter determination module 65 is used to determine the scraper conveyor speed and chain tension adjustment based on the reasonable range of chain tension for each transport segment;

[0099] The adjustment module 66 is used to adjust the coal mine scraper conveyor based on the scraper conveyor's adjustable conveying speed and the chain's adjustable tension.

[0100] It should be noted that, in order to simplify the descriptions disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments of this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0101] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method of on-line detection of the operation of a coal mine scraper, characterized in that, The method comprises the following steps: obtaining a running video of a coal mine scraper under first running parameters, the first running parameters of the coal mine scraper comprising a first conveying speed of the scraper and a first tension of the chain; determining coal mine transportation information of each transportation section based on the running video of the coal mine scraper under the first running parameters using a transportation information determination model, the transportation information determination model being a recurrent neural network; constructing a transportation graph in the order of transportation sections from head to tail, the transportation graph comprising a plurality of transportation section nodes and a plurality of edges between the transportation section nodes, the node features of each transportation section node comprising the coal mine transportation information of the transportation section, and the edges between different transportation section nodes being coal load difference values; processing the transportation graph based on a graph neural network to determine a reasonable chain tension interval of each transportation section of the coal mine scraper; determining a scraper adjusted conveying speed and a chain adjusted tension based on the reasonable chain tension interval of each transportation section, which comprises: determining a reasonable conveying speed interval of each transportation section based on the reasonable chain tension interval of each transportation section and the coal mine transportation information of each transportation section; determining a transportation pressure level, a transportation stability requirement, and a scraper running resistance coefficient of each transportation section based on the coal mine transportation information of each transportation section; determining a speed safety boundary, a speed efficiency boundary, a plurality of simulated speed schemes, and a feasibility score of each simulated scheme of each transportation section based on the transportation pressure level, the transportation stability requirement, the scraper running resistance coefficient, and the reasonable chain tension interval of each transportation section; determining a reasonable conveying speed interval of each transportation section based on the speed safety boundary, the speed efficiency boundary, the plurality of simulated speed schemes, and the feasibility score of each simulated scheme of each transportation section; generating a plurality of second running parameters of the coal mine scraper based on the reasonable chain tension interval of each transportation section and the reasonable conveying speed interval of each transportation section; determining running information of each second running parameter based on the plurality of second running parameters of the coal mine scraper and the coal mine transportation information of each transportation section, the running information of each second running parameter comprising a total stability, a transportation section load balancing degree, a chain life consumption degree, and a conveying efficiency; determining a scraper adjusted conveying speed and a chain adjusted tension based on the running information of each second running parameter; adjusting the coal mine scraper based on the scraper adjusted conveying speed and the chain adjusted tension.

2. A coal mine scraper operation online detection system for performing the coal mine scraper operation online detection method of claim 1, characterized in that, The method comprises the following steps: an obtaining module, configured to obtain a running video of a coal mine scraper under first running parameters, the first running parameters of the coal mine scraper comprising a first conveying speed of the scraper and a first tension of the chain; The transport information determination module is configured to determine coal mine transport information of each transport section based on the running video of the coal mine scraper under the first running parameter and using a transport information determination model. The construction module is configured to construct a transport graph in accordance with the sequence of the transport sections from head to tail, the transport graph comprising a plurality of transport section nodes and a plurality of edges between the transport section nodes, the node features of each transport section node comprising the coal mine transport information of the transport section, and the edges between different transport section nodes being coal load difference values. The tension degree analysis module is configured to determine a chain tension degree reasonable interval of each transport section of the coal mine scraper based on processing of the transport graph by a graph neural network. The running parameter determination module is configured to determine scraper conveyor speed adjustment and chain tension degree adjustment based on the chain tension degree reasonable interval of each transport section. The adjustment module is configured to adjust the coal mine scraper based on the scraper conveyor speed adjustment and the chain tension degree adjustment.

3. The coal mine scraper operation online detection system of claim 2, wherein, The running parameter determination module is further configured to: determine a conveyor speed reasonable interval of each transport section based on the chain tension degree reasonable interval of each transport section and the coal mine transport information of each transport section; generate a plurality of second running parameters of the coal mine scraper based on the chain tension degree reasonable interval of each transport section and the conveyor speed reasonable interval of each transport section; determine running information of each second running parameter based on the plurality of second running parameters of the coal mine scraper and the coal mine transport information of each transport section; and determine scraper conveyor speed adjustment and chain tension degree adjustment based on the running information of each second running parameter.

4. An electronic device, comprising: The computer program is stored in the memory and configured to be executed by the processor to implement the coal mine scraper running online detection method of claim 1. The program is executed by the processor to implement the coal mine scraper running online detection method of claim 1. ​ ​ 5. A computer-readable storage medium having stored thereon a computer program, characterized in that, ​

Images