An intelligent locking method for coal mine power supply system based on active defense

Abstract

The application relates to the technical field of coal mine power supply system safety, and particularly discloses a coal mine power supply system intelligent locking method based on active defense, which can comprehensively understand the fault occurrence process by continuously collecting the operation data of the coal mine power supply system, identifying abnormal events in combination with preset abnormal event judgment criteria, further tracing the abnormal events and forming an abnormal evolution path. Through quantitative analysis of the abnormal events, quantitative indexes for locking decision are obtained, and the operation risk level of the coal mine power supply system is determined according to preset risk level judgment rules, so that the system operation risk is evaluated, and the limitation of traditional manual experience parameter setting is avoided. Finally, when the operation risk level meets the active locking trigger condition, a locking instruction is issued to the isolation equipment of the coal mine power supply system, active intervention on the fault risk is realized, the safety of the coal mine power supply system is improved, and the safe operation and production of the mine are ensured.

Description

Technical Field

[0001] This application relates to the field of safety technology for coal mine power supply systems, and more specifically, to an intelligent interlocking method for coal mine power supply systems based on active defense. Background Technology

[0002] In coal mine production, the power supply system plays a crucial role, serving as the foundation for ensuring safe mine operation and smooth production. However, the complexity of the underground environment, coupled with frequent fluctuations in equipment load, presents severe challenges to the power supply system. Any electrical anomaly can quickly lead to power outages, damage expensive equipment, and even endanger the lives of workers. Therefore, ensuring the stability, reliability, and high safety of the power supply system is directly critical to safe production in underground operations.

[0003] Currently, most coal mine power supply systems employ protection measures, such as those for overcurrent, leakage, or short circuits, primarily by cutting off power after a fault has occurred. This approach has several significant drawbacks: First, protection actions often only activate after an accident, failing to detect the evolving trend of a fault as it develops. Second, based on the traditional protection feedback model, when an anomaly is detected, the feedback mechanism only activates after a fault occurs. This typically only issues an alarm, creating a dilemma of "being able to issue an alarm but not actively cutting off power," making it difficult to effectively handle complex situations such as sudden circuit overloads or equipment malfunctions. Furthermore, the parameter settings for protection devices rely on manual experience, making it difficult for the system to flexibly adjust according to changes in actual operating conditions. This can easily lead to unnecessary cascading power outages, disrupting normal production.

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

[0005] The purpose of this application is to provide an intelligent interlocking method for coal mine power supply systems based on active defense, which aims to solve the problems of passive protection measures, inability to provide early warning, difficulty in actively cutting off power supply, and reliance on manual experience for parameter setting in existing coal mine power supply systems.

[0006] To solve the above problems, the solution proposed in this application is as follows: A smart interlocking method for a coal mine power supply system based on active defense includes the following steps: Step S1: Continuously collect the operating data of the coal mine power supply system, wherein the operating data includes current data, voltage data, equipment temperature data and insulation resistance data; Step S2: Based on the collected operational data and combined with the preset abnormal event judgment criteria, identify whether there are abnormal events in the coal mine power supply system. If there are abnormal events, assign a timestamp to each identified abnormal event. Step S3: In response to any identified abnormal event, trace back all abnormal events before the abnormal event within the set time window, and combine and sort the abnormal events in chronological order to form an abnormal evolution path. Step S4: Identify each abnormal event in the abnormal evolution path, extract the equipment operation monitoring parameters associated with each abnormal event and the deviation of the equipment operation monitoring parameter values ​​from the baseline value, and perform quantitative analysis to obtain quantitative indicators for lockout decision-making. Step S5: Using quantitative indicators, the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of each abnormal event, determine the operation risk level of the coal mine power supply system according to the preset risk level judgment rules. Step S6: When the operational risk level meets the active interlocking trigger condition, an interlocking command is issued to the isolation equipment of the coal mine power supply system.

[0007] Furthermore, step S1 includes: Step S11a: Use multiple acquisition units deployed in the coal mine power supply system to collect the operation data of the coal mine power supply system; wherein, each acquisition unit is pre-set with the device number, circuit name and physical location it monitors when it is deployed, and after each acquisition unit completes the collection of operation data, it is stored in combination with the device number, circuit name and physical location corresponding to each acquisition unit.

[0008] Furthermore, step S1 includes: Step S11b: Obtain the raw operating data from the coal mine power supply system; Step S12b: Obtain the mechanical sensor data and environmental sensor data corresponding to the original operating data. The mechanical sensor data includes equipment vibration data and cable wear data, and the environmental sensor data includes downhole humidity data and rock conductivity data. Step S13b: Using the historical operating data of the coal mine power supply system and the quantitative correlation formula between the original operating data and mechanical and environmental factors established by the regression analysis algorithm, calculate and obtain the interference contribution of mechanical factors to the original operating data and the interference contribution of environmental factors to the original operating data. Among them, mechanical factors correspond to the mechanical state of the equipment reflected by mechanical sensor data, and environmental factors correspond to the underground environmental state reflected by environmental sensor data. Step S14b: Combining the interference contribution of mechanical factors to the original operating data, the interference contribution of environmental factors to the original operating data, and the collected mechanical sensor data and environmental sensor data, calculate the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data. Extract the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data from the original operating data to obtain the operating data of the coal mine power supply system used to characterize the electrical sensor data.

[0009] Furthermore, step S2 includes: Step S21: Compare the collected running data with the preset normal operating data range, and fit the trend of the running data in multiple consecutive collection periods to determine whether there is any collected running data that exceeds the preset normal operating data range, and whether the rate of change of the collected running data meets the preset data mutation rate. If it exists, determine that there is a corresponding abnormal event, and assign a timestamp to the identified abnormal event; The preset abnormal event judgment criteria include a preset normal operation data range and a preset data mutation rate.

[0010] Furthermore, after step S3, the following steps are included: Step A1: Identify and distinguish the causal abnormal event, the result abnormal event, and multiple process abnormal events from the abnormal evolution path. The result abnormal event is the abnormal event identified in response to the abnormal evolution path. The causal abnormal event is the abnormal event that initially occurs in the time sequence of the abnormal evolution path. The process abnormal event is the abnormal event that occurs between the result abnormal event and the causal abnormal event in the abnormal evolution path. Step A2: Identify the content of the causal abnormal event and the result abnormal event, and preliminarily confirm the process abnormal event that occurs when the causal abnormal event evolves into the result abnormal event based on the prior database; Step A3: Perform consistency matching between the confirmed process anomaly events and the process anomaly events identified in the anomaly evolution path; Step A4: When the matching results are consistent or the process abnormal event identified in the abnormal evolution path contains the confirmed process abnormal event, the abnormal evolution path is corrected. The correction of the abnormal evolution path is indicated by replacing the process abnormal event identified in the abnormal evolution path with the confirmed process abnormal event.

[0011] Furthermore, after step S3, the following steps are included: Step B1: Extract the time difference between the occurrence of two adjacent abnormal events and the difference in related parameters in the abnormal evolution path; Step B2: Determine whether the time difference between adjacent abnormal events is less than a preset time threshold and whether the difference in the correlation parameters exceeds a set causal correlation threshold. If yes, then determine that the adjacent abnormal events have a reasonable causal correlation; if no, then determine that the adjacent abnormal events are accidental and remove the latter abnormal event from the abnormal evolution path. Step B3: Perform the above causal verification on each of the adjacent abnormal events in the abnormal evolution path, and correct the abnormal evolution path to obtain a new abnormal evolution path.

[0012] Furthermore, step S4 includes: Step S41: Identify the timestamps corresponding to each abnormal event in the abnormal evolution path, and match the equipment operation monitoring parameters associated with each abnormal event from the continuously collected operation data. Step S42: Obtain the preset reference value corresponding to the equipment operation monitoring parameter, and calculate the offset of the equipment operation monitoring parameter relative to the preset reference value; Step S43: Use the equipment operation monitoring parameters associated with each abnormal event to match the score weight corresponding to each abnormal event from the expert experience knowledge base, and use the deviation of the equipment operation monitoring parameter values ​​associated with each abnormal event relative to the benchmark value to match the score corresponding to each abnormal event from the expert experience knowledge base. Use a weighted method to obtain the abnormal score of each abnormal event. The expert experience knowledge base is based on the original data associated with abnormal operation data obtained from the historical operation database of the coal mine power supply system. The original data is fused using a data fusion algorithm and combined with expert experience recognition. Step S44: Accumulate the abnormal scores corresponding to each abnormal event to obtain a quantitative indicator for locking decision.

[0013] Furthermore, step S5 includes: Step S51: Use quantitative indicators combined with preset lockout decision confidence thresholds to confirm whether there are operational risks in the coal mine power supply system; Step S52: When there is a risk, obtain the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, the offset of the equipment operation monitoring parameter values ​​relative to the baseline value, and the duration of the abnormal event. Step S53: Based on the equipment operation monitoring parameters, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of the abnormal event, the current operation risk level of the coal mine power supply system is obtained by matching from the preset risk level judgment rules.

[0014] Furthermore, in step S53, the preset risk level judgment rule specifically includes: The coal mine power supply system is pre-defined to have four risk levels: low risk, medium risk, high risk, and emergency risk. For each risk level, corresponding parameters are configured to include the number of abnormal equipment operation monitoring parameters, the threshold for the deviation of equipment operation monitoring parameter values ​​from the benchmark value, and the threshold for the duration of abnormal events. If only a single device has abnormal operating parameters, and its deviation from the baseline value does not exceed the preset low-amplitude deviation threshold and the duration of the corresponding abnormal event is less than the preset short-term threshold, it is judged as low risk. When multiple or a single device has abnormal operating monitoring parameters, and the deviation of its value relative to the benchmark value exceeds the preset low-amplitude deviation threshold but does not exceed the preset medium-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset short-term threshold and the preset medium-term threshold, it is judged as medium risk. When multiple devices have abnormal operating monitoring parameters, and their deviation from the benchmark value exceeds the preset medium-amplitude deviation threshold but does not exceed the preset high-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset medium-time threshold and the preset high-time threshold, it is judged as high risk; When the deviation of a device's operating monitoring parameter value from the baseline value exceeds a preset high deviation threshold, and the duration of the corresponding abnormal event reaches a preset high time threshold, it is determined to be an emergency risk.

[0015] Furthermore, in step S6, the locking command specifically includes: Power outages are implemented on electrical equipment corresponding to abnormal events located at the end of the time sequence in the abnormal evolution path.

[0016] This application provides an intelligent interlocking method for coal mine power supply systems based on active defense. By continuously collecting operational data of the coal mine power supply system and identifying abnormal events in conjunction with preset abnormal event judgment criteria, it solves the problems of passive protection and inability to provide early warning in existing technologies. Furthermore, this method traces abnormal events and forms an abnormal evolution path, enabling a comprehensive understanding of the fault occurrence process. Through quantitative analysis of abnormal events, quantitative indicators for interlocking decisions are obtained, and the operational risk level of the coal mine power supply system is determined according to preset risk level judgment rules, achieving an assessment of system operational risk and avoiding the limitations of traditional manual experience-based parameter setting. Finally, when the operational risk level meets the active interlocking trigger conditions, an interlocking command is issued to the isolation equipment of the coal mine power supply system, realizing proactive intervention in fault risk, thereby improving the safety of the coal mine power supply system and ensuring the safe operation and smooth production of the mine. Attached Figure Description

[0017] Figure 1 A flowchart illustrating an intelligent interlocking method for a coal mine power supply system based on active defense, provided in an embodiment of this application. Detailed Implementation

[0018] To better illustrate the present invention, the invention will now be described in further detail with reference to the accompanying drawings.

[0019] 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.

[0020] The following description uses at least one specific embodiment as an example. In this embodiment: In this embodiment, as Figure 1 As shown, an intelligent interlocking method for a coal mine power supply system based on active defense includes the following steps: Step S1: Continuously collect the operating data of the coal mine power supply system, wherein the operating data includes current data, voltage data, equipment temperature data and insulation resistance data; Step S2: Based on the collected operational data and combined with the preset abnormal event judgment criteria, identify whether there are abnormal events in the coal mine power supply system. If there are abnormal events, assign a timestamp to each identified abnormal event. Step S3: In response to any identified abnormal event, trace back all abnormal events before the abnormal event within the set time window, and combine and sort the abnormal events in chronological order to form an abnormal evolution path. Step S4: Identify each abnormal event in the abnormal evolution path, extract the equipment operation monitoring parameters associated with each abnormal event and the deviation of the equipment operation monitoring parameter values ​​from the baseline value, and perform quantitative analysis to obtain quantitative indicators for lockout decision-making. Step S5: Using quantitative indicators, the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of each abnormal event, determine the operation risk level of the coal mine power supply system according to the preset risk level judgment rules. Step S6: When the operational risk level meets the active interlocking trigger condition, an interlocking command is issued to the isolation equipment of the coal mine power supply system.

[0021] This embodiment proposes an intelligent interlocking method for coal mine power supply systems based on proactive defense. It continuously collects operational data from the coal mine power supply system and identifies abnormal events by combining this data with preset abnormal event judgment criteria. When an abnormal event occurs, it traces and combines these events to form an abnormal evolution path. Then, it identifies the abnormal events within the path and extracts the associated equipment operation monitoring parameters and their offset amplitudes, performing quantitative analysis to obtain quantitative indicators for interlocking decisions. Finally, based on the quantitative indicators, equipment operation monitoring parameters, offset amplitudes, and the duration of the abnormal event, it determines the operational risk level of the coal mine power supply system. When the proactive interlocking trigger conditions are met, it issues an interlocking command to the isolated equipment, aiming to achieve early warning and proactive intervention for anomalies in the coal mine power supply system.

[0022] The intelligent interlocking method for coal mine power supply systems based on active defense proposed in this embodiment can be implemented as follows: In step S1, it is necessary to continuously collect operational data from the coal mine power supply system. This operational data is fundamental for assessing the system's health. For example, independent sensors, such as current sensors, voltage sensors, temperature sensors, and insulation resistance sensors, are deployed at various key nodes of the coal mine power supply system. These sensors can operate independently, transmitting the collected data to a local data storage unit via wired or wireless means.

[0023] In step S2, based on the collected operational data and a preset abnormal event judgment benchmark, it is determined whether there are any abnormal events in the coal mine power supply system. The collected operational data is compared with preset fixed thresholds. For example, upper and lower limits can be set for current, voltage, equipment temperature, and insulation resistance. When any operational data exceeds these preset ranges, it is determined that an abnormal event exists. If an abnormal event is identified, a timestamp needs to be assigned to each identified abnormal event, for example, recording the specific date and time of the abnormal event for subsequent time-series analysis.

[0024] In step S3, in response to any identified abnormal event, all abnormal events preceding that event within a set time window are traced back, and these events are combined and sorted chronologically to form an abnormal evolution path. For example, when an abnormal temperature increase of a device is detected, a time window can be set, such as the past hour or 24 hours, to trace whether any other abnormal events have occurred within this time window for the device or its associated devices. Then, these traced abnormal events are arranged chronologically to form an abnormal evolution path. For example, the path might be "decreasing insulation resistance -> current fluctuation -> increased device temperature". This helps to understand the causal relationships and development trends between abnormal events.

[0025] In step S4, each abnormal event located in the abnormal evolution path is identified, and the equipment operation monitoring parameters associated with each abnormal event, as well as the deviation of the equipment operation monitoring parameter values ​​relative to the baseline value, are extracted and quantitatively analyzed to obtain a quantitative index for lockout decision-making. For example, for each abnormal event in the abnormal evolution path, its corresponding equipment operation monitoring parameter can be identified; for example, the monitoring parameter corresponding to "equipment temperature rise" is "equipment temperature," and the monitoring parameter corresponding to "current fluctuation" is "current." Then, the specific values ​​of these monitoring parameters when the abnormality occurs are obtained, and their deviation relative to the normal operating baseline value is calculated. For example, if the normal equipment temperature baseline value is 60℃, and it reaches 80℃ during the abnormality, the deviation is 20℃. Next, a weight or score can be assigned to each abnormal event and its deviation. For example, the weight of temperature abnormality may be higher than that of current abnormality, or the larger the deviation, the higher the score. Finally, these scores are accumulated or weighted averaged to obtain a comprehensive quantitative index that reflects the severity of the entire abnormal evolution path.

[0026] In step S5, the operational risk level of the coal mine power supply system is determined using quantitative indicators, equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of each abnormal event, based on a preset risk level judgment rule. For example, a set of risk level judgment rules can be preset, which may include multiple conditions. For instance, when the quantitative indicator exceeds a certain threshold, and the path contains significant deviations in multiple key equipment parameters, and the duration of the abnormal event is relatively long, it is judged as "high risk"; when the quantitative indicator is low, and only a single equipment parameter has a slight deviation, and the duration is short, it is judged as "low risk".

[0027] In step S6, when the operational risk level meets the active blocking triggering conditions, a blocking command is issued to the isolation equipment of the coal mine power supply system. When the system determines that the current operational risk level meets these conditions, the system will automatically generate and send a blocking command to the corresponding isolation equipment, such as a circuit breaker or disconnector.

[0028] The intelligent interlocking method for coal mine power supply systems based on proactive defense proposed in this embodiment works by identifying abnormal events in system operation data through a recognition mechanism combined with preset judgment criteria. The evolution path of these abnormal events is then traced, connecting the abnormal points into a logically and temporally ordered event chain. By quantitatively analyzing key parameters in the abnormal evolution path, the severity of the current risk can be objectively assessed. Finally, based on the quantitative assessment results, an interlocking command is proactively triggered to intervene in potential dangers at an early stage. This shift from passive response to proactive prevention of risks enables coal mine power supply systems to effectively cope with the complex and ever-changing underground environment.

[0029] In some of the above implementation methods, continuous collection of operational data from coal mine power supply systems is proposed. However, in practical applications, if the collection process is not managed in a refined manner, the data source may be unclear, making it difficult to effectively trace the equipment and location where abnormal events occur, thereby affecting the accuracy of subsequent abnormal event identification and risk assessment.

[0030] Furthermore, step S1 includes: Step S11a: Use multiple acquisition units deployed in the coal mine power supply system to collect the operation data of the coal mine power supply system; wherein, each acquisition unit is pre-set with the device number, circuit name and physical location it monitors when it is deployed, and after each acquisition unit completes the collection of operation data, it is stored in combination with the device number, circuit name and physical location corresponding to each acquisition unit.

[0031] Specifically, the multiple acquisition units refer to sensors or data acquisition modules deployed at different key nodes or on equipment in the coal mine power supply system. These acquisition units are configured to monitor and acquire operational data of their assigned areas or equipment in real time, such as current data, voltage data, equipment temperature data, and insulation resistance data. During deployment, each acquisition unit is pre-set with the device number, circuit name, and physical location it monitors. The device number uniquely identifies the monitored equipment, the circuit name specifies the electrical circuit to which the equipment belongs, and the physical location provides the equipment's specific spatial coordinates within the mine. After the acquisition unit completes data acquisition, this data is combined with the corresponding device number, circuit name, and physical location for storage, ensuring that each piece of operational data has a clear source and contextual information.

[0032] By implementing refined management of operational data during the data acquisition phase, the tracing difficulties that may exist in traditional data acquisition are resolved. By deploying multiple acquisition units and pre-setting the device number, loop name, and physical location for each unit, each piece of operational data can be clearly associated with a specific device, loop, and location. Therefore, when an abnormal event is identified in the subsequent step S2, the associated device and location can be quickly and accurately traced, providing a solid data foundation for subsequent anomaly evolution path analysis, quantitative indicator calculation, and risk level determination.

[0033] In some of the above-described implementations, operational data of the coal mine power supply system is continuously collected. However, in actual coal mine environments, the directly collected raw operational data, especially electrical sensor data, is often affected by mechanical and environmental factors. These non-electrical interferences can distort the operational data, thereby affecting the accuracy of subsequent anomaly identification.

[0034] Furthermore, step S1 includes: Step S11b: Obtain the raw operating data from the coal mine power supply system; Step S12b: Obtain the mechanical sensor data and environmental sensor data corresponding to the original operating data. The mechanical sensor data includes equipment vibration data and cable wear data, and the environmental sensor data includes downhole humidity data and rock conductivity data. Step S13b: Using the historical operating data of the coal mine power supply system and the quantitative correlation formula between the original operating data and mechanical and environmental factors established by the regression analysis algorithm, calculate and obtain the interference contribution of mechanical factors to the original operating data and the interference contribution of environmental factors to the original operating data. Among them, mechanical factors correspond to the mechanical state of the equipment reflected by mechanical sensor data, and environmental factors correspond to the underground environmental state reflected by environmental sensor data. Step S14b: Combining the interference contribution of mechanical factors to the original operating data, the interference contribution of environmental factors to the original operating data, and the collected mechanical sensor data and environmental sensor data, calculate the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data. Extract the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data from the original operating data to obtain the operating data of the coal mine power supply system used to characterize the electrical sensor data.

[0035] Specifically, in step S11b, the raw operating data refers to the unprocessed initial data directly obtained from various sensors in the coal mine power supply system, which includes various information such as electrical parameters, mechanical parameters, and environmental parameters. This data forms the basis for subsequent analysis.

[0036] In step S12b, mechanical sensor data can be understood as data reflecting the mechanical state of the equipment, such as equipment vibration data and cable wear data, which indicate the physical operating condition of the equipment. Environmental sensor data is used to reflect the underground environmental conditions, such as underground humidity data and rock conductivity data, which reflect the impact of the underground coal mine environment on equipment operation. Acquiring this data is to quantify the interference of mechanical and environmental factors on the original operating data.

[0037] In step S13b, the quantitative correlation formula is established through regression analysis of historical operating data of the coal mine power supply system. The regression analysis algorithm reveals the statistical relationship between the original operating data and mechanical and environmental factors, thereby establishing a mathematical model that can quantify the interference contribution of these factors. Mechanical factors correspond to the mechanical state of the equipment reflected by mechanical sensor data, and environmental factors correspond to the underground environmental state reflected by environmental sensor data. Using this formula, the interference contribution of each mechanical and environmental factor to the original operating data can be calculated.

[0038] In step S14b, the actual interference amounts of the mechanical sensor data and the environmental sensor data are calculated based on the interference contribution obtained in step S13b and the currently collected mechanical and environmental sensor data. Stripping away these actual interference amounts aims to remove fluctuations and deviations caused by non-electrical factors from the raw operating data, thereby obtaining operating data for the coal mine power supply system that characterizes the electrical sensor data. This operating data accurately reflects the electrical operating status of the coal mine power supply system itself.

[0039] By introducing the quantification and stripping of interference from mechanical and environmental factors, this method solves the problem of inaccurate identification caused by non-electrical interference in operational data in traditional methods. First, by acquiring raw operational data along with corresponding mechanical and environmental sensor data, a comprehensive data foundation is provided for subsequent interference analysis. Second, by using historical operational data to establish quantitative correlation formulas between raw operational data and mechanical and environmental factors, the impact of these interference factors on the raw operational data can be accurately calculated. Finally, by calculating and stripping the actual interference from mechanical and environmental sensor data, the final operational data ensures that it more accurately characterizes the electrical state of the coal mine power supply system, thus providing high-quality input for subsequent anomaly identification.

[0040] In this embodiment, step S2 includes: Step S21: Compare the collected running data with the preset normal operating data range, and fit the trend of the running data in multiple consecutive collection periods to determine whether there is any collected running data that exceeds the preset normal operating data range, and whether the rate of change of the collected running data meets the preset data mutation rate. If it exists, determine that there is a corresponding abnormal event, and assign a timestamp to the identified abnormal event; The preset abnormal event judgment criteria include a preset normal operation data range and a preset data mutation rate.

[0041] Specifically, the preset normal operating data range refers to the reasonable range of values ​​that various operating data (such as current data, voltage data, equipment temperature data, and insulation resistance data) should fall within under normal operating conditions of the coal mine power supply system. This range can be determined through historical operating data statistical analysis, expert experience setting, or system design specifications. When the collected operating data exceeds this range, it is considered that there may be an anomaly. The preset data mutation rate refers to the rate or magnitude of change of operating data values ​​within a continuous collection period reaching a preset critical value. For example, a large fluctuation in current or voltage within a short period of time, or a sharp increase in equipment temperature within a very short period of time. This rate of change is set to capture abnormal situations that, although they may still be within the normal value range, indicate potential faults due to their changing trends.

[0042] In practical applications, numerical comparison refers to comparing real-time collected operational data with the upper and lower limits of a preset normal operating data range to determine whether the data has exceeded the limits. Trend fitting refers to analyzing operational data over multiple consecutive collection periods using mathematical models (such as linear regression, curve fitting, etc.) to predict its future trend or identify its current rate of change. When the collected operational data not only exceeds the preset normal operating data range but also its rate of change meets the preset data mutation rate, the system will determine that a corresponding abnormal event exists. Once an abnormal event is identified, a timestamp is assigned to the event to record its exact time of occurrence for subsequent traceability and analysis.

[0043] By comparing the collected operational data with preset normal operating data ranges, it is possible to directly determine whether the current operating parameters deviate from the normal range. Simultaneously, by fitting the changing trends of operational data over multiple consecutive collection cycles, the rate of change of operational data can be dynamically monitored, thereby promptly detecting potential and rapidly developing anomalies, such as short circuits, overloads, or insulation degradation. This method, combining numerical boundary judgment with changing trend analysis, allows for a comprehensive consideration of both the absolute value and dynamic characteristics of the data in addressing anomalies. Therefore, it is possible to more comprehensively capture various anomalies that may occur in coal mine power supply systems, providing accurate initial data for subsequent anomaly evolution path analysis and risk level determination.

[0044] In some embodiments described above, this application proposes a method to respond to any identified abnormal event by tracing back all abnormal events prior to that event within a set time window, and then combining and sorting these abnormal events in chronological order to form an abnormal evolution path. However, in practical applications, an abnormal evolution path formed solely based on chronological order may not fully and accurately reflect the true evolution logic and causal relationships between abnormal events, and may include some accidental or non-critical abnormal events, thereby affecting the accuracy of subsequent risk assessments and the reliability of lockout decisions.

[0045] Therefore, the scheme for correcting the abnormal evolution path after step S3 specifically includes: Step A1: Identify and distinguish the causal abnormal event, the result abnormal event, and multiple process abnormal events from the abnormal evolution path. The result abnormal event is the abnormal event identified in response to the abnormal evolution path. The causal abnormal event is the abnormal event that initially occurs in the time sequence of the abnormal evolution path. The process abnormal event is the abnormal event that occurs between the result abnormal event and the causal abnormal event in the abnormal evolution path. Step A2: Identify the content of the causal abnormal event and the result abnormal event, and preliminarily confirm the process abnormal event that occurs when the causal abnormal event evolves into the result abnormal event based on the prior database; Step A3: Perform consistency matching between the confirmed process anomaly events and the process anomaly events identified in the anomaly evolution path; Step A4: When the matching results are consistent or the process abnormal event identified in the abnormal evolution path contains the confirmed process abnormal event, the abnormal evolution path is corrected. The correction of the abnormal evolution path is indicated by replacing the process abnormal event identified in the abnormal evolution path with the confirmed process abnormal event.

[0046] Specifically, in step A1, the initially formed anomaly evolution path is first subjected to a structured analysis, classifying the anomaly events into three categories: causal anomalies, outcome anomalies, and process anomalies. Causal anomalies are the earliest anomalies occurring in the anomaly evolution path, typically the starting point of the entire anomaly chain. Outcome anomalies are those identified in response to the anomaly evolution path, usually the currently relevant or most recently occurring anomalies. Process anomalies refer to all anomalies occurring between the causal and outcome anomalies, constituting the key links in the anomaly's development. This classification facilitates targeted verification and correction of the anomaly evolution path in the subsequent process.

[0047] In step A2, the system identifies the specific content of the causal and consequential abnormal events and, in conjunction with a pre-defined prior database, preliminarily confirms which process abnormal events typically occur as the causal abnormal event evolves into the consequential abnormal event. The prior database can be understood as containing information such as historical failure modes, expert experience and knowledge, and equipment fault trees, used to store and describe known evolutionary patterns and causal relationships between different abnormal events. By querying the prior database, an abnormal evolution path based on experience and knowledge can be obtained, serving as a benchmark for subsequent verification.

[0048] Based on this, in step A3, the process anomalies confirmed through the prior database are matched for consistency with the process anomalies identified from the actual anomaly evolution paths in step A1. This matching process aims to compare the similarities and differences between the actually observed anomaly evolution process and the ideal evolution process inferred based on historical experience and expert knowledge.

[0049] Finally, in step A4, the anomaly evolution path is corrected based on the matching results. When the matching results are consistent—that is, when the actually observed process anomaly events completely match the process anomaly events confirmed by the prior database—the current anomaly evolution path is considered reliable. Even when the process anomaly events identified in the anomaly evolution path include those confirmed by the prior database, the path is still considered to have some rationality. However, to improve the accuracy and simplicity of the path, the confirmed process anomaly events will replace the process anomaly events identified in the anomaly evolution path. This replacement operation can remove potentially redundant, irrelevant, or accidental anomaly events from the path, thus making the anomaly evolution path more accurately reflect the actual fault evolution logic.

[0050] In some preferred embodiments, it is assumed that the coal mine power supply system experiences the following sequence of abnormal events over a certain period of time: first, "overload of a feeder current" (cause abnormal event), followed by "local temperature rise of the transformer," "slight decrease in insulation resistance," "abnormal fan vibration," and "bus voltage fluctuation" (process abnormal events), ultimately leading to "relay protection tripping" (result abnormal event). If only based on chronological order, the abnormal evolution path may include all these events. However, using the modified method of this application, in step A1, "feeder current overload" is first identified as the cause abnormal event, "relay protection tripping" as the result abnormal event, and the rest as process abnormal events. In step A2, the system queries a priori database and finds that the typical evolution path from "feeder current overload" to "relay protection tripping" is usually "feeder current overload" leading to "local temperature rise of the transformer," which in turn causes "decrease in insulation resistance," ultimately triggering "relay protection tripping." In step A3, the actually identified process anomalies ("local transformer temperature rise", "slight decrease in insulation resistance", "abnormal fan vibration", "bus voltage fluctuation") are matched with the process anomalies confirmed in the prior database ("local transformer temperature rise", "decline in insulation resistance"). In step A4, since the actual path includes key events in the confirmed path, and the direct causal link between "abnormal fan vibration" and "bus voltage fluctuation" and the current fault may be weak, the system will correct the anomaly evolution path and replace it with the path confirmed in the prior database, i.e., "feeder current overload" → "local transformer temperature rise" → "decline in insulation resistance" → "relay protection trip". This correction eliminates non-critical or coincidental events, making the anomaly evolution path clearer and more accurate, providing a more reliable basis for subsequent risk assessment.

[0051] In some of the above implementations, in response to an identified anomalous event, all anomalous events preceding that event within a set time window are traced back, and these events are combined and sorted chronologically to form an anomalous evolution path. However, this simple chronological combination method may result in the anomalous evolution path containing some coincidentally related anomalous events rather than events with a true causal relationship, thereby affecting the accuracy of subsequent risk assessments and the reliability of lockout decisions.

[0052] Therefore, after step S3, the following steps are included: Step B1: Extract the time difference between the occurrence of two adjacent abnormal events and the difference in related parameters in the abnormal evolution path; Step B2: Determine whether the time difference between adjacent abnormal events is less than a preset time threshold and whether the difference in the correlation parameters exceeds a set causal correlation threshold. If yes, then determine that the adjacent abnormal events have a reasonable causal correlation; if no, then determine that the adjacent abnormal events are accidental and remove the latter abnormal event from the abnormal evolution path. Step B3: Perform the above causal verification on each of the adjacent abnormal events in the abnormal evolution path, and correct the abnormal evolution path to obtain a new abnormal evolution path.

[0053] Specifically, in step B1, extracting the time difference between two adjacent abnormal events in the abnormal evolution path refers to calculating the time interval between the occurrence of the first abnormal event and the occurrence of the second abnormal event. Simultaneously, extracting the difference in associated parameters involves obtaining the equipment operation monitoring parameters associated with each of these two adjacent abnormal events and calculating the numerical differences or changes in these parameters before and after the two events. For example, if the first abnormal event is "current overload" and the second abnormal event is "equipment temperature rise," then the difference in associated parameters could refer to the change in current or temperature during the two events.

[0054] In step B2, it is determined whether the time difference between adjacent anomalous events is less than a preset time threshold, aiming to assess the temporal closeness of the two events. If the time interval is too long, the probability of a causal relationship is low. It is also determined whether the difference in the correlation parameters exceeds a set causal relationship threshold, aiming to assess whether the parameter changes are significant enough to support a causal relationship. For example, if one event causes the other, it will usually show significant changes in the relevant parameters that exceed random fluctuations. If both the time difference is less than the preset time threshold and the difference in the correlation parameters exceeds the set causal relationship threshold, then a reasonable causal relationship is considered to exist between the two adjacent anomalous events. Conversely, if neither condition is met, it is determined to be a random correlation, and the latter anomalous event in the anomalous evolution path is removed from the path to avoid interference from non-causal events in the path analysis.

[0055] In practical applications, step B3 involves performing the aforementioned causal verification on each adjacent abnormal event in the abnormal evolution path, ensuring the logical consistency of the entire abnormal evolution path. Through verification and correction, a new abnormal evolution path that more accurately reflects the actual fault evolution process can be obtained.

[0056] In some preferred embodiments, it is assumed that the coal mine power supply system experienced the following sequence of abnormal events over a certain period of time: Event 1: A sudden drop in voltage in a certain circuit (time T1, associated parameter: voltage); Event 2: Temperature of adjacent device A increases (Time T2, associated parameter: temperature); Event 3: Abnormal current in another circuit device B (Time T3, associated parameter: current).

[0057] Based on step S3 above, an abnormal evolution path may form: "sudden voltage drop -> temperature rise of device A -> abnormal current of device B".

[0058] At this point, the scheme in this application will further perform causal verification: First, in step B1, the time difference (T2-T1) between the "voltage drop" and the "temperature rise of device A" is extracted, along with the differences in the associated parameters (voltage change and temperature change). Next, in step B2, it is determined whether T2-T1 is less than a preset time threshold, and whether the voltage change and temperature change exceed set causal correlation thresholds. If the determination results in these values, a reasonable causal correlation is established between the two events.

[0059] Subsequently, the time difference (T3-T2) between "temperature rise in device A" and "current anomaly in device B" and the difference in related parameters (temperature change and current change) are extracted. If, in step B2, it is determined that T3-T2 is too long, or there is no significant correlation between the temperature change and the current change (for example, the temperature rise in device A does not directly cause the current anomaly in device B; the two may just occur at similar times but have no direct causal relationship), then "temperature rise in device A" and "current anomaly in device B" are determined to be an accidental correlation, and "current anomaly in device B" is removed from the abnormal evolution path.

[0060] Finally, in step B3, after the above causal verification and correction, the abnormal evolution path may be corrected to "voltage drop -> device A temperature rise", thus obtaining a new abnormal evolution path that is more accurate and more causally logical, providing a more reliable basis for subsequent risk assessment and lockout decisions.

[0061] In some of the above implementations, after identifying each anomalous event in the abnormal evolution path and extracting the associated parameters, quantitative analysis is required to obtain quantitative indicators for locking decisions. However, if the quantitative analysis process lacks refinement and professional consideration, the accuracy of the quantitative indicators may be insufficient, thereby affecting the reliability of subsequent risk level determination and potentially delaying the timing of proactive locking.

[0062] In this regard, step S4 includes: S41. Based on the timestamps corresponding to the abnormal events, match the equipment operation monitoring parameters associated with each abnormal event from the continuously collected operation data; S42. Obtain the preset reference value corresponding to the equipment operation monitoring parameter, and calculate the offset of the equipment operation monitoring parameter relative to the preset reference value; S43. The score weights corresponding to each abnormal event are obtained by matching the equipment operation monitoring parameters associated with each abnormal event from the expert experience knowledge base, and the scores corresponding to each abnormal event are obtained by matching the deviation of the equipment operation monitoring parameter values ​​associated with each abnormal event from the expert experience knowledge base. The abnormal scores of each abnormal event are obtained by weighting. The expert experience knowledge base is based on the original data associated with abnormal operation data obtained from the historical operation database of the coal mine power supply system. The original data is fused by using a data fusion algorithm and combining expert experience identification. S44. The abnormal scores corresponding to each abnormal event are summed to obtain a quantitative indicator for locking decision.

[0063] Specifically, in step S41, it is first necessary to identify the timestamp of each abnormal event in the abnormal evolution path. These timestamps are the precise moments when the abnormal events were detected in the system. Subsequently, based on these timestamps, the equipment operation monitoring parameters directly associated with each abnormal event are matched and extracted from the continuously collected coal mine power supply system operation data. These parameters can be current data, voltage data, equipment temperature data, and insulation resistance data, etc., which directly reflect the specific operating status of the equipment when the abnormal event occurred.

[0064] In step S42, for the matched equipment operation monitoring parameters, their corresponding preset benchmark values ​​need to be obtained. The preset benchmark values ​​are typically the parameter range or typical values ​​of the equipment under normal and stable operating conditions. By comparing the current monitoring parameters with the benchmark values, the deviation of the equipment operation monitoring parameters relative to the preset benchmark values ​​can be calculated. This deviation directly reflects the degree to which the equipment's operating state deviates from the normal level and is an important basis for assessing the severity of the anomaly.

[0065] In practical applications, in step S43, to more accurately quantify the impact of abnormal events, this application introduces an expert experience knowledge base. This knowledge base is constructed based on raw data associated with operational anomalies obtained from the historical operation database of the coal mine power supply system, and then identified and fused using a data fusion algorithm combined with expert experience. Specifically, the score weight corresponding to each abnormal event is obtained from the expert experience knowledge base by matching the equipment operation monitoring parameters associated with each abnormal event. Simultaneously, the corresponding score is obtained from the expert experience knowledge base based on the deviation of the equipment operation monitoring parameter values ​​from the benchmark value. By weighting the score weights and scores, an abnormality score for each abnormal event can be obtained. This weighting method comprehensively considers the type, severity, and expert experience of the abnormal event, making the abnormality scores more representative and accurate.

[0066] Furthermore, in step S44, the abnormal scores of all abnormal events in the abnormal evolution path are accumulated to obtain a quantitative indicator for the locking decision. This indicator reflects the severity and potential risks of the entire abnormal evolution path and provides a key quantitative basis for subsequent risk level determination.

[0067] In some preferred embodiments, it is assumed that a partial discharge anomaly occurs on a feeder in the coal mine power supply system.

[0068] First, in step S41, the system identifies the timestamp of the partial discharge abnormal event and matches the equipment operation monitoring parameters associated with the event from the continuously collected operating data, such as the partial discharge amount of the feeder, the insulation resistance value, and the temperature of the related equipment.

[0069] Next, in step S42, the system obtains the preset reference values ​​of these parameters and calculates the deviation of the partial discharge amount from the normal value (for example, from 0pC to 500pC, the deviation is 500pC), the decrease of the insulation resistance value from the normal value, and the increase of the temperature from the normal value.

[0070] Subsequently, in step S43, the system uses the monitoring parameters of these devices (partial discharge quantity, insulation resistance, and temperature) to match the corresponding score weights from a pre-built expert experience knowledge base. For example, partial discharge quantity may be assigned a higher weight because it directly indicates insulation degradation; a decrease in insulation resistance also has a higher weight; and a temperature increase may have a slightly lower weight because it may be the result of multiple factors. Simultaneously, based on the offset magnitude of these parameters, a corresponding score is matched from the knowledge base. For example, a partial discharge quantity of 500 pC may correspond to a higher score, while a slight temperature increase corresponds to a lower score. Through weighted calculation, the anomalous score of the partial discharge anomaly event is obtained.

[0071] Finally, in step S44, if the partial discharge event is the only event in the abnormal evolution path, its anomalous score is the quantitative indicator used for the latch-up decision; if the path contains other abnormal events, the anomalous scores of all abnormal events are summed to obtain the final quantitative indicator. In this way, the system can perform a refined and professional quantitative assessment of partial discharge anomalies, providing an accurate basis for subsequent risk assessment and latch-up decisions.

[0072] Furthermore, in the above-mentioned intelligent interlocking method for coal mine power supply systems based on active defense, step S5 includes: S51. Use quantitative indicators combined with preset lockout decision confidence thresholds to confirm whether there are operational risks in the coal mine power supply system; S52. When there is a risk, obtain the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, as well as the offset of the equipment operation monitoring parameter values ​​relative to the baseline value and the duration of the abnormal event. S53. Based on the equipment operation monitoring parameters, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of the abnormal event, the current operation risk level of the coal mine power supply system is obtained by matching from the preset risk level judgment rules.

[0073] Step S51 aims to quickly determine whether there are potential operational risks in the coal mine power supply system through a preliminary assessment of quantitative indicators. Quantitative indicators can be understood as a comprehensive numerical representation of the overall abnormal state of the system, with their magnitude reflecting the severity of the anomaly. The lockout decision confidence threshold is a pre-set key parameter used to define at what level the quantitative indicators should be considered as indicating operational risk in the system. When the calculated quantitative indicators exceed this pre-set threshold, the system is preliminarily confirmed to have operational risk, thus triggering the subsequent detailed risk assessment process.

[0074] Furthermore, in step S52, once an operational risk is confirmed, more detailed risk assessment data will be collected. This includes extracting equipment operation monitoring parameters directly related to each abnormal event from the abnormal evolution path, such as current data, voltage data, equipment temperature data, and insulation resistance data. Simultaneously, the deviation of these equipment operation monitoring parameters' real-time values ​​from their preset reference values ​​will be obtained; this deviation reflects the degree to which the parameters deviate from their normal state. In addition, the duration of the abnormal event is also an important consideration, indicating the length of time the abnormal state persists.

[0075] Specifically, based on step S52, step S53 determines the risk level. Equipment operation monitoring parameters, their deviations from baseline values, and the duration of abnormal events are taken as input and matched according to preset risk level judgment rules. These preset risk level judgment rules are established based on expert experience and historical fault data analysis, and are used to map different parameter combinations, deviation ranges, and durations to specific risk levels, such as low risk, medium risk, high risk, or emergency risk. Through this matching mechanism, the current operational risk level of the coal mine power supply system can be identified and output.

[0076] Specifically, in step S53, the preset risk level judgment rule includes: The coal mine power supply system is pre-defined to have four risk levels: low risk, medium risk, high risk, and emergency risk. For each risk level, corresponding parameters are configured to include the number of abnormal equipment operation monitoring parameters, the threshold for the deviation of equipment operation monitoring parameter values ​​from the benchmark value, and the threshold for the duration of abnormal events. If only a single device has abnormal operating parameters, and its deviation from the baseline value does not exceed the preset low-amplitude deviation threshold and the duration of the corresponding abnormal event is less than the preset short-term threshold, it is judged as low risk. When multiple or a single device has abnormal operating monitoring parameters, and the deviation of its value relative to the benchmark value exceeds the preset low-amplitude deviation threshold but does not exceed the preset medium-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset short-term threshold and the preset medium-term threshold, it is judged as medium risk. When multiple devices have abnormal operating monitoring parameters, and their deviation from the benchmark value exceeds the preset medium-amplitude deviation threshold but does not exceed the preset high-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset medium-time threshold and the preset high-time threshold, it is judged as high risk; When the deviation of a device's operating monitoring parameter value from the baseline value exceeds a preset high deviation threshold, and the duration of the corresponding abnormal event reaches a preset high time threshold, it is determined to be an emergency risk.

[0077] In some of the above embodiments, when the operational risk level of the coal mine power supply system meets the active interlocking triggering conditions, an interlocking command needs to be issued to the isolation equipment of the coal mine power supply system. This application further clarifies the specific content of the interlocking command.

[0078] Specifically, in step S6, the locking command includes: Power outages are implemented on electrical equipment corresponding to abnormal events located at the end of the time sequence in the abnormal evolution path.

[0079] The interlocking command refers to the control signal sent by the system to the isolation equipment in the coal mine power supply system after determining that the operational risk level has reached the active interlocking trigger condition. The core of this command is to specifically operate on the electrical equipment that causes or is associated with the highest risk at present.

[0080] Specifically, the system identifies the most recent, end-of-time anomaly within the anomaly evolution path formed in step S3. This anomaly is typically the direct cause or final manifestation of an increased system risk level. Subsequently, the system determines the electrical equipment directly associated with this end-of-time anomaly. Once the target electrical equipment is identified, a blocking command is issued to the isolation device controlling that equipment to perform a power-off operation and lock it, thereby cutting off the power supply to the equipment and preventing further escalation of the fault or the occurrence of a more serious accident.

[0081] This embodiment provides an intelligent interlocking method for coal mine power supply systems based on active defense. By continuously collecting operational data of the coal mine power supply system and identifying abnormal events in conjunction with preset abnormal event judgment criteria, it solves the problems of passive protection and inability to provide early warnings in existing technologies. Furthermore, this method traces abnormal events and forms abnormal evolution paths, enabling a comprehensive understanding of the fault occurrence process. Through quantitative analysis of abnormal events, quantitative indicators for interlocking decisions are obtained, and the operational risk level of the coal mine power supply system is determined according to preset risk level judgment rules, achieving an assessment of system operational risk and avoiding the limitations of traditional manual experience-based parameter setting. Finally, when the operational risk level meets the active interlocking trigger conditions, an interlocking command is issued to the isolation equipment of the coal mine power supply system, realizing proactive intervention in fault risks to improve the safety of the coal mine power supply system and ensure the safe operation and smooth production of the mine.

[0082] 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 smart interlocking method for a coal mine power supply system based on active defense, characterized in that, Includes the following steps: Step S1: Continuously collect the operating data of the coal mine power supply system, wherein the operating data includes current data, voltage data, equipment temperature data and insulation resistance data; Step S2: Based on the collected operational data and combined with the preset abnormal event judgment criteria, identify whether there are abnormal events in the coal mine power supply system. If there are abnormal events, assign a timestamp to each identified abnormal event. Step S3: In response to any identified abnormal event, trace back all abnormal events before the abnormal event within the set time window, and combine and sort the abnormal events in chronological order to form an abnormal evolution path. Step S4: Identify each abnormal event in the abnormal evolution path, extract the equipment operation monitoring parameters associated with each abnormal event and the deviation of the equipment operation monitoring parameter values ​​from the baseline value, and perform quantitative analysis to obtain quantitative indicators for lockout decision-making. Step S5: Using quantitative indicators, the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of each abnormal event, determine the operation risk level of the coal mine power supply system according to the preset risk level judgment rules. Step S6: When the operational risk level meets the active interlocking triggering conditions, an interlocking command is issued to the isolation equipment of the coal mine power supply system.

2. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, Step S1 includes: Step S11a: Use multiple acquisition units deployed in the coal mine power supply system to collect the operation data of the coal mine power supply system; wherein, each acquisition unit is pre-set with the device number, circuit name and physical location it monitors when it is deployed, and after each acquisition unit completes the collection of operation data, it is stored in combination with the device number, circuit name and physical location corresponding to each acquisition unit.

3. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, Step S1 includes: Step S11b: Obtain the raw operating data from the coal mine power supply system; Step S12b: Obtain the mechanical sensor data and environmental sensor data corresponding to the original operating data. The mechanical sensor data includes equipment vibration data and cable wear data, and the environmental sensor data includes downhole humidity data and rock conductivity data. Step S13b: Using the historical operating data of the coal mine power supply system and the quantitative correlation formula between the original operating data and mechanical and environmental factors established by the regression analysis algorithm, calculate and obtain the interference contribution of mechanical factors to the original operating data and the interference contribution of environmental factors to the original operating data. Among them, mechanical factors correspond to the mechanical state of the equipment reflected by mechanical sensor data, and environmental factors correspond to the underground environmental state reflected by environmental sensor data. Step S14b: Combining the interference contribution of mechanical factors to the original operating data, the interference contribution of environmental factors to the original operating data, and the collected mechanical sensor data and environmental sensor data, calculate the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data. Extract the actual interference amount of the mechanical sensor data and the actual interference amount of the environmental sensor data from the original operating data to obtain the operating data of the coal mine power supply system used to characterize the electrical sensor data.

4. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, Step S2 includes: Step S21: Compare the collected running data with the preset normal operating data range, and fit the trend of the running data in multiple consecutive collection periods to determine whether there is any collected running data that exceeds the preset normal operating data range, and whether the rate of change of the collected running data meets the preset data mutation rate. If it exists, determine that there is a corresponding abnormal event, and assign a timestamp to the identified abnormal event; The preset abnormal event judgment criteria include a preset normal operation data range and a preset data mutation rate.

5. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, After step S3, the following is included: Step A1: Identify and distinguish the causal abnormal event, the result abnormal event, and multiple process abnormal events from the abnormal evolution path. The result abnormal event is the abnormal event identified in response to the abnormal evolution path. The causal abnormal event is the abnormal event that initially occurs in the time sequence of the abnormal evolution path. The process abnormal event is the abnormal event that occurs between the result abnormal event and the causal abnormal event in the abnormal evolution path. Step A2: Identify the content of the causal abnormal event and the result abnormal event, and preliminarily confirm the process abnormal event that occurs when the causal abnormal event evolves into the result abnormal event based on the prior database; Step A3: Perform consistency matching between the confirmed process anomaly events and the process anomaly events identified in the anomaly evolution path; Step A4: When the matching results are consistent or the process abnormal event identified in the abnormal evolution path contains the confirmed process abnormal event, the abnormal evolution path is corrected. The correction of the abnormal evolution path is indicated by replacing the process abnormal event identified in the abnormal evolution path with the confirmed process abnormal event.

6. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, After step S3, the following is included: Step B1: Extract the time difference between the occurrence of two adjacent abnormal events and the difference in related parameters in the abnormal evolution path; Step B2: Determine whether the time difference between adjacent abnormal events is less than a preset time threshold and whether the difference in the correlation parameters exceeds a set causal correlation threshold. If yes, then determine that the adjacent abnormal events have a reasonable causal correlation; if no, then determine that the adjacent abnormal events are accidental and remove the latter abnormal event from the abnormal evolution path. Step B3: Perform the above causal verification on each of the adjacent abnormal events in the abnormal evolution path, and correct the abnormal evolution path to obtain a new abnormal evolution path.

7. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, Step S4 includes: Step S41: Identify the timestamps corresponding to each abnormal event in the abnormal evolution path, and match the equipment operation monitoring parameters associated with each abnormal event from the continuously collected operation data. Step S42: Obtain the preset reference value corresponding to the equipment operation monitoring parameter, and calculate the offset of the equipment operation monitoring parameter relative to the preset reference value; Step S43: Use the equipment operation monitoring parameters associated with each abnormal event to match the score weight corresponding to each abnormal event from the expert experience knowledge base, and use the deviation of the equipment operation monitoring parameter values ​​associated with each abnormal event relative to the benchmark value to match the score corresponding to each abnormal event from the expert experience knowledge base. Use a weighted method to obtain the abnormal score of each abnormal event. The expert experience knowledge base is based on the original data associated with abnormal operation data obtained from the historical operation database of the coal mine power supply system. The original data is fused using a data fusion algorithm and combined with expert experience recognition. Step S44: Accumulate the abnormal scores corresponding to each abnormal event to obtain a quantitative indicator for locking decision.

8. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, Step S5 includes: Step S51: Use quantitative indicators combined with preset lockout decision confidence thresholds to confirm whether there are operational risks in the coal mine power supply system; Step S52: When there is a risk, obtain the equipment operation monitoring parameters associated with each abnormal event in the abnormal evolution path, as well as the offset of the equipment operation monitoring parameter values ​​relative to the baseline value and the duration of the abnormal event. Step S53: Based on the equipment operation monitoring parameters, the deviation of the equipment operation monitoring parameter values ​​from the benchmark value, and the duration of the abnormal event, the current operation risk level of the coal mine power supply system is obtained by matching from the preset risk level judgment rules.

9. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 8, characterized in that, In step S53, the preset risk level judgment rule specifically includes: The coal mine power supply system is pre-defined as having four risk levels: low risk, medium risk, high risk, and emergency risk. For each risk level, corresponding parameters are configured, including the number of abnormal equipment operation monitoring parameters, the threshold for the deviation of equipment operation monitoring parameter values ​​from the benchmark value, and the threshold for the duration of abnormal events. If only a single device has abnormal operating parameters, and its deviation from the baseline value does not exceed the preset low-amplitude deviation threshold and the duration of the corresponding abnormal event is less than the preset short-term threshold, it is judged as low risk. When multiple or a single device has abnormal operating monitoring parameters, and the deviation of its value relative to the benchmark value exceeds the preset low-amplitude deviation threshold but does not exceed the preset medium-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset short-term threshold and the preset medium-term threshold, it is judged as medium risk. When multiple devices have abnormal operating monitoring parameters, and their deviation from the benchmark value exceeds the preset medium-amplitude deviation threshold but does not exceed the preset high-amplitude deviation threshold, and the duration of the corresponding abnormal event is between the preset medium-time threshold and the preset high-time threshold, it is judged as high risk; When the deviation of a device's operating monitoring parameter value from the baseline value exceeds a preset high deviation threshold, and the duration of the corresponding abnormal event reaches a preset high time threshold, it is determined to be an emergency risk.

10. The intelligent interlocking method for coal mine power supply systems based on active defense according to claim 1, characterized in that, In step S6, the locking command specifically includes: Power outages are implemented on electrical equipment corresponding to abnormal events located at the end of the time sequence in the abnormal evolution path.

Images