A coal mine power supply system fault monitoring method based on a smart grid

Abstract

This invention relates to the field of fault monitoring technology for coal mine power supply systems, specifically disclosing a fault monitoring method for coal mine power supply systems based on a smart grid. This method comprehensively monitors the operating status of key equipment by collecting operational information from multiple sensing parameters in real time. An initial health score is then set for the key equipment. Abnormal signals are identified when sensing parameters do not reach a preset fault alarm threshold but exceed a preset normal operating threshold. The abnormal score is then confirmed based on a scoring benchmark set by an expert experience database, and the initial health score is adjusted accordingly to obtain an actual health score. The actual health score is used to determine whether there is a risk of fault in the key equipment. By adjusting the health score, this application, through multi-dimensional data collection and expert experience-based decision-making, can provide early warnings of hidden faults from operational data, reducing the risk of unplanned power outages and improving the operational stability of the coal mine power supply system.

Description

Technical Field

[0001] This application relates to the field of fault monitoring technology for coal mine power supply systems, and more specifically, to a method and system for fault monitoring of coal mine power supply systems using a smart grid. Background Technology

[0002] Coal mine power supply systems are critical infrastructure for ensuring mine production and safety, and their operational reliability directly affects underground work safety and production efficiency. With the deep integration of smart grid technology and intelligent mine construction, higher demands are placed on the safe, continuous, and stable operation of coal mine power supply systems. However, traditional fault monitoring and protection methods, due to their reliance on fixed thresholds and human experience, are inadequate in the face of drastic load changes, grid structure adjustments, and complex operating conditions commonly encountered in coal mine environments. They struggle to detect problems promptly and respond accurately, severely impacting power supply stability.

[0003] Specifically, the current fault monitoring of coal mine power supply systems mainly suffers from the following problems: First, it relies heavily on manual inspections and periodic debugging, making it impossible to respond in real time to instantaneous changes in the power grid status, resulting in delays in fault diagnosis. Second, protection settings are usually based on offline calculations and static settings, and the settings cannot be adaptively adjusted, making it impossible to respond to early characteristics of faults, thus leading to a high risk of unplanned power outages.

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] This application aims to solve the technical problems mentioned in the background art by providing a fault monitoring method for coal mine power supply systems based on smart grids, which aims to solve the problem of the lag in traditional fault monitoring methods for coal mine power supply systems.

[0006] To solve the above-mentioned technical problems, the solution proposed in this application is as follows:

[0007] A fault monitoring method for a coal mine power supply system based on a smart grid is disclosed. This method is applied to a coal mine power supply system comprising multiple key devices, including at least one of a transformer, a high-voltage switch, and a main cable.

[0008] Real-time operational information of various sensing parameters is collected using multiple types of sensing devices deployed on critical equipment, wherein the multiple sensing parameters include at least one of current, voltage, power factor, temperature, vibration, and insulation status.

[0009] Set an initial health score for key equipment, and identify abnormal signals such as sensing parameters not reaching the preset fault alarm threshold and sensing parameters exceeding the preset normal operation threshold based on the real-time operation information of the key equipment.

[0010] If an abnormality exists, the abnormal score is confirmed by comparing the identified abnormal signal with the scoring benchmark set based on the expert experience database. The initial health score is then adjusted based on the abnormal score to obtain the actual health score.

[0011] The actual health score is used to determine whether critical equipment is at risk of failure.

[0012] Furthermore, the use of various types of sensing devices deployed on key equipment to collect real-time operational information of various sensing parameters includes:

[0013] Determine the type of critical equipment and extract the fault detection parameters associated with that critical equipment from the prior database based on the type of critical equipment;

[0014] Multiple types of sensing devices deployed on key equipment based on fault detection parameters are used to collect real-time operating information of various sensing parameters according to a preset collection cycle. Among them, after the real-time operating information of various sensing parameters is collected, a digital filtering algorithm is used to remove electromagnetic interference noise, and a cyclic redundancy check algorithm is used to verify the integrity of the real-time operating information of various sensing parameters after noise removal. When the verification fails, the real-time operating information of the sensing parameters that failed the verification is confirmed, and the real-time operating information of the sensing parameters that failed the verification is re-collected.

[0015] Furthermore, the step of confirming the abnormal score based on the identified abnormal signal and a scoring benchmark set based on an expert experience database, and adjusting the initial health score based on the abnormal score to obtain the actual health score, includes:

[0016] Identify the duration and amplitude of abnormal signals;

[0017] The abnormal score is determined by comparing the duration and fluctuation amplitude of the abnormal signal with a scoring benchmark set based on an expert experience database.

[0018] The initial health score is adjusted based on the abnormal score to obtain the actual health score.

[0019] Furthermore, the initial health score is set based on the equipment model, operating years, historical fault records, and maintenance frequency of the key equipment;

[0020] Furthermore, the initial health score is calibrated and updated once according to the set calibration update cycle.

[0021] Furthermore, in the step of setting an initial health score for key equipment and identifying abnormal signals where sensing parameters have not reached a preset fault alarm threshold and have exceeded a preset normal operation threshold based on the real-time operating information of the key equipment, the identified abnormal signals are adjusted according to the following steps:

[0022] Acquire external operating status information associated with key equipment and synchronize the external operating status information with the real-time operating information of the key equipment. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information.

[0023] The parameter interference signals caused by load fluctuations, power grid fluctuations, and environmental fluctuations are extracted from the external operating status information.

[0024] By removing parameter interference signals from the identified abnormal signals, the true abnormal signals of the key equipment can be obtained.

[0025] Furthermore, in the step of setting an initial health score for key equipment and identifying abnormal signals where sensing parameters have not reached a preset fault alarm threshold and have exceeded a preset normal operation threshold based on the real-time operating information of the key equipment, the preset fault alarm threshold and the preset normal operation threshold are adjusted according to the following steps:

[0026] Within a set time window, real-time operating information of key equipment and external operating status information associated with the key equipment are acquired. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information.

[0027] A reference operating condition set is constructed based on the real-time operating information of key equipment and the external operating status information associated with the key equipment. The reference operating condition set includes multiple reasonable operating ranges of various sensing parameters of the key equipment to adapt to the current operating conditions.

[0028] Adjust the preset fault alarm threshold and preset normal operation threshold according to the reference operating condition set.

[0029] Furthermore, the determination of whether critical equipment has a failure risk based on the actual health score includes:

[0030] The actual health score is obtained using the set acquisition period interval, and the trend of the actual health score is obtained.

[0031] Based on the changing trend of the actual health score, match the abnormal signals that are associated with the changing trend of the current actual health score from the identified abnormal signals;

[0032] The current actual health score change trend and associated abnormal signals are combined and compared with the preset risk evolution path. Based on the comparison results, it is determined whether there is a failure risk in the key equipment. The preset risk evolution path is a pre-established failure development model corresponding to the decline trend of health score and the occurrence of abnormal signals.

[0033] Furthermore, the process of combining the current actual health score trend with associated abnormal signals and comparing it with a preset risk evolution path, and determining whether key equipment has a failure risk based on the comparison results, includes:

[0034] Sort the associated abnormal signals according to their occurrence order;

[0035] The current actual health score change trend is combined with the sorted abnormal signals and compared with the preset risk evolution path. The preset risk evolution path of key equipment includes the time sequence of different abnormal signals.

[0036] The comparison results are used to determine whether there is a risk of failure in key equipment.

[0037] Furthermore, the determination of whether critical equipment has a failure risk based on the actual health score includes:

[0038] The actual health score is compared with the preset warning score, and the comparison results are used to determine whether there is a risk of failure in key equipment.

[0039] Specifically, when the comparison result indicates that the actual health score is lower than the preset warning score, an early warning is triggered and a preventive maintenance report containing key equipment, fault risk types, abnormal signals, and maintenance recommendations is generated.

[0040] Furthermore, the step of comparing the actual health score with the preset warning score and determining whether there is a risk of failure in the key equipment based on the comparison result includes:

[0041] The preventive maintenance report is pushed to the management terminal of the coal mine power supply system. The operation and maintenance personnel who operate the management terminal of the coal mine power supply system confirm receipt and record the receipt time.

[0042] Specifically, if a preventive maintenance report is not confirmed within a preset time window, the alert level of the currently pushed preventive maintenance report will be upgraded and it will be re-pushed, repeating this cycle until the maintenance personnel confirm it.

[0043] This application discloses a fault monitoring method for coal mine power supply systems based on smart grids. By deploying various types of sensing devices on key equipment, it collects real-time operational information on multiple sensing parameters, including current, voltage, power factor, temperature, vibration, and insulation status, achieving comprehensive and real-time monitoring of the operational status of key equipment in the coal mine power supply system. The method sets an initial health score for key equipment and identifies abnormal signals based on real-time operational information—that is, situations where sensing parameters do not reach a preset fault alarm threshold but exceed a preset normal operation threshold. If an abnormal signal exists, the abnormal score is confirmed according to a scoring benchmark set by an expert experience database, and the initial health score is adjusted based on this abnormal score to obtain an actual health score. Finally, the actual health score is used to determine whether the key equipment has a fault risk. This application, by adjusting the health score, can reflect changes in equipment operating status, overcoming the shortcomings of static protection settings that cannot adapt to complex operating conditions. Simultaneously, through multi-dimensional data collection and expert experience-based decision-making, it can provide early warnings of hidden faults from operational data, reducing the risk of unplanned power outages and improving the operational stability of the coal mine power supply system. Attached Figure Description

[0044] Figure 1 A flowchart illustrating a fault monitoring method for a coal mine power supply system based on a smart grid, provided in this application embodiment. Detailed Implementation

[0045] To better illustrate the present invention, the invention will now be described in further detail with reference to the accompanying drawings.

[0046] It should be understood that, in order to make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The components of the embodiments of this disclosure described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed disclosure, but merely represents selected embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure.

[0047] The following description uses at least one specific embodiment as an example. In this embodiment:

[0048] like Figure 1As shown, a fault monitoring method for a coal mine power supply system based on a smart grid is provided. This method is applied to a coal mine power supply system, which includes multiple key devices, including at least one of a transformer, a high-voltage switch, and a main cable. The method includes the following steps:

[0049] S1. Use multiple types of sensing devices deployed on key equipment to collect real-time operating information of multiple sensing parameters, wherein the multiple sensing parameters include at least one of current, voltage, power factor, temperature, vibration and insulation status.

[0050] S2. Set an initial health score for key equipment and identify abnormal signals such as sensing parameters not reaching the preset fault alarm threshold and sensing parameters exceeding the preset normal operation threshold based on the real-time operating information of the key equipment.

[0051] S3. If it exists, the abnormal score is confirmed by comparing the identified abnormal signal with the scoring benchmark set based on the expert experience database. The initial health score is adjusted based on the abnormal score to obtain the actual health score.

[0052] S4. Determine whether critical equipment is at risk of failure based on the actual health score.

[0053] This application introduces various types of sensing devices to collect real-time operational information of key equipment and combines it with a health score assessment mechanism to achieve early warning of fault risks in coal mine power supply systems and improve the safety of power supply systems.

[0054] First, various types of sensing devices deployed on critical equipment are used to collect real-time operational information on a variety of sensing parameters. Specifically, various sensors can be installed on critical equipment, such as temperature and vibration sensors on transformers, current and voltage sensors on high-voltage switches, and insulation status sensors on main cables. These sensors can periodically collect real-time operational data from the equipment, for example, data every minute.

[0055] Next, an initial health score is set for key equipment, and based on the real-time operating information of the key equipment, abnormal signals are identified where the sensed parameters have not reached the preset fault alarm threshold, but have exceeded the preset normal operation threshold. The initial health score can be set according to factors such as the type, model, and years of operation of the equipment. For example, a newly commissioned transformer can be set with a higher initial health score. During the monitoring process, the system continuously receives real-time operating information collected by the sensing devices and compares it with the preset normal operation threshold and fault alarm threshold. For example, if the temperature value collected by the temperature sensor of a transformer is consistently higher than the preset normal operation threshold, but has not yet reached the preset fault alarm threshold, it can be identified as an abnormal signal.

[0056] If an abnormal signal is detected, the system determines the abnormal score by comparing the identified abnormal signal with a scoring benchmark set based on an expert experience database. This abnormal score is then used to adjust the initial health score to obtain the actual health score. For example, when an abnormal signal of persistently high transformer temperature is detected, the system will query the expert experience database. This database may contain scoring benchmarks such as "a persistently high temperature for 1 hour results in an abnormal score of 5; a persistently high temperature for 3 hours results in an abnormal score of 15." Based on these benchmarks, an abnormal score can be determined. Then, the initial health score is subtracted from this abnormal score to obtain the actual health score of the equipment.

[0057] Finally, the system determines whether critical equipment is at risk of failure based on the actual health score. For example, if the actual health score of equipment is lower than a preset risk threshold, the system can determine that the equipment is at risk of failure and trigger the corresponding early warning mechanism.

[0058] The fault monitoring method for coal mine power supply systems based on smart grid proposed in this application achieves early warning of fault risks of key equipment in coal mine power supply systems through real-time acquisition of multi-dimensional sensing parameters, dynamic evaluation of health scores, and intelligent identification and processing of abnormal signals.

[0059] This application, by deploying various types of sensing devices, can acquire real-time and comprehensive operational information of key equipment, including current, voltage, power factor, temperature, vibration, and insulation status, thereby improving the real-time nature and richness of data acquisition.

[0060] Furthermore, this application introduces a health score mechanism, setting an initial health score for critical equipment and dynamically adjusting it based on real-time operational information to obtain the actual health score. This health score assessment method can quantitatively reflect the health status of the equipment, avoiding the over-reliance on single parameters or fixed thresholds in traditional methods. When a sensed parameter becomes abnormal, the system can identify abnormal signals that have exceeded the preset normal operation threshold but have not reached the preset fault alarm threshold. This allows the system to issue warnings before a fault occurs, i.e., when the equipment operating parameters deviate slightly, thus buying time for preventative maintenance.

[0061] In some of the embodiments described above in this application, it is proposed to use various types of sensing devices deployed on key equipment to collect real-time operational information of various sensing parameters. However, if the collection process is not managed in a refined manner and data quality control is not performed during its implementation, the collected operational information may be subject to noise interference, incomplete data, or mismatch with equipment characteristics, thereby affecting the accuracy of subsequent fault identification.

[0062] To this end, the use of various types of sensing devices deployed on key equipment to collect real-time operational information on various sensing parameters includes:

[0063] Determine the type of critical equipment and extract the fault detection parameters associated with that critical equipment from the prior database based on the type of critical equipment;

[0064] Multiple types of sensing devices deployed on key equipment based on fault detection parameters are used to collect real-time operating information of various sensing parameters according to a preset collection cycle. Among them, after the real-time operating information of various sensing parameters is collected, a digital filtering algorithm is used to remove electromagnetic interference noise, and a cyclic redundancy check algorithm is used to verify the integrity of the real-time operating information of various sensing parameters after noise removal. When the verification fails, the real-time operating information of the sensing parameters that failed the verification is confirmed, and the real-time operating information of the sensing parameters that failed the verification is re-collected.

[0065] Specifically, determining the type of critical equipment refers to identifying whether the critical equipment to be monitored is a specific type of equipment such as a transformer, high-voltage switch, or main cable. Based on the identified critical equipment type, fault detection parameters closely related to that equipment type are extracted from a pre-established prior database.

[0066] The use of various types of sensing devices deployed on key equipment based on fault detection parameters to collect real-time operational information of multiple sensing parameters at preset collection cycles can be understood as selectively selecting and deploying corresponding sensing devices based on the extracted fault detection parameters. For example, if the fault detection parameters include temperature, a temperature sensor is deployed; if they include vibration, a vibration sensor is deployed. These sensing devices continuously collect real-time operational information of the corresponding sensing parameters at preset collection cycles, such as per second, per minute, or longer time intervals.

[0067] In practical applications, digital filtering algorithms are used to remove electromagnetic interference clutter from the real-time operational information of various sensing parameters after acquisition. The purpose is to eliminate the impact of electromagnetic noise, which is common in industrial environments, on data quality. For example, digital filtering techniques such as low-pass filters, median filters, or Kalman filters can be used to process the raw acquired data to obtain smoother and more accurate operational information.

[0068] Furthermore, a cyclic redundancy check (CRC) algorithm is used to verify the integrity of the real-time operating information of various denoised sensing parameters. The purpose is to detect whether errors or corruption have occurred during data transmission or storage. When the verification fails, the real-time operating information of the failed sensing parameter is confirmed, and this information is re-acquired. Once the CRC check detects an error, the system identifies which part of the sensing parameter's real-time operating information is problematic and triggers a re-acquisition command to ensure that the operating information of all critical devices is accurate and complete.

[0069] By identifying key equipment types and extracting associated fault detection parameters, the deployment of sensing devices and data collection become more targeted, avoiding the blind collection of irrelevant data and thus improving the efficiency and effectiveness of data acquisition. Simultaneously, digital filtering algorithms effectively remove electromagnetic interference noise, ensuring the purity of the original data. Based on this, a cyclic redundancy check algorithm is used to verify the integrity of the denoised data, and data that fails the check is re-collected, constructing a data quality assurance mechanism to ensure the accuracy and reliability of real-time operational information.

[0070] In some embodiments described above, this application proposes a method for setting an initial health score for critical equipment, identifying abnormal signals based on real-time operational information, and then adjusting the initial health score according to the abnormal signals to obtain the actual health score. However, in practical applications, simply identifying the presence of abnormal signals cannot fully and accurately reflect the severity and urgency of equipment failures. For example, a brief, small-amplitude abnormal signal has a drastically different impact on the equipment's health status compared to a long-lasting, highly fluctuating abnormal signal, which may lead to insufficient accuracy in adjusting the health score, thereby affecting the reliability of fault risk assessment.

[0071] In response, the process of confirming abnormal scores by comparing the identified abnormal signals with a scoring benchmark set based on an expert experience database, and adjusting the initial health score based on these abnormal scores to obtain the actual health score, includes:

[0072] Identify the duration and amplitude of abnormal signals;

[0073] The abnormal score is determined by comparing the duration and fluctuation amplitude of the abnormal signal with a scoring benchmark set based on an expert experience database.

[0074] The initial health score is adjusted based on the abnormal score to obtain the actual health score.

[0075] Specifically, identifying the duration and amplitude of abnormal signals means that after detecting an abnormal signal in which the sensing parameters have not reached the preset fault alarm threshold and have exceeded the preset normal operation threshold, the system will continuously monitor the duration of the abnormal signal and record its maximum or average fluctuation amplitude during the duration. The duration reflects the persistence of the abnormal state, while the fluctuation amplitude reflects the degree to which the abnormality deviates from the normal range.

[0076] Furthermore, the anomaly score is determined by comparing the duration and amplitude of the abnormal signal with a scoring benchmark established based on an expert experience database. This expert experience database pre-stores a large amount of historical fault data, equipment operating patterns, and expert evaluation rules regarding the relationship between different abnormal signal characteristics (such as duration and amplitude) and fault risk levels. By comparing the duration and amplitude of the currently identified abnormal signal with the scoring benchmark in the database, an anomaly score can be quantitatively determined. For example, the longer the duration and the greater the amplitude of the abnormal signal, the higher its corresponding anomaly score, indicating a greater fault risk.

[0077] Therefore, the initial health score is adjusted based on the confirmed anomaly score to obtain the actual health score. This adjustment process can be implemented using various mathematical models or algorithms. For example, the anomaly score can be subtracted from the initial health score as a deduction, or a weighted average can be used for fusion. The goal is to ensure that the health score accurately reflects the true health status of the device after an anomaly signal occurs.

[0078] In some preferred embodiments, a specific example is given below. Suppose that during the operation of a transformer, its temperature sensing parameter suddenly rises, exceeding a preset normal operating threshold but not yet reaching a preset fault alarm threshold, and this is identified as an abnormal signal. The solution of this application further monitors the duration and fluctuation amplitude of this abnormal signal. For example, if the abnormal temperature signal lasts for 5 minutes and the fluctuation amplitude is 10% of the normal operating temperature, the system will determine a lower abnormal score, such as 5 points, based on the scoring benchmark set in the expert experience database. If the abnormal temperature signal lasts for 30 minutes and the fluctuation amplitude reaches 25% of the normal operating temperature, a higher abnormal score, such as 20 points, will be determined.

[0079] Specifically, assume the transformer's initial health score is 90.

[0080] In the first scenario (lasting 5 minutes, fluctuating by 10%), if the confirmed abnormal score is 5, the actual health score is adjusted to 85.

[0081] In the second scenario (lasting 30 minutes, fluctuating by 25%), if the confirmed abnormal score is 20, the actual health score is adjusted to 70.

[0082] In this way, the system can dynamically and accurately adjust the health scores of critical equipment based on the actual severity of abnormal signals, thus providing a more reliable basis for subsequent fault risk assessment. For example, a health score of 85 may indicate a minor risk, while a health score of 70 may indicate a moderate risk, requiring initial intervention.

[0083] Preferably, the initial health score is set based on the equipment model, operating years, historical fault records, and maintenance frequency of the key equipment; and the initial health score is calibrated and updated once according to the set calibration update cycle.

[0084] The initial health score is understood as a baseline assessment of the current health status of critical equipment. Multiple dimensions of data are considered when setting this score. Specifically, the equipment model is a crucial factor. Different models may differ in design life, material properties, rated parameters, and known defects, and these inherent attributes directly affect their initial health level. For example, transformers may have higher initial reliability due to their design features. Service life refers to the length of time the equipment has been in use. As service life increases, equipment typically experiences natural wear and tear, aging, and performance degradation. Therefore, equipment with longer service lives may have a lower initial health score. Historical fault records provide direct evidence of anomalies or failures that occurred during past operation. Equipment with a history of faults, even after repair, should have a lower initial health score than equipment that has never failed, reflecting its potential risks. Maintenance frequency reflects the level of investment in routine maintenance and repair. Regular and standardized maintenance helps extend equipment life and maintain good operating conditions; therefore, equipment with high maintenance frequency usually achieves a relatively higher initial health score.

[0085] Furthermore, to ensure the timeliness and accuracy of the initial health score, it is set to be calibrated and updated once according to a preset calibration update cycle. This calibration update cycle can be determined based on the type of equipment, its importance, operating environment, and empirical data; for example, it can be set monthly, quarterly, or annually. At the end of each calibration update cycle, the system reassesses information such as the equipment model, years of operation, latest historical fault records, and maintenance frequency of key equipment. Combining this with the latest operating data and expert experience, the initial health score is adjusted and corrected to more accurately reflect the true health status of the equipment at the current stage.

[0086] Traditional fault monitoring methods for coal mine power supply systems rely primarily on comparing sensing parameters with preset thresholds to identify abnormal signals from critical equipment. However, in actual operation, the real-time operating information of critical equipment is affected by external factors such as load fluctuations in the coal mine power supply system, changes in power grid operating parameters, and environmental parameter fluctuations. These external factors may cause non-fault-related fluctuations in sensing parameters, resulting in false anomalies caused by external interference in the initially identified abnormal signals, or masking genuine equipment malfunctions.

[0087] In response, this application proposes a method for adjusting the identified abnormal signals. By acquiring external operating status information associated with key equipment and synchronizing this information in time, parameter interference signals caused by external factors are extracted and eliminated, thereby obtaining more realistic abnormal signals and improving the accuracy of fault monitoring.

[0088] In response, the steps of setting an initial health score for key equipment and identifying abnormal signals where sensing parameters either fail to reach a preset fault alarm threshold or exceed a preset normal operation threshold based on the real-time operating information of the key equipment are adjusted according to the following steps:

[0089] Acquire external operating status information associated with key equipment and synchronize the external operating status information with the real-time operating information of the key equipment. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information.

[0090] The parameter interference signals caused by load fluctuations, power grid fluctuations, and environmental fluctuations are extracted from the external operating status information.

[0091] By removing parameter interference signals from the identified abnormal signals, the true abnormal signals of the key equipment can be obtained.

[0092] Specifically, external operating status information refers to non-equipment internal sensing parameters related to the operating environment and conditions of critical equipment. This information is used to understand the operating context of critical equipment. For example, load status information of a coal mine power supply system includes the total load and the variation curves of the load of each branch, reflecting the system's demand for electricity; power grid operating parameter information includes fluctuations in grid voltage and frequency, reflecting the stability of the external power grid; environmental parameter information includes ambient temperature, humidity, dust concentration, etc., in the area where the equipment is located, and these factors directly affect the equipment's heat dissipation, insulation performance, etc.

[0093] Synchronizing external operational status information with the real-time operational information of key equipment refers to ensuring consistency between these two types of information on the timeline for accurate correlation analysis. This can be achieved through a unified timestamp mechanism, GPS time synchronization, or Network Time Protocol (NTP), ensuring that the internal sensing parameters of the equipment collected at the same point in time correspond precisely with the external operational status information. The purpose is to provide an accurate time reference for subsequent extraction of parameter interference signals.

[0094] In practical applications, parameter interference signals refer to non-fault-related effects on the sensing parameters of critical equipment caused by factors such as load fluctuations, power grid fluctuations, and environmental fluctuations in external operating status information. For example, when the load in a coal mine suddenly increases, the current and temperature of the transformer may rise accordingly, but this is not a manifestation of a fault in the transformer itself, but rather a response under normal operating conditions.

[0095] Furthermore, removing parameter interference signals from the identified abnormal signals to obtain the true abnormal signals of key equipment refers to correcting these signals using the extracted parameter interference signals after initially identifying abnormal signals where the perceived parameters exceed a preset threshold. For example, if an abnormal current signal is identified at a certain time point, and a significant load fluctuation causing current interference signal is also found at that time point, this interference signal can be subtracted from or compensated for in the abnormal signal. This allows for a determination of whether the remaining signal still constitutes an abnormality or whether it is a true precursor to equipment failure. The aim is to eliminate the interference of external factors on the judgment of abnormal signals and improve the authenticity and accuracy of abnormal signals.

[0096] In some preferred embodiments, a specific example is given below. Suppose that in a coal mine power supply system, a transformer's winding temperature sensing parameter remains consistently higher than a preset normal operation threshold for a certain period, but has not yet reached a preset fault alarm threshold, and is initially identified as an abnormal signal. To determine whether this abnormal signal is a genuine equipment malfunction, the system simultaneously acquires external operating status information associated with the transformer. Specifically, the system acquires the load status information of the coal mine power supply system during that period (e.g., a sudden increase in total load due to the start-up of underground mining equipment), grid operating parameter information (e.g., slight fluctuations in grid voltage), and environmental parameter information (e.g., an increase in the ambient temperature of the transformer room). This external operating status information is synchronized with the transformer's real-time operating information. Subsequently, the system analyzes and extracts the parameter interference signals of the transformer winding temperature caused by the sudden increase in load, grid voltage fluctuations, and ambient temperature increases, based on historical data and a preset model. For example, the model calculates that under the current load and environmental conditions, a 5°C increase in transformer winding temperature is a normal response. If the initial identified abnormal signal indicates an 8°C increase in winding temperature, then after eliminating the 5°C interference signal, the actual abnormal signal indicates a 3°C increase in winding temperature. The system will then use this 3°C actual abnormal signal to adjust the subsequent health score and assess fault risk. In this way, the system can distinguish between a normal temperature increase caused by changes in the external environment and a genuine abnormality caused by a potential fault inside the transformer (such as localized overheating), thus avoiding misjudgments caused by external factors.

[0097] Traditional fault monitoring methods for coal mine power supply systems typically employ preset fixed fault alarm thresholds and normal operation thresholds to identify abnormal signals. However, the operating conditions of coal mine power supply systems are complex and variable, influenced by various external factors such as load fluctuations, changes in power grid operating parameters, and environmental factors (e.g., temperature and humidity), causing the operating parameters of critical equipment to change dynamically. Relying solely on static thresholds may lead to false alarms (identifying normal fluctuations as abnormalities) or missed alarms (failing to detect true anomalies in a timely manner) under specific operating conditions, thus affecting the accuracy and reliability of fault monitoring. To address this, this application proposes a method for dynamically adjusting preset fault alarm thresholds and preset normal operation thresholds to improve the adaptability and accuracy of fault monitoring.

[0098] Furthermore, in the step of setting an initial health score for key equipment and identifying abnormal signals where sensing parameters have not reached a preset fault alarm threshold and have exceeded a preset normal operation threshold based on the real-time operating information of the key equipment, the preset fault alarm threshold and the preset normal operation threshold are adjusted according to the following steps:

[0099] Within a set time window, real-time operating information of key equipment and external operating status information associated with the key equipment are acquired. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information.

[0100] A reference operating condition set is constructed based on the real-time operating information of key equipment and the external operating status information associated with the key equipment. The reference operating condition set includes multiple reasonable operating ranges of various sensing parameters of the key equipment to adapt to the current operating conditions.

[0101] Adjust the preset fault alarm threshold and preset normal operation threshold according to the reference operating condition set.

[0102] Specifically, setting a time window refers to a predetermined period of time, such as the most recent hour, day, or week, used to collect sufficient data to reflect the recent operational characteristics of critical equipment. By acquiring real-time operational information of critical equipment within this time window, the timeliness of the analyzed data can be ensured.

[0103] External operating status information refers to non-equipment parameters that affect the operating status of critical equipment. These include load status information of the coal mine power supply system, power grid operating parameters, and environmental parameters. Load status information reflects the current electricity demand of coal mine production, such as total power and load distribution across branches. Power grid operating parameters can include grid voltage, frequency, and harmonic content, reflecting the stability of the power supply. Environmental parameters include ambient temperature, humidity, and altitude, factors that directly affect the equipment's heat dissipation performance and insulation characteristics. This external information is crucial for understanding the normal operating range of critical equipment under specific conditions.

[0104] In practical applications, a "reference operating condition set" is constructed by comprehensively analyzing the real-time operating information of key equipment and its associated external operating status information. This reference operating condition set includes multiple reasonable operating ranges corresponding to various sensing parameters (such as current, voltage, temperature, vibration, etc.) of key equipment under different external operating conditions. For example, under high temperature and high load conditions, the normal operating temperature range of a transformer may be higher than that under low temperature and low load conditions.

[0105] Therefore, once a reference set of operating conditions is established, the preset fault alarm threshold and preset normal operation threshold can be matched with the corresponding reasonable operating range from the reference set according to the current operating conditions of the key equipment, and dynamically adjusted accordingly, so as to achieve adaptive adjustment based on the actual operating environment and load conditions.

[0106] In some preferred embodiments, a specific example is given below. Suppose that in a coal mine power supply system, a main transformer needs to be monitored for faults. Traditional methods might set a fixed normal operating temperature threshold (e.g., below 75°C) and a fault alarm threshold (e.g., above 85°C). However, during hot summer months and when coal mine production loads are high, the transformer's normal operating temperature may reach 78°C. If a fixed threshold is still used, 78°C might be misjudged as abnormal, triggering unnecessary alarms.

[0107] The proposed solution will be adjusted as follows:

[0108] First, within a set time window, the system continuously acquires real-time operating information of the transformer (such as winding temperature, oil temperature, load current, etc.) and external operating status information. External operating status information may include ambient temperature (e.g., 35°C in summer), total coal mine load (e.g., 80% of rated load), and grid voltage stability.

[0109] Secondly, based on this real-time operating information and external operating status information, the system will construct or update the reference operating condition set for the transformer. For example, under the operating condition of "ambient temperature 30-40°C, load 70%-90%", the reference operating condition set may indicate that the reasonable operating range for the transformer winding temperature is 70-82°C.

[0110] Finally, based on the current actual operating conditions (e.g., ambient temperature 35°C, load 80%), the system will match the corresponding reasonable operating range from the reference operating condition set, and dynamically adjust the preset normal operating threshold to 82°C and the preset fault alarm threshold to 88°C accordingly.

[0111] In this way, when the transformer winding temperature reaches 78°C, the system will not trigger an abnormal alarm because it is still within the dynamically adjusted normal operating range (70-82°C), thus avoiding false alarms. Meanwhile, if the temperature further rises to 85°C, the system will accurately identify the anomaly because it exceeds the normal operating threshold under the current conditions but has not yet reached the fault alarm threshold, indicating a potential risk. This dynamic adjustment mechanism makes fault monitoring more accurate and intelligent.

[0112] In some embodiments described above, this application proposes determining the risk of failure in critical equipment based on actual health scores. However, in practice, relying solely on a single actual health score threshold may fail to accurately capture early signs of failure evolution and differentiate between different types of failure risks, potentially leading to misjudgments or delayed warnings. For example, if the actual health score of critical equipment is slowly declining but has not yet reached a preset warning score, neglecting its trend and accompanying abnormal signals could result in missing the optimal time for preventative maintenance. Failure to address these issues could lead to insufficient accuracy and foresight in failure risk assessment, impacting the safe and stable operation of coal mine power supply systems.

[0113] In this regard, determining whether critical equipment has a risk of failure based on actual health scores includes:

[0114] The actual health score is obtained using the set acquisition period interval, and the trend of the actual health score is obtained.

[0115] Based on the changing trend of the actual health score, match the abnormal signals that are associated with the changing trend of the current actual health score from the identified abnormal signals;

[0116] The current actual health score change trend and associated abnormal signals are combined and compared with the preset risk evolution path. Based on the comparison results, it is determined whether there is a failure risk in the key equipment. The preset risk evolution path is a pre-established failure development model corresponding to the decline trend of health score and the occurrence of abnormal signals.

[0117] Specifically, acquiring actual health scores using a set acquisition period interval means that the system periodically retrieves the actual health scores of critical equipment from the health score acquisition module at predetermined time intervals (e.g., hourly, daily, or weekly). By continuously acquiring these scores, a time series data can be constructed, and then data analysis methods (e.g., moving average, regression analysis, or trend line fitting) can be used to calculate and identify the changing trend of the actual health score. This trend indicates whether the health score is stable, declining slowly, declining rapidly, or fluctuating. The acquisition period interval can be flexibly adjusted according to the operating characteristics, importance, and data acquisition capabilities of the critical equipment to ensure timely reflection of changes in health status.

[0118] Furthermore, matching abnormal signals associated with the current actual health score's changing trend from the identified abnormal signals means that after obtaining the actual health score's changing trend, the system reviews and analyzes all abnormal signals identified by the abnormal signal identification module within the same time period or a period close to the time when the trend occurred. Using techniques such as timestamps, event correlation, or pattern matching, abnormal signals that are temporally synchronized with or causally related to the health score's changing trend are filtered out as abnormal signals associated with the current actual health score's changing trend. For example, if the health score shows a continuous downward trend, and abnormal vibrations at a specific frequency or a local temperature increase are detected during this period, these abnormal signals are considered to be associated with the downward trend in the health score.

[0119] The process of combining the current actual health score trend with associated abnormal signals and comparing it with a preset risk evolution path involves integrating the obtained actual health score trend (e.g., slow decline, accelerated decline) and associated abnormal signals (e.g., continuous exceedance of specific parameters, sequence of abnormal events) to form a comprehensive fault characteristic pattern. This comprehensive pattern is then input into the fault risk assessment module and compared with the preset risk evolution path. The preset risk evolution path is pre-established based on a large amount of historical data, expert experience, and fault mechanism analysis. It describes the typical process of different types of faults from their inception to development, including how the health score declines and the types and sequences of abnormal signals that may appear at different stages. For example, a preset risk evolution path might be described as: slow decline in health score → intermittent current fluctuations → accelerated decline in health score → continuous temperature anomalies → ultimately leading to equipment failure. By matching the current actual operating mode of the equipment with these preset paths, the current stage of fault development and potential fault type of the equipment can be determined.

[0120] In some preferred embodiments, it is assumed that a transformer's health score has shown a slow but continuous downward trend over the past month. During this period, the operation information acquisition module detects a slight but frequent increase in the partial discharge signal intensity of the transformer within a specific time period. Simultaneously, the oil temperature sensor records a slight increase in oil temperature within the normal fluctuation range, with an increased fluctuation frequency. The fault risk assessment module combines this information (health score decline trend, abnormal partial discharge signal, and abnormal oil temperature fluctuation) and compares it with a preset risk evolution path for transformer winding insulation aging faults. This preset path may be described as: slow decline in health score → appearance of partial discharge signal → slight increase in oil temperature and increased fluctuation → potentially leading to insulation breakdown. Through comparison, the system finds that the current operating status of the transformer highly matches the preset path, thereby determining that the transformer has a fault risk of winding insulation aging and predicting its possible development direction, triggering preventive maintenance recommendations in a timely manner.

[0121] In some of the embodiments described above in this application, although combining the changing trend of the actual health score with the associated abnormal signals and comparing it with a preset risk evolution path can preliminarily determine whether there is a risk of failure in critical equipment, in the actual failure evolution process, different types of abnormal signals often appear in a specific time sequence, forming a unique failure development pattern. If abnormal signals are simply combined without considering their order of appearance, the identification of the failure evolution path may be inaccurate, thereby affecting the accuracy and timeliness of the failure risk assessment.

[0122] In response, the process involves combining the current actual health score trend with associated abnormal signals and comparing it with a preset risk evolution path. Based on the comparison results, it is determined whether critical equipment has a failure risk, including:

[0123] Sort the associated abnormal signals according to their occurrence order;

[0124] The current actual health score change trend is combined with the sorted abnormal signals and compared with the preset risk evolution path. The preset risk evolution path of key equipment includes the time sequence of different abnormal signals.

[0125] The comparison results are used to determine whether there is a risk of failure in key equipment.

[0126] Specifically, after acquiring abnormal signals associated with the actual health score change trend, these abnormal signals are arranged according to their chronological order of appearance on the time axis. For example, if a temperature abnormality is detected first, followed by a vibration abnormality, then after sorting, the temperature abnormality will be ranked before the vibration abnormality. This temporal sorting ensures that the combination of abnormal signals can accurately reflect the dynamic process of fault development. The preset risk evolution path for critical equipment is a pre-established fault development model, which not only includes the correspondence between the health score decline trend and abnormal signals, but more importantly, clearly defines the temporal order of different abnormal signals. For example, a certain type of bearing failure might manifest as initial slight vibration abnormality, followed by a temperature increase, and finally severe vibration. The preset risk evolution path will contain this specific temporal information. In practical applications, the sorted abnormal signals are combined with the actual health score change trend to form a comprehensive feature containing time-series information. Subsequently, this comprehensive feature is precisely compared with the preset risk evolution path.

[0127] In some preferred embodiments, it is assumed that a transformer's health score shows a continuous downward trend during operation. The system first identifies associated abnormal signals, including: a partial discharge signal detected at time T1, a winding temperature increase detected at time T2, and a decrease in insulation resistance detected at time T3. According to the scheme of this application, these abnormal signals are sorted according to their occurrence sequence, i.e., partial discharge signal → winding temperature increase → insulation resistance decrease. Subsequently, the actual health score change trend is combined with the sorted abnormal signals and compared with a preset risk evolution path. If the preset "transformer insulation aging fault" risk evolution path clearly defines the timing pattern of "partial discharge signal before winding temperature increase, and then before insulation resistance decrease," then the system will be able to accurately match this fault pattern and determine that the transformer has a fault risk caused by insulation aging. In contrast, if these abnormal signals are simply combined without considering the timing, it may not be possible to accurately distinguish whether it is insulation aging or other types of faults, or it may not be a risk but rather a signal error, thus affecting the accuracy of the judgment.

[0128] In some of the embodiments described above in this application, it is proposed to determine whether there is a risk of failure in critical equipment based on the actual health score. However, in practical applications, simply determining whether there is a risk of failure may not be sufficient to provide timely and specific maintenance guidance, and it is difficult to achieve proactive preventive maintenance.

[0129] In this regard, determining whether critical equipment has a risk of failure based on actual health scores includes:

[0130] The actual health score is compared with the preset warning score, and the comparison results are used to determine whether there is a risk of failure in key equipment.

[0131] Specifically, when the comparison result indicates that the actual health score is lower than the preset warning score, an early warning is triggered and a preventive maintenance report containing key equipment, fault risk types, abnormal signals, and maintenance recommendations is generated.

[0132] Specifically, the preset warning score can be understood as a health score threshold set by the system based on factors such as historical data, equipment type, operating environment, and security level. It defines the critical point at which equipment transitions from normal operation to a potential fault state. When the actual health score of critical equipment falls below this preset warning score, it indicates that the equipment's health status has reached a level requiring attention and action. Triggering a warning means the system automatically issues an alarm message, such as through audible and visual alarms, SMS notifications, emails, or highlighting on the monitoring interface, to alert relevant maintenance personnel. Furthermore, the generation of preventative maintenance reports aims to provide maintenance personnel with detailed fault risk information and specific maintenance recommendations. This report includes the identification information of critical equipment, the currently identified fault risk type (e.g., overheating risk, insulation aging risk), abnormal signals leading to the risk (e.g., continuous high temperature, abnormal voltage fluctuations), and specific maintenance recommendations for these risks and abnormal signals (e.g., recommending checking the cooling system, recommending insulation testing).

[0133] In some preferred embodiments, suppose that during the operation of a transformer, the sensing parameter of its winding temperature continuously rises, causing its health score to gradually decrease from 90 points. When the actual health score of the transformer drops below a preset warning score (e.g., 70 points), the system will immediately trigger an audible and visual warning and notify the on-duty personnel via SMS. Simultaneously, the system will automatically generate a preventative maintenance report, which will clearly indicate that the transformer has an overheating risk, the abnormal signal being a persistently high winding temperature, and recommend that maintenance personnel immediately check the transformer's cooling system, such as whether the cooling fan is working properly and whether the oil level is sufficient. Following the report's guidance, maintenance personnel will promptly maintain the cooling system, thereby effectively preventing transformer failures due to overheating and ensuring the continuity of power supply.

[0134] Furthermore, the step of comparing the actual health score with the preset warning score and determining whether there is a risk of failure in the key equipment based on the comparison result includes:

[0135] The preventive maintenance report is pushed to the management terminal of the coal mine power supply system. The operation and maintenance personnel who operate the management terminal of the coal mine power supply system confirm receipt and record the receipt time.

[0136] Specifically, if a preventive maintenance report is not confirmed within a preset time window, the alert level of the currently pushed preventive maintenance report will be upgraded and it will be re-pushed, repeating this cycle until the maintenance personnel confirm it.

[0137] After comparing the actual health score with the preset warning score to determine the corresponding fault risk of critical equipment, the system automatically generates and pushes a preventive maintenance report to the corresponding management terminal of the coal mine power supply system for maintenance personnel to review. Maintenance personnel confirm and obtain the preventive maintenance report by operating the management terminal, and the system simultaneously records the confirmation and acquisition time for subsequent tracking of the receipt of warning information.

[0138] The system has a pre-set time window threshold. If a preventative maintenance report is not confirmed by maintenance personnel within this preset time window, the system automatically upgrades the warning level of the report and re-pushes it to the management terminal according to the upgraded warning level. This warning level upgrade and re-pushing process is repeated until maintenance personnel confirm the preventative maintenance report, thus forming a complete closed-loop control of warning information. This ensures that fault risk warning information effectively reaches maintenance personnel, preventing abnormal operation of critical equipment in the coal mine power supply system due to information omissions or untimely responses, and guaranteeing the safe and stable operation of the coal mine power supply system.

[0139] Based on the above, this application discloses a fault monitoring method for coal mine power supply systems based on smart grids. By deploying various types of sensing devices on key equipment, it collects real-time operational information on multiple sensing parameters such as current, voltage, power factor, temperature, vibration, and insulation status, achieving comprehensive and real-time monitoring of the operational status of key equipment in the coal mine power supply system. This method sets an initial health score for key equipment and identifies abnormal signals based on real-time operational information, i.e., situations where the sensing parameters do not reach the preset fault alarm threshold but exceed the preset normal operation threshold. If an abnormal signal exists, the abnormal score is confirmed according to the scoring benchmark set by the expert experience database, and the initial health score is adjusted based on this abnormal score to obtain the actual health score. Finally, the actual health score is used to determine whether there is a fault risk in the key equipment. This application, by adjusting the health score, can reflect changes in the equipment's operational status, overcoming the shortcomings of static protection settings that cannot adapt to complex operating conditions. Simultaneously, through multi-dimensional data collection and expert experience-based decision-making, it can provide early warnings of hidden faults from operational data, reducing the risk of unplanned power outages and improving the operational stability of the coal mine power supply system.

[0140] Finally, it should be noted that the above-described embodiments are merely specific implementations of this disclosure, used to illustrate the technical solutions of this disclosure, and not to limit them. The protection scope of this disclosure is not limited thereto. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features within the scope of the technology disclosed in this disclosure. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure, and should all be covered within the protection scope of this disclosure.

Claims

1. A fault monitoring method for a coal mine power supply system based on a smart grid, wherein the method is applied in a coal mine power supply system, the coal mine power supply system comprising multiple key devices, wherein, The plurality of key equipment includes at least one of transformers, high-voltage switches, and main cables, characterized in that it includes: Real-time operational information of various sensing parameters is collected using multiple types of sensing devices deployed on critical equipment, wherein the multiple sensing parameters include at least one of current, voltage, power factor, temperature, vibration, and insulation status. Set an initial health score for key equipment, and identify abnormal signals such as sensing parameters not reaching the preset fault alarm threshold and sensing parameters exceeding the preset normal operation threshold based on the real-time operation information of the key equipment. If an abnormality exists, the abnormal score is confirmed by comparing the identified abnormal signal with the scoring benchmark set based on the expert experience database. The initial health score is then adjusted based on the abnormal score to obtain the actual health score. The actual health score is used to determine whether critical equipment is at risk of failure.

2. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, The use of various types of sensing devices deployed on key equipment to collect real-time operational information on various sensing parameters includes: Determine the type of critical equipment and extract the fault detection parameters associated with that critical equipment from the prior database based on the type of critical equipment; Multiple types of sensing devices deployed on key equipment based on fault detection parameters are used to collect real-time operating information of various sensing parameters according to a preset collection cycle. Among them, after the real-time operating information of various sensing parameters is collected, a digital filtering algorithm is used to remove electromagnetic interference noise, and a cyclic redundancy check algorithm is used to verify the integrity of the real-time operating information of various sensing parameters after noise removal. When the verification fails, the real-time operating information of the sensing parameters that failed the verification is confirmed, and the real-time operating information of the sensing parameters that failed the verification is re-collected.

3. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, The process of confirming abnormal scores based on the identified abnormal signals and a scoring benchmark set according to an expert experience database, and adjusting the initial health score based on these abnormal scores to obtain the actual health score, includes: Identify the duration and amplitude of abnormal signals; The abnormal score is determined by comparing the duration and fluctuation amplitude of the abnormal signal with a scoring benchmark set based on an expert experience database. The initial health score is adjusted based on the abnormal score to obtain the actual health score.

4. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 3, characterized in that: The initial health score is set based on the equipment model, operating years, historical fault records, and maintenance frequency of the key equipment; Furthermore, the initial health score is calibrated and updated once according to the set calibration update cycle.

5. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, In the step of setting an initial health score for key equipment and identifying abnormal signals where sensing parameters have not reached a preset fault alarm threshold and have exceeded a preset normal operation threshold based on the real-time operating information of the key equipment, the identified abnormal signals are adjusted according to the following steps: Acquire external operating status information associated with key equipment and synchronize the external operating status information with the real-time operating information of the key equipment. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information. The parameter interference signals caused by load fluctuations, power grid fluctuations, and environmental fluctuations are extracted from the external operating status information. By removing parameter interference signals from the identified abnormal signals, the true abnormal signals of the key equipment can be obtained.

6. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, In the step of setting an initial health score for critical equipment and identifying abnormal signals where sensing parameters have not reached a preset fault alarm threshold and have exceeded a preset normal operation threshold based on the real-time operating information of the critical equipment, the preset fault alarm threshold and the preset normal operation threshold are adjusted according to the following steps: Within a set time window, real-time operating information of key equipment and external operating status information associated with the key equipment are acquired. The external operating status information includes load status information of the coal mine power supply system, power grid operating parameter information, and environmental parameter information. A reference operating condition set is constructed based on the real-time operating information of key equipment and the external operating status information associated with the key equipment. The reference operating condition set includes multiple reasonable operating ranges of various sensing parameters of the key equipment to adapt to the current operating conditions. Adjust the preset fault alarm threshold and preset normal operation threshold according to the reference operating condition set.

7. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, The method of determining whether critical equipment has a risk of failure based on the actual health score includes: The actual health score is obtained using the set acquisition period interval, and the trend of the actual health score is obtained. Based on the changing trend of the actual health score, match the abnormal signals that are associated with the changing trend of the current actual health score from the identified abnormal signals; The current actual health score change trend and associated abnormal signals are combined and compared with the preset risk evolution path. Based on the comparison results, it is determined whether there is a failure risk in the key equipment. The preset risk evolution path is a pre-established failure development model corresponding to the decline trend of health score and the occurrence of abnormal signals.

8. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 7, characterized in that, The process of combining the current actual health score change trend with associated abnormal signals and comparing it with a preset risk evolution path, and determining whether key equipment has a failure risk based on the comparison results, includes: Sort the associated abnormal signals according to their occurrence order; The current actual health score change trend is combined with the sorted abnormal signals and compared with the preset risk evolution path. The preset risk evolution path of key equipment includes the time sequence of different abnormal signals. The comparison results are used to determine whether there is a risk of failure in key equipment.

9. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 1, characterized in that, The method of determining whether critical equipment has a risk of failure based on the actual health score includes: The actual health score is compared with the preset warning score, and the comparison results are used to determine whether there is a risk of failure in key equipment. Specifically, when the comparison result indicates that the actual health score is lower than the preset warning score, an early warning is triggered and a preventive maintenance report containing key equipment, fault risk types, abnormal signals, and maintenance recommendations is generated.

10. The fault monitoring method for coal mine power supply systems based on smart grids according to claim 9, characterized in that, The process of comparing the actual health score with the preset warning score and determining whether there is a risk of failure in key equipment based on the comparison results includes: The preventive maintenance report is pushed to the management terminal of the coal mine power supply system. The operation and maintenance personnel who operate the management terminal of the coal mine power supply system confirm receipt and record the receipt time. Specifically, if a preventive maintenance report is not confirmed within a preset time window, the alert level of the currently pushed preventive maintenance report will be upgraded and it will be re-pushed, repeating this cycle until the maintenance personnel confirm it.

Images