Anomaly Detection and Early Warning Response Analysis System and Method for Coal Mine Underground Operation Data

Abstract

This invention relates to the field of intelligent analysis technology for coal mine electromechanical monitoring and drainage systems. It provides a system and method for anomaly perception and early warning response analysis of underground coal mine operation data. The system includes: acquiring the current operating condition, historical monitoring data, and initial weighting of voltage fluctuations for a target pump in the underground coal mine drainage system; determining whether the target pump is under a specified operating condition characterized by high head, single-pump operation, and long-distance power supply based on the current operating condition; and, in response to the target pump being under the specified operating condition, sequentially performing sample classification, current fluctuation segment selection, and fluctuation amplitude calculation based on the historical monitoring data, and then correcting the initial weighting of voltage fluctuations using a correction function. This invention can detect early performance degradation without triggering traditional anomaly conditions, thereby improving the prediction accuracy and response timeliness of underground coal mine drainage systems in complex environments, and significantly enhancing the operational safety and reliability of the drainage system.

Description

Technical Field

[0001] This invention belongs to the field of intelligent analysis technology for coal mine electromechanical monitoring and drainage systems, and particularly relates to an anomaly perception and early warning response analysis system and method for coal mine underground operation data. Background Technology

[0002] In underground coal mine production operations, drainage systems are one of the crucial infrastructure components ensuring mine safety and normal production. Due to the hot, humid, dusty, and high-load environment underground, coupled with long electrical power supply distances, drainage systems must operate under high-intensity, high-lift conditions for extended periods. Existing coal mine drainage systems typically rely on monitoring and control devices to collect real-time operating parameters such as voltage, current, flow rate, and pressure, and use set alarm thresholds or expert experience models to assess the operating status of drainage pumps. When voltage fluctuations exceed limits, flow rates drop excessively, or pump loads become abnormal, the system issues an early warning signal, alerting maintenance personnel to perform repairs or adjustments. This threshold-based monitoring method is simple in structure and easy to implement, and has been widely applied in the safe operation management of coal mine electromechanical equipment.

[0003] Due to the long underground power supply lines and significant voltage loss, the system commonly exhibits phenomena such as minor voltage fluctuations and slight flow rate fluctuations. However, these fluctuations are often considered "normal deviations within the allowable range" and are not further analyzed. On the other hand, existing systems typically use a single parameter exceeding its limit as an anomaly criterion, failing to identify potential efficiency degradation issues that may arise under combinations of parameters such as voltage fluctuations, stable current, and decreased flow rate. In particular, when the pump current remains stable and no alarm is triggered, the decrease in pump efficiency is often overlooked by the system, leading to a gradual deterioration of drainage performance over long periods of operation without being detected. This can ultimately result in safety hazards such as drainage delays, overflow risks, and even equipment damage. Summary of the Invention

[0004] The purpose of this invention is to provide an anomaly perception and early warning response analysis system and method for underground coal mine operation data, aiming to solve the problems mentioned in the background art.

[0005] In a first aspect, the present invention provides a method for anomaly detection and early warning response analysis of underground coal mine operation data, including:

[0006] Obtain the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of the coal mine.

[0007] Based on the current operating conditions, determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply;

[0008] In response to the target pump being in the specified operating condition, after sequentially classifying samples, selecting current fluctuation ranges, and calculating fluctuation amplitude based on the historical monitoring data, the initial weight ratio of the influence of voltage fluctuation is corrected by combining the correction function;

[0009] The corrected weighting is applied to the preset model for generating drainage reliability values, the weight of the voltage fluctuation impact term is corrected, and the overflow risk assessment results of the coal mine underground drainage system are updated.

[0010] Furthermore, in response to the target pump being under the specified operating condition, after sequentially performing sample classification, current fluctuation range selection, and fluctuation amplitude calculation based on the historical monitoring data, the initial weighting of the voltage fluctuation impact is corrected using a correction function, including:

[0011] The historical monitoring data is analyzed, and several sample data are selected that are consistent with the current background environment of the target pump and whose voltage fluctuation and flow rate drop fluctuation are within the allowable and preset range. An equal number of first-class samples and second-class samples are selected, where the first-class samples correspond to the specified operating conditions and the second-class samples correspond to non-specified operating conditions.

[0012] After determining the current data of each first-class sample and second-class sample during operation, the current fluctuation segment in each second-class sample that is relatively stable compared to the corresponding first-class sample is selected, and the current fluctuation amplitude of each current fluctuation segment compared to the stable current waveform is calculated.

[0013] The average fluctuation amplitude of all the fluctuation amplitudes is calculated to obtain the average fluctuation amplitude of the current change. This average fluctuation amplitude is then used as a correction factor to optimize the initial weight ratio, resulting in the optimized weight ratio.

[0014] Furthermore, the consistency with the current background environment of the target pump includes: the environmental conditions at the time of sample data collection are matched with the external operating conditions, electrical power supply conditions and hydraulic load conditions of the target pump at present, and the differences between the parameters do not exceed the preset tolerance range.

[0015] Furthermore, the voltage fluctuation and flow rate drop fluctuation are both within the allowable and preset range, including: the voltage fluctuation amplitude is lower than the preset abnormal alarm threshold and within the rated voltage tolerance range, while the flow rate drop amplitude is less than the preset allowable error range and does not trigger the preset drainage performance abnormality judgment condition.

[0016] Furthermore, the correction function is:

[0017]

[0018] in, This refers to the optimized weighting percentage. This refers to the initial weighting percentage. This represents the total number of current fluctuation segments identified in all second-class samples. It refers to the first The second type of sample current data corresponding to each current fluctuation segment. This refers to the steady-state current data of the first type of sample under a steady-state current waveform. This refers to the preset correction strength coefficient.

[0019] Furthermore, the historical monitoring data is collected by various sensors and monitoring terminals installed in pump rooms, power distribution cabinets and pipeline nodes, and transmitted in real time to the ground monitoring center through the mine monitoring system or industrial Ethernet. This data includes electrical parameters, hydraulic parameters, operating condition parameters and environmental parameters.

[0020] Furthermore, the initial weighting percentages for the impact of voltage fluctuations include: the initial set percentages of each influencing factor's contribution to the overall system reliability when calculating the drainage reliability value using the generated model.

[0021] Secondly, the present invention provides an anomaly detection and early warning response analysis system for underground coal mine operation data, used to implement the anomaly detection and early warning response analysis method for underground coal mine operation data as described above, including:

[0022] Data acquisition module: used to acquire the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of coal mines;

[0023] Operating condition detection module: used to determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply based on the current operating condition;

[0024] Weight correction module: In response to the target pump being in the specified operating condition, based on the historical monitoring data, sample classification, current fluctuation range selection and fluctuation amplitude calculation are performed in sequence, and then the initial weight ratio of the influence of voltage fluctuation is corrected in combination with the correction function.

[0025] Weight update module: This module applies the corrected weight ratios to the preset drainage reliability value generation model, corrects the weights of the voltage fluctuation impact term, and updates the overflow risk assessment results of the coal mine underground drainage system.

[0026] Thirdly, the present invention provides an anomaly perception and early warning response analysis device for underground coal mine operation data, including a processor and a storage medium;

[0027] The storage medium is used to store instructions;

[0028] The processor is configured to operate according to the instructions to perform the steps of the method according to any of the foregoing.

[0029] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] This invention proposes an anomaly detection and early warning response analysis method based on a combination of historical monitoring data pattern identification and dynamic weight correction, specifically addressing the unique operating conditions of coal mine underground drainage systems characterized by "high head, single pump operation, and long-distance power supply." This method identifies potential pump efficiency degradation patterns—characterized by "micro-voltage fluctuations, stable current, and decreased flow"—that exist only under specific conditions by comparing subtle changes in voltage, flow rate, and current under different operating conditions. Based on this, the weighting of the voltage fluctuation impact term in the drainage reliability value calculation model is adaptively optimized, enabling early identification of latent degradation characteristics in the system. Compared to existing monitoring methods that rely on alarm thresholds, this invention can detect early performance degradation without triggering traditional anomaly conditions, thereby improving the prediction accuracy and response timeliness of coal mine underground drainage systems in complex environments and significantly enhancing the operational safety and reliability of the drainage system. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the abnormal perception and early warning response analysis method for coal mine underground operation data provided in Embodiment 1 of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0034] Please see Figure 1 This embodiment proposes a method for anomaly detection and early warning response analysis of underground coal mine operation data, including the following steps:

[0035] Step S100: When the target pump in the coal mine underground drainage system is identified to be in a specified operating condition of high head, single pump operation and long-distance power supply, the historical monitoring data of the drainage system and the initial weight ratio of the voltage fluctuation effect are obtained to generate the drainage reliability value for the coal mine underground drainage system.

[0036] In this embodiment of the invention, the underground drainage system in a coal mine refers to an electromechanical integrated system used to collect and drain accumulated water in underground roadways, working faces, pump rooms, and low-lying areas. It typically includes a bottom water tank, a booster pump, a main drainage pipeline, valve assemblies, power supply lines, and monitoring and control devices. The main function of this system is to achieve continuous drainage during coal mine production or sudden seepage or water inrush accidents, maintaining the safe operation of roadways and working faces. The target pump refers to the core pump group that undertakes the main drainage task in this drainage system; that is, the pump equipment selected as the research object during the current monitoring period. It is usually the main drainage pump or booster pump installed in the main drainage roadway, and its operating status directly affects the reliability of the entire drainage system.

[0037] "Designated operating conditions of high head, single pump operation, and long-distance power supply" represent a typical high-load operating state in coal mine drainage systems. When the elevation difference between the underground water accumulation point and the surface discharge outlet is significant, the drainage pump must provide a high head to deliver water to the surface or intermediate water tank. When the system operates independently with only one main pump, all drainage loads are borne by that pump, significantly increasing operational risks. "Long-distance power supply" means long power lines, large voltage drops, and more pronounced fluctuations in power quality. This operating condition often occurs in deep mines or remote mining areas, and is closely related to underground coal mine operations because the reliability of drainage capacity directly determines the safety conditions and flood prevention levels of the mining area. If the pump operates abnormally or its efficiency degrades under this condition, it will directly lead to underground water accumulation, equipment shutdown, and even disaster risks.

[0038] Historical monitoring data originates from online monitoring and automated control devices in coal mine underground drainage systems. This data is typically collected by various sensors and monitoring terminals installed in pump rooms, distribution cabinets, and pipeline nodes, and transmitted in real-time to the ground monitoring center via the mine monitoring system or industrial Ethernet. The data generally includes electrical parameters (voltage, current, power factor, power supply frequency), hydraulic parameters (flow rate, head, outlet pressure, pipeline differential pressure), operating condition parameters (pump start-up and shutdown time, load rate, operating time), and environmental parameters (underground temperature, humidity, power supply line impedance, etc.).

[0039] The drainage reliability value is a comprehensive evaluation index used to reflect the ability of a coal mine underground drainage system to maintain normal drainage under specific operating conditions. It represents the probability or health level of the system's continuous and stable drainage within a given time. This index is usually calculated by comprehensively weighting multi-source monitoring data and can be used to determine the system's safety level or predict potential failure risks. The drainage reliability value itself is a mature existing technology concept, widely used in fields such as mine electromechanical equipment health management and water supply pump station status assessment. In existing technologies, this index is sometimes also referred to as the "drainage system health index," "pump station reliability coefficient," or "equipment status evaluation value." Traditional methods of obtaining this value mainly rely on expert experience weighting methods or reliability models based on statistical characteristics. However, this invention achieves higher evaluation accuracy through adaptive correction of the weighting of voltage fluctuation effects.

[0040] The initial weighting of voltage fluctuations refers to the initial proportion of each influencing factor (such as voltage fluctuations and power transmission equipment stability) contributing to the overall system reliability when calculating drainage reliability values. This weight reflects the sensitivity of each parameter to the reliability results. In the initial model, voltage fluctuations are usually assigned a fixed weight to characterize the impact of power supply quality on pump performance and system stability. Based on this, this invention modifies the initial weighting proportion through sample comparison and pattern identification to better reflect actual operating characteristics. In addition to voltage fluctuations, the drainage reliability model can also include multiple indicators such as flow rate attenuation weight, motor temperature rise weight, and vibration noise weight, which together constitute a weighted evaluation system for drainage reliability values.

[0041] Step S200: Analyze historical monitoring data and select a number of sample data that are consistent with the current background environment of the target pump and whose voltage fluctuation and flow rate drop fluctuation are within the allowable and preset range. Select an equal number of first-class samples and second-class samples, where the first-class samples correspond to the specified operating conditions and the second-class samples correspond to non-specified operating conditions.

[0042] It should be noted that "consistent with the current background environment of the target pump" means that the environmental conditions at the time of sample data collection match the external operating conditions, electrical power supply conditions, and hydraulic load conditions of the target pump at the time of operation, and the differences between the parameters do not exceed the preset tolerance range.

[0043] The statement that both voltage fluctuation and flow rate drop fluctuation are within acceptable limits means that the voltage fluctuation amplitude is lower than the preset abnormal alarm threshold and within the rated voltage tolerance range, while the flow rate drop amplitude is less than the preset allowable error range and does not trigger the preset drainage performance abnormality judgment condition.

[0044] In this embodiment of the invention, the core function of step S200 is to construct a comparative sample set that can truly reflect the performance differences of the target pump under different operating conditions through refined analysis and rigorous screening of historical monitoring data, providing a reliable data foundation for subsequent pattern identification and weight correction. This step is not only a data preparation step, but also a key step in ensuring the credibility of the analysis results and the effectiveness of pattern identification.

[0045] In practical implementation, the first step is to perform multi-dimensional analysis of historical monitoring data from the coal mine underground drainage system. Key parameters such as voltage, current, flow rate, head, and pipeline pressure are extracted from the raw dataset, and time-series synchronization and data cleaning are performed to eliminate abnormal, missing, or noisy data. Subsequently, historical data is filtered according to the current operating conditions of the target pump to ensure that the selected samples are consistent with the current operating environment of the target pump.

[0046] The phrase "consistent with the current background environment of the target pump" means that the external environmental conditions at the time of sample data collection match the current operating conditions, electrical power supply conditions, and hydraulic load conditions of the target pump. For example, factors such as the voltage level of the power supply busbar, line length, power supply topology, drainage pipe pressure, head height, and operating water temperature of the pump house should all be within the same or equivalent operating condition range. This type of screening ensures that different samples are compared on the same environmental basis, so that the subsequent analysis results reflect the differences in the operating status of the target pump, rather than statistical biases caused by differences in external conditions. Controlling the differences between parameters within a preset tolerance range (e.g., voltage difference not exceeding ±3%, head difference not exceeding ±5%) ensures the consistency of background conditions.

[0047] After screening, the samples were divided into two categories: the first category corresponded to the operating data of the target pump under specified conditions of "high head, single pump operation, and long-distance power supply," while the second category corresponded to the operating data under unspecified conditions. Although these samples shared a consistent external environment, the internal operating states of the pumps differed. For example, when the mine water level dropped or the standby pump was activated, the pump's load characteristics might change from "high head, single pump operation" to "dual pumps in parallel, low head." In other words, under the same power supply line and hydraulic system environment, the target pump might experience different operating modes at different times, thus forming comparable samples of "same environment, different operating conditions." This design allows subsequent pattern identification to be performed only on the differences in pump operating states, under the premise of consistent external conditions, thereby ensuring the effectiveness and relevance of pattern identification.

[0048] The requirement that "voltage fluctuations and flow rate drops are both within allowable and preset ranges" is to avoid interference from abnormal events or extreme disturbances in the analysis of patterns. In actual operation, when voltage fluctuations exceed the rated tolerance range or flow rate drops exceed the alarm threshold, the system usually triggers protection or alarm mechanisms. This type of data cannot reflect the true state of the pump, which is "normal but potentially degraded." This invention aims to identify subtle abnormal features that show signs of efficiency degradation but have not yet triggered alarms. Therefore, limiting voltage fluctuations and flow rate drops to allowable ranges ensures that the analyzed samples are all microscopic changes under normal operating conditions. This screening method helps to discover potential problems that are "seemingly normal but actually degraded," improving the sensitivity of anomaly detection and early warning response.

[0049] In addition to the screening criteria mentioned above, auxiliary screening parameters such as pump runtime, start-stop frequency, load rate, and power factor can be introduced during implementation. For example, the sample collection time can be limited to the stable operation phase of the pump (e.g., data collected 30 minutes after start-stop), or data on short-term load fluctuations caused by external scheduling can be excluded. These screening criteria further improve the consistency and analytical accuracy of the dataset, enabling the present invention to extract representative operating patterns based on high-confidence samples, providing solid support for subsequent weight correction and risk assessment.

[0050] Step S300: Analyze whether there is a specified pattern, that is, in the first type of sample, when there are voltage fluctuations and flow rate decreases, the power supply current of the target pump remains stable, while in the second type of sample, although the power supply current of the target pump remains stable in some second type of sample, this phenomenon is random and the frequency of occurrence is lower than the preset threshold.

[0051] When the existence of a specified pattern is verified in several first-class and second-class samples, it indicates that in the underground drainage system of a coal mine, under the specified operating conditions of high head, single pump operation and long-distance power supply, although the voltage fluctuation and flow rate drop are within the allowable range, an abnormal flow rate drop occurs when the power supply current of the target pump remains stable. This indicates that the motor control system of the target pump is not responsive enough to small voltage fluctuations, which indirectly verifies that the pump efficiency of the target pump degrades under the specified operating conditions.

[0052] In this embodiment of the invention, step S300 is the core and innovative part of the entire method. Its main purpose is to identify a unique operating pattern that only appears under the specific working condition of "high head, single pump operation, and long-distance power supply" through comparative analysis of different types of samples, thereby revealing the potential efficiency degradation mechanism of the target pump in the coal mine drainage system under specific conditions. The discovery and verification of this pattern breaks through the inherent understanding of "normal fluctuations equal safety" in existing technologies, and provides a new dimension for anomaly identification in the health monitoring of underground pumping stations.

[0053] In underground coal mine drainage systems, minor voltage fluctuations and slight decreases in drainage flow are common phenomena due to factors such as long power supply line distances, high line impedance, humid and hot environments, complex ventilation, and electromagnetic interference. Existing monitoring systems generally consider voltage fluctuations below the abnormal alarm threshold and flow rate decreases within the allowable error range as normal operating fluctuations, which will not trigger warnings or interventions. This logic is reasonable in most operating conditions because short-term fluctuations in the system do not affect the overall pump performance.

[0054] However, after comparing and analyzing a large amount of operational data from coal mines, those skilled in the art discovered that under the specified operating conditions of "high head, single pump operation, and long-distance power supply," the target pump exhibits a peculiar behavior: although there are slight voltage fluctuations within the permissible range, the pump's supply current remains stable for an extended period, without adjusting accordingly to voltage changes. Simultaneously, the flow rate shows a slight decreasing trend. This indicates that the motor's control system or speed control unit fails to respond sensitively to minor voltage fluctuations, leading to a decrease in the pump's output mechanical power and thus a potential degradation in pump efficiency. In other words, the voltage fluctuations are "masked" by the system; although the pump does not exhibit obvious abnormalities, its energy efficiency characteristics have deviated from the normal curve.

[0055] This phenomenon is not significant in ordinary low-head or dual-pump parallel operation conditions because, under these conditions, the impact of system voltage fluctuations on pump operating characteristics can be mitigated through load sharing or pipeline compensation. However, in single-pump, high-head, long-feed operation conditions, voltage fluctuations are amplified by line impedance during long-distance power transmission, but the monitoring end can only detect the superficial appearance of slight voltage fluctuations and stable current, making it difficult to detect the hidden losses in pump efficiency. Therefore, the identification of this specified law is highly scenario-specific and is a unique phenomenon under the complex operating conditions of underground coal mine drainage systems.

[0056] To verify the existence and universality of this pattern, step S300 combines the two types of samples selected in step S200 for systematic analysis. During sample selection, it was ensured that the background environments of the two types of samples were consistent, eliminating interference from differences in external conditions, so that the comparison results only reflect the inherent differences in the target pump's operating state. Based on this, the correlation pattern among "voltage fluctuation, flow rate decrease, and current change" in the two types of samples was analyzed:

[0057] In the first type of sample, the target pump's supply current remained stable throughout when voltage fluctuated and flow rate decreased, demonstrating a high degree of consistency. In the second type of sample, although similar situations occasionally occurred, the phenomenon was randomly distributed, and its frequency was significantly lower than the preset threshold. This difference indicates that the correlation between voltage fluctuations, current stability, and flow rate decrease is statistically significant only under specified operating conditions, constituting an identifiable characteristic pattern.

[0058] If cross-validation of several sets of samples confirms the existence of this pattern, it can be inferred that under specified operating conditions, the target pump's motor control system exhibits hysteresis or mismatch in response to minute voltage fluctuations, leading to reduced energy conversion efficiency and resulting in abnormal flow rate decline. This not only reflects pump efficiency degradation but also implies a gradual decrease in drainage reliability over a continuous operating period.

[0059] Therefore, the significance of step S300 is as follows:

[0060] On the one hand, it enables early identification of potential performance degradation of pumps by verifying the patterns in the samples;

[0061] On the other hand, it provides data for subsequent weight correction, enabling the dynamic amplification of the weight of voltage fluctuation factors when calculating drainage reliability values, and more realistically reflecting the actual operating health of the target pump under specific working conditions.

[0062] In summary, steps S300 and S200 form a logical closed loop: the former ensures data comparability through rigorous sample screening, while the latter achieves scientific identification of pump efficiency degradation through pattern verification. The synergistic effect of these two steps constitutes the core innovation of this invention: under complex underground coal mine conditions, through feature comparison and pattern identification of multi-source data, it reveals hidden performance degradation phenomena that traditional monitoring systems have failed to identify, thereby achieving higher-precision anomaly perception and early warning response analysis.

[0063] Step S400: If a specified pattern exists, calculate the average fluctuation range of the current change of the second type of samples compared to the first type of samples, and optimize the initial weight ratio accordingly, specifically including:

[0064] Step S401: After determining that a specified pattern exists, determine the current data of each first-class sample and the second-class sample during operation.

[0065] Step S402: Select the current fluctuation segment in each second type of sample that has a stable current waveform compared to the corresponding first type of sample, and calculate the current fluctuation amplitude of each current fluctuation segment compared to the stable current waveform.

[0066] Step S403: Average all the fluctuation amplitudes to obtain the average fluctuation amplitude of the current change, and use the average fluctuation amplitude as a correction factor to optimize the initial weight ratio to obtain the optimized weight ratio.

[0067] Step S404: Apply the optimized weight ratio to the generation model of drainage reliability value, and correct the weight of the voltage fluctuation impact item (adverse) to update the overflow risk assessment result of the coal mine underground drainage system.

[0068] The current waveforms of several samples of the first type remain consistent and exhibit a stable flat state under specified operating conditions.

[0069] When optimizing the initial weight ratio, a preset correction function is used, which is defined as follows:

[0070] ;

[0071] in, This refers to the optimized weighting percentage. This refers to the initial weighting percentage. This represents the total number of current fluctuation segments identified in all second-class samples. It refers to the first The second type of sample current data corresponding to each current fluctuation segment. This refers to the steady-state current data of the first type of sample under a steady-state current waveform. This refers to the average fluctuation range. This refers to the preset correction strength coefficient, and Greater than 0.

[0072] In this embodiment of the invention, step S400 constitutes a crucial step in achieving adaptive weight optimization. It is a key step in transitioning from feature recognition to dynamic correction. Its main purpose is to propose a dynamic weight correction mechanism based on current fluctuation differences, thereby enabling the drainage reliability value of the underground coal mine drainage system to be adaptively optimized according to the degradation trend of pump efficiency under specific operating conditions during the calculation process. Through this step, the invention transforms the specified pattern of "voltage fluctuation - current stability - flow rate decrease" identified in step S300 into quantitative correction parameters, realizing a closed-loop feedback mechanism from anomaly identification to weight optimization.

[0073] In a specific implementation, steps S401 to S404 constitute the complete logical chain of the correction process.

[0074] In step S401, after determining the existence of a specified pattern, the sample data is first standardized to determine the current data sequences of each type of sample (first category and second category) within the corresponding operating cycle. To ensure data comparability, existing time-series alignment algorithms (such as the timestamp-based Dynamic Time Warping (DTW) algorithm or interpolation alignment algorithm) can be used to ensure that the current waveforms of different samples correspond to the same sampling time point or drainage stage. This step is a mature data processing technology and can be directly implemented through existing industrial monitoring systems or data analysis platforms.

[0075] In step S402, the system identifies and selects current fluctuation segments in each second-type sample that differ from the stable current waveform of the corresponding first-type sample. This identification can be achieved through fluctuation detection algorithms, such as current deviation judgment based on a sliding window, variance threshold judgment, or derivative rate of change detection. When the current value in a certain time segment deviates from the corresponding stable waveform by more than a small fluctuation judgment threshold, that segment can be identified as a current fluctuation segment. Subsequently, the fluctuation amplitude of each segment relative to the stable waveform is calculated, and the obtained amplitude values ​​are all positive, used to quantify the degree of deviation of the current change.

[0076] In step S403, the system averages the amplitude values ​​of all identified current fluctuation segments to obtain the average fluctuation amplitude of the current change. This average fluctuation amplitude reflects the overall deviation intensity of the current in the second type of samples compared to the stable waveform of the first type of samples, and is a key indicator characterizing the stability of pump operation. Next, the system uses this average fluctuation amplitude as a correction factor to optimize the initial weight ratio, obtaining the optimized weight ratio. The correction method can employ linear functions, proportional functions, or adaptive functions based on confidence weights, etc. This invention preferably uses a proportional correction model, that is, using the product of the average fluctuation amplitude and the correction intensity coefficient as the correction value, thereby achieving better response sensitivity while maintaining model simplicity. This calculation method is intuitive and operable, and can be directly implemented in the data analysis module through mathematical calculations or programming algorithms.

[0077] In step S404, the optimized weighting is applied to the generation model of the drainage reliability value, and the weights of the voltage fluctuation impact term are corrected. The corrected model can more realistically reflect the actual impact of voltage fluctuations on pump operation, thereby dynamically updating the overflow risk assessment results of the coal mine underground drainage system. This model update process can be implemented through existing multi-factor weighted evaluation frameworks, such as reliability calculation modules based on weighted average, Bayesian fusion, or fuzzy comprehensive evaluation methods, which are existing and directly implementable calculation methods. It should be noted that this embodiment adopts a conditional discriminative generation model, dynamically selecting the calculation path based on system performance. The model expression is:

[0078]

[0079] In the formula, R is the drainage reliability value. S represents the initial weight percentage of the i-th parameter. i Let $t$ be the base state value of the $i$-th parameter, and $\{t}$ be the time decay factor. The fusion coefficient is... Let f(S) be the initial weight percentage of the i-th parameter. i ) is a nonlinear mapping function, D is a deviation function, M represents the system state, M=1 indicates that the system is in a latent abnormal mode, and M=0 indicates that the system is in a normal or explicit abnormal mode.

[0080] The weight correction function employed in this invention is intuitive and engineering-feasible. The idea is to quantify the mismatch in current response by performing a differential analysis on the current fluctuation range of the second type of sample and the stable range of the first type of sample, and then correct the weight of the voltage fluctuation influence term using the average fluctuation amplitude. This method can be considered a "self-correction mechanism based on relative deviation," with the advantage of requiring no additional sensors or complex modeling, relying solely on existing monitoring data to achieve dynamic weight optimization. Furthermore, besides the proportional function, this invention can also employ other mathematical expressions for weight correction, such as a correction function based on exponential decay, a regression correction model based on confidence intervals, or a nonlinear mapping correction achieved through a neural network fitting function—all alternative solutions achievable by those skilled in the art.

[0081] By implementing step S400, this invention achieves closed-loop control from feature pattern identification to quantitative correction, enabling the drainage reliability assessment model to have adaptive learning capabilities. This mechanism effectively overcomes the analytical deficiency in existing technologies that considers "voltage and flow fluctuations within allowable ranges as safe," and can proactively detect and respond to pump operating states that are "appearing normal but potentially degraded."

[0082] Therefore, the overall beneficial effect of this invention is that, through dynamic correlation analysis of the power supply current, drainage flow rate, and voltage fluctuations of the target pump under specific operating conditions, an adaptively adjustable drainage reliability calculation mechanism is constructed. This mechanism not only improves the sensitivity of anomaly detection but also enables the quantitative identification of pump efficiency degradation, thereby enhancing the risk prevention and control capabilities of underground drainage systems in coal mines under complex environments.

[0083] From an application perspective, this invention can be widely applied to the operational health monitoring and early warning analysis of main drainage systems in coal mines, local drainage systems in mining areas, and surface pumping stations. Its method does not rely on specific hardware modifications, possesses good compatibility and scalability, and can be directly embedded into existing mine automation monitoring platforms, industrial IoT systems, or remote pumping station diagnostic systems, demonstrating significant promotional value and practical significance.

[0084] Example 2:

[0085] This embodiment proposes an anomaly detection and early warning response analysis system for underground coal mine operation data, which can realize the anomaly detection and early warning response analysis method for underground coal mine operation data described in Embodiment 1, including:

[0086] Data acquisition module: used to acquire the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of coal mines;

[0087] Operating condition detection module: used to determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply based on the current operating condition;

[0088] Weight correction module: In response to the target pump being in the specified operating condition, based on the historical monitoring data, sample classification, current fluctuation range selection and fluctuation amplitude calculation are performed in sequence, and then the initial weight ratio of the influence of voltage fluctuation is corrected in combination with the correction function.

[0089] Weight update module: This module applies the corrected weight ratios to the preset drainage reliability value generation model, corrects the weights of the voltage fluctuation impact term, and updates the overflow risk assessment results of the coal mine underground drainage system.

[0090] Example 3:

[0091] This invention also provides an anomaly perception and early warning response analysis device for underground coal mine operation data, which can realize the anomaly perception and early warning response analysis method for underground coal mine operation data described in Embodiment 1, including a processor and a storage medium;

[0092] The storage medium is used to store instructions;

[0093] The processor is configured to operate according to the instructions to perform the steps of the following method:

[0094] Obtain the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of the coal mine.

[0095] Based on the current operating conditions, determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply;

[0096] In response to the target pump being in the specified operating condition, after sequentially classifying samples, selecting current fluctuation ranges, and calculating fluctuation amplitude based on the historical monitoring data, the initial weight ratio of the influence of voltage fluctuation is corrected by combining the correction function;

[0097] The corrected weighting is applied to the preset model for generating drainage reliability values, the weight of the voltage fluctuation impact term is corrected, and the overflow risk assessment results of the coal mine underground drainage system are updated.

[0098] Example 4:

[0099] This invention also provides a computer-readable storage medium that can implement the anomaly detection and early warning response analysis method for underground coal mine operation data as described in Embodiment 1. The medium stores a computer program that, when executed by a processor, performs the following steps:

[0100] Obtain the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of the coal mine.

[0101] Based on the current operating conditions, determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply;

[0102] In response to the target pump being in the specified operating condition, after sequentially classifying samples, selecting current fluctuation ranges, and calculating fluctuation amplitude based on the historical monitoring data, the initial weight ratio of the influence of voltage fluctuation is corrected by combining the correction function;

[0103] The corrected weighting is applied to the preset model for generating drainage reliability values, the weight of the voltage fluctuation impact term is corrected, and the overflow risk assessment results of the coal mine underground drainage system are updated.

[0104] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0105] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0106] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0108] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for anomaly detection and early warning response analysis of underground coal mine operation data, characterized in that, include: Obtain the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of the coal mine. Based on the current operating conditions, determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply; In response to the target pump being in the specified operating condition, after sequentially classifying samples, selecting current fluctuation ranges, and calculating fluctuation amplitude based on the historical monitoring data, the initial weight ratio of the influence of voltage fluctuation is corrected by combining the correction function; This step includes: The historical monitoring data is analyzed, and several sample data are selected that are consistent with the current background environment of the target pump and whose voltage fluctuation and flow rate drop fluctuation are within the allowable and preset range. An equal number of first-class samples and second-class samples are selected, where the first-class samples correspond to the specified operating conditions and the second-class samples correspond to non-specified operating conditions. After determining the current data of each first-class sample and second-class sample during operation, the current fluctuation segment in each second-class sample that is relatively stable compared to the corresponding first-class sample is selected, and the current fluctuation amplitude of each current fluctuation segment compared to the stable current waveform is calculated. The average fluctuation amplitude of all the fluctuation amplitudes is obtained by averaging all the fluctuation amplitudes. This average fluctuation amplitude is then used as a correction factor to optimize the initial weight ratio, resulting in the optimized weight ratio. The correction function is: ； in, This refers to the optimized weighting percentage. This refers to the initial weighting percentage. This represents the total number of current fluctuation segments identified in all second-class samples. It refers to the first The second type of sample current data corresponding to each current fluctuation segment. This refers to the steady-state current data of the first type of sample under a steady-state current waveform. This refers to the preset correction strength coefficient; The corrected weighting is applied to the preset model for generating drainage reliability values, the weight of the voltage fluctuation impact term is corrected, and the overflow risk assessment results of the coal mine underground drainage system are updated.

2. The method for anomaly perception and early warning response analysis of underground coal mine operation data according to claim 1, characterized in that, The consistency with the current background environment of the target pump includes: the environmental conditions at the time of sample data collection are consistent with the external operating conditions, electrical power supply conditions and hydraulic load conditions of the target pump at present, and the differences between the parameters do not exceed the preset tolerance range.

3. The method for anomaly perception and early warning response analysis of underground coal mine operation data according to claim 1, characterized in that, The voltage fluctuation and flow rate drop fluctuation are both within the allowable and preset range, including: the voltage fluctuation amplitude is lower than the preset abnormal alarm threshold and within the rated voltage tolerance range, while the flow rate drop amplitude is less than the preset allowable error range and does not trigger the preset drainage performance abnormality judgment condition.

4. The method for anomaly perception and early warning response analysis of underground coal mine operation data according to claim 1, characterized in that, The historical monitoring data is collected by various sensors and monitoring terminals installed in pump rooms, power distribution cabinets and pipeline nodes, and transmitted in real time to the ground monitoring center through the mine monitoring system or industrial Ethernet. The data includes electrical parameters, hydraulic parameters, operating condition parameters and environmental parameters.

5. The method for anomaly perception and early warning response analysis of underground coal mine operation data according to claim 1, characterized in that, The initial weighting percentages for the impact of voltage fluctuations include: the initial set percentages of each influencing factor's contribution to the overall system reliability when calculating the drainage reliability value using the generated model.

6. An anomaly detection and early warning response analysis system for underground coal mine operation data, characterized in that, The method for anomaly detection and early warning response analysis of underground coal mine operation data as described in any one of claims 1-5 includes: Data acquisition module: used to acquire the current operating status, historical monitoring data, and initial weighting of the impact of voltage fluctuations on the target pump in the underground drainage system of coal mines; Operating condition detection module: used to determine whether the target pump is in a specified operating condition of high head, single pump operation and long-distance power supply based on the current operating condition; Weight correction module: In response to the target pump being in the specified operating condition, based on the historical monitoring data, sample classification, current fluctuation range selection and fluctuation amplitude calculation are performed in sequence, and then the initial weight ratio of the influence of voltage fluctuation is corrected in combination with the correction function. Weight update module: This module applies the corrected weight ratios to the preset drainage reliability value generation model, corrects the weights of the voltage fluctuation impact term, and updates the overflow risk assessment results of the coal mine underground drainage system.

7. An anomaly detection and early warning response analysis device for underground coal mine operation data, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 5.

Images