Air compressor unit intelligent operation and maintenance management planning method and system

Abstract

The application discloses a kind of air compressor unit intelligent operation and maintenance management planning method and system, it is related to air compressor intelligent operation and maintenance technical field, comprising: obtaining the real-time operation data of mine mobile air compressor multistage air intake filter device, wherein multistage air intake filter device includes rough filter and fine filter;According to rough filter, fine filter differential pressure before and after and real-time operation data, determine rough filter self-cleaning device working state;When fine filter differential pressure rise rate exceeds first threshold and rough filter differential pressure drop amplitude is lower than second threshold, determine rough filter self-cleaning device failure;Calculate rough filter effective dust saturation and plan rough filter cleaning maintenance timing;Predict fine filter remaining dust capacity, determine fine filter replacement warning moment, and generate operation and maintenance management planning scheme.The problem that filter core maintenance is carried out in prior art by relying on fixed cycle or artificial experience, filter core pollution degree and remaining life cannot be accurately evaluated, leading to filter core overuse damage host or early replacement causing spare parts waste problem.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for air compressors, specifically to an intelligent operation and maintenance management planning method and system for air compressor units. Background Technology

[0002] As the operational environments in fields such as mining and construction become increasingly complex, mobile air compressors are widely used in on-site air supply and equipment operation. During operation, air compressors require an intake filter to filter the air entering the compressor, reducing the adverse effects of dust, particulate matter, and other impurities on the internal components. Due to the high dust concentration in mining and other working environments, dust gradually accumulates in the intake filter over long-term operation, increasing filtration resistance and affecting airflow and equipment performance.

[0003] In existing technologies, intake filters are typically inspected and maintained periodically based on equipment operating time, cumulative mileage, or maintenance personnel experience. However, in actual operation, due to differences in environmental dust concentration, load conditions, and operating conditions, the contamination rate of filters may vary significantly between different devices or even at different stages of the same device's operation. Fixed maintenance cycles are difficult to fully adapt to actual operating conditions.

[0004] Furthermore, during equipment operation, the contamination status of the filter unit usually requires shutdown for inspection or disassembly for testing before it can be further confirmed, making it difficult for maintenance personnel to promptly grasp the current operating status of the filter unit. When the filter unit is highly contaminated, it may affect the equipment's air intake efficiency and operational stability; while performing maintenance when the filter unit still has some usability will increase the number of equipment shutdowns and maintenance costs.

[0005] Therefore, how to rationally plan the maintenance needs of the air compressor intake filter based on its actual operating status, so as to improve the maintenance management level during equipment operation, has become an urgent technical problem to be solved. Summary of the Invention

[0006] This invention provides an intelligent operation and maintenance management planning method and system for air compressor units to solve the technical problem that in the prior art, filter element maintenance mainly relies on fixed cycles or manual experience judgment, which cannot accurately identify the actual degree of filter element contamination and remaining service life, resulting in excessive use of filter elements damaging the main unit or premature replacement causing waste of spare parts.

[0007] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: In a first aspect, the present invention provides an intelligent operation and maintenance management planning method for air compressor units, comprising: Real-time operating data of a multi-stage air intake filtration device for a mobile air compressor in a mine is obtained, wherein the multi-stage air intake filtration device includes a coarse filter and a fine filter; Based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with the real-time operating data, the working status of the coarse filter self-cleaning device is determined. When the rate of increase of the pressure difference before and after the fine filter exceeds the first threshold and the rate of decrease of the pressure difference before and after the coarse filter does not reach the second threshold within a preset time window, the coarse filter self-cleaning device is determined to be faulty. When it is determined that the coarse filter self-cleaning device has failed, the current effective dust holding capacity saturation of the coarse filter is calculated, and the cleaning and maintenance sequence of the coarse filter is planned according to the current effective dust holding capacity saturation. Predict the remaining dust holding capacity of the fine filter, and determine the early warning time for replacing the fine filter based on the remaining dust holding capacity of the fine filter and the current effective dust holding saturation, and output the operation and maintenance management plan.

[0008] Secondly, the present invention provides an intelligent operation and maintenance management planning system for air compressor units, comprising: The operation data acquisition module is used to acquire real-time operation data of the multi-stage air intake filtration device of the mobile air compressor in the mine, wherein the multi-stage air intake filtration device includes a coarse filter and a fine filter; The self-cleaning status analysis module is used to determine the working status of the coarse filter self-cleaning device based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with the real-time operating data. The coarse filter maintenance planning module is used to determine that the coarse filter self-cleaning device has failed when the rate of increase of the pressure difference before and after the fine filter exceeds a first threshold and the rate of decrease of the pressure difference before and after the coarse filter does not reach a second threshold within a preset time window. The fine filter capacity prediction module is used to calculate the current effective dust holding capacity saturation of the coarse filter when it is determined that the coarse filter self-cleaning device has failed, and to plan the cleaning and maintenance sequence of the coarse filter based on the current effective dust holding capacity saturation. The joint operation and maintenance planning module is used to predict the remaining dust holding capacity of the fine filter, and determine the fine filter replacement warning time based on the remaining dust holding capacity of the fine filter and the current effective dust holding saturation, and output the operation and maintenance management plan.

[0009] One or more technical solutions provided in this invention have at least the following technical effects or advantages: This invention acquires real-time operating data of the coarse and fine filters in a multi-stage air intake filtration system, and combines this data with pressure difference changes before and after the coarse and fine filters, as well as self-cleaning information, to analyze the working status of the coarse filter self-cleaning device, enabling accurate identification of abnormal coarse filter conditions. Furthermore, by combining parameters such as coarse filter pressure difference change trends, effective dust saturation, and operating time, the remaining operating time of the coarse filter is predicted, allowing for reasonable planning of the coarse filter cleaning and maintenance sequence. Simultaneously, by establishing a dust capacity prediction model for the fine filter, integrating the coarse filter contamination status, fine filter pressure difference growth trends, and cumulative equipment operating time, the remaining dust capacity of the fine filter is predicted, and the fine filter replacement warning time is determined. Finally, by combining the coarse filter cleaning and maintenance time with the fine filter replacement warning time, a joint shutdown maintenance window is generated, achieving coordinated planning of the coarse and fine filter maintenance processes. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of a method for intelligent operation and maintenance management planning of air compressor units provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the operating status monitoring principle of the multi-stage air intake filtration device provided in this embodiment of the invention; Figure 3 This is a schematic diagram of the structure of an intelligent operation and maintenance management planning system for air compressor units provided in an embodiment of the present invention.

[0011] In the attached diagram, the components represented by each number are as follows: The system includes a data acquisition module 11, a self-cleaning status analysis module 12, a coarse filter maintenance planning module 13, a fine filter capacity prediction module 14, and a joint operation and maintenance planning module 15. Detailed Implementation

[0012] This invention provides an intelligent operation and maintenance management planning method and system for air compressor units, which specifically solves the technical problem that in the prior art, filter element maintenance mainly relies on fixed cycles or manual experience judgment, which cannot accurately identify the actual degree of pollution and remaining service life of the filter element, resulting in excessive use of the filter element damaging the main unit or premature replacement causing waste of spare parts.

[0013] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0014] Example 1, as Figure 1 and Figure 2 As shown, this invention provides an intelligent operation and maintenance management planning method for air compressor units, including: S100: Acquire real-time operating data of the multi-stage air intake filtration device of the mobile air compressor in the mine, wherein the multi-stage air intake filtration device includes coarse filter and fine filter; Step S100 in the method provided in this embodiment of the invention includes: Collect the upstream and downstream pressures of the coarse filter and calculate the pressure difference before and after the coarse filter. Collect the upstream and downstream pressures of the fine filter and calculate the pressure difference before and after the fine filter. Record the cumulative number of actions of the coarse filter self-cleaning device since the last coarse filter cleaning and reset, as well as the time of each action. Read the factory-calibrated dust holding capacity and the termination pressure difference of the coarse filter. The factory-calibrated dust holding capacity is the total mass of test dust that the coarse filter can hold when it reaches the termination pressure difference under standard test conditions. The termination pressure difference is used to calculate the current effective dust holding capacity saturation of the coarse filter by combining the current pressure difference value before and after the coarse filter. Record the cumulative runtime since the last coarse filter cleaning and reset.

[0015] In this embodiment, the mobile air compressor for mining operates in a mining environment for extended periods. Since the air in the working area contains a large amount of particulate matter such as mine dust and coal dust, the air entering the compressor needs to be filtered. To reduce the impact of dust particles on the compressor's operating performance, a multi-stage intake filtration system is installed before the air enters the compressor.

[0016] The multi-stage air intake filtration device includes a coarse filter assembly and a fine filter assembly arranged sequentially along the airflow direction. The coarse filter assembly is used to perform preliminary filtration of large-particle dust, and the fine filter assembly is used to further filter the air after coarse filtration.

[0017] Specifically, real-time operating data of the multi-stage air intake filtration device is obtained, including the pressure difference before and after the coarse filter, the pressure difference before and after the fine filter, the operation data of the coarse filter self-cleaning device, the factory calibration parameters of the coarse filter, and the cumulative running time of the coarse filter.

[0018] The real-time operating data is collected by pressure detection sensors, self-cleaning control modules, and air compressor control systems installed on the multi-stage intake filter.

[0019] First, the upstream pressure and downstream pressure of the coarse filter module are collected, and the pressure difference before and after the coarse filter is calculated.

[0020] Specifically, a first pressure detection sensor is installed on the air inlet side of the coarse filter assembly, and a second pressure detection sensor is installed on the air outlet side of the coarse filter assembly. The first pressure detection sensor is used to obtain the upstream pressure value of the coarse filter in real time, and the second pressure detection sensor is used to obtain the downstream pressure value of the coarse filter in real time.

[0021] The pressure detection sensor collects pressure data according to a preset sampling period, which is set according to the data processing capability of the air compressor control system. In this embodiment, it is set to 5 seconds.

[0022] As the coarse filter continuously filters dust particles from the air, the airflow resistance increases as dust gradually adheres to the filter media surface, creating a pressure difference between the upstream and downstream pressures of the coarse filter. The pressure difference across the coarse filter is calculated as follows: Pressure difference before and after coarse filtration = upstream pressure of coarse filtration - downstream pressure of coarse filtration.

[0023] For example, during the operation of a mobile air compressor in a mine, the pressure sensor collects the upstream pressure value of the coarse filter as 0.102 MPa and the downstream pressure value as 0.096 MPa. Then, the pressure difference before and after the coarse filter is 0.102 MPa - 0.096 MPa = 0.006 MPa, which means that the current pressure difference before and after the coarse filter is 6 kPa.

[0024] The pressure difference before and after the coarse filter is used to characterize the current filtration resistance of the coarse filter assembly. When the pressure difference before and after the coarse filter continues to increase, it indicates that the degree of dust accumulation inside the coarse filter is increasing and the filtration capacity is gradually decreasing.

[0025] Furthermore, the upstream pressure and downstream pressure of the fine filter module are collected, and the pressure difference before and after fine filtration is calculated.

[0026] Specifically, pressure sensors are installed on both the inlet and outlet sides of the fine filter assembly to acquire the upstream and downstream pressure values ​​of the fine filter, respectively. The pressure difference across the fine filter is calculated as follows: Pressure difference before and after fine filtration = upstream pressure of fine filtration - downstream pressure of fine filtration.

[0027] For example, under a certain operating condition, the upstream pressure of the fine filter is detected to be 0.095 MPa and the downstream pressure is detected to be 0.086 MPa. Then the pressure difference before and after the fine filter is 0.095 MPa - 0.086 MPa = 0.009 MPa, that is, the current pressure difference before and after the fine filter is 9 kPa.

[0028] By simultaneously acquiring the pressure difference values ​​before and after coarse filtration and fine filtration, the relationship between dust load changes between the coarse and fine filtration components can be analyzed. When the coarse filtration capacity decreases, the amount of dust not effectively trapped increases, causing the fine filtration component to bear a greater filtration load, thus leading to a faster increase in the pressure difference value before and after fine filtration.

[0029] Furthermore, the cumulative number of actions of the coarse filter self-cleaning device since the last coarse filter cleaning and reset, as well as the timing of each action, are recorded.

[0030] In this embodiment, the coarse filter assembly is equipped with a self-cleaning device, which uses a back-blowing method to clean the dust on the surface of the coarse filter material.

[0031] The coarse filter cleaning and reset process includes the process of the control system re-recording the coarse filter's operating status after manual cleaning, filter element replacement, or one effective automatic cleaning.

[0032] To determine the actual operating effect of the self-cleaning device, the cumulative number of cleaning actions performed by the self-cleaning device from the completion of the last coarse filter cleaning and reset to the current time is recorded, along with the time corresponding to each cleaning action.

[0033] For example, during the operation of a mobile air compressor in a mine, the control system recorded the self-cleaning device's activation times as July 3, 2026, at 09:00, 12:00, 15:00, and 18:00, respectively, since the last coarse filter cleaning and reset. Therefore, the current cumulative activation count of the coarse filter self-cleaning device is 4 times.

[0034] The number of actions and the timing of the actions mentioned above are used in subsequent steps to match and analyze the pressure difference changes before and after coarse filtration, in order to determine whether each self-cleaning action effectively reduces the coarse filtration resistance.

[0035] Furthermore, the factory-calibrated dust holding capacity and termination pressure difference of the coarse filter component are read.

[0036] The factory-calibrated dust holding capacity and the termination pressure differential are pre-stored in the equipment parameter database of the air compressor control system, which originates from standard test data provided by the coarse filter component manufacturer. The factory-calibrated dust holding capacity is the total mass of test dust that the coarse filter component can hold when operating under standard test conditions up to the termination pressure differential; the termination pressure differential is the maximum allowable pressure differential value corresponding to the coarse filter component reaching its design maintenance state, and is used as a benchmark parameter for evaluating the dust load level of the coarse filter.

[0037] Furthermore, based on the current pressure difference before and after the coarse filter, the final pressure difference, and the factory-calibrated dust holding capacity, the current effective dust holding capacity saturation of the coarse filter is calculated.

[0038] The current dust holding capacity is calculated as follows: Current consumed dust capacity = factory rated dust capacity of coarse filter × (current pressure difference before and after coarse filter ÷ final pressure difference of coarse filter).

[0039] Furthermore, the current effective dust holding capacity saturation of the coarse filter is calculated as follows: The current effective dust holding capacity saturation of the coarse filter = (current consumed dust holding capacity ÷ factory rated dust holding capacity of the coarse filter) × 100%.

[0040] For example, the coarse filter assembly used in a mobile air compressor in a mine has a factory-calibrated dust holding capacity of 800g and a final pressure difference of 20kPa. The current measured pressure difference before and after the coarse filter is 12kPa. Therefore, the currently consumed dust holding capacity = 800g × (12kPa ÷ 20kPa) = 480g. Further calculation shows that the current effective dust holding capacity saturation of the coarse filter = (480g ÷ 800g) × 100% = 60%.

[0041] Furthermore, the cumulative runtime since the last coarse filter cleaning and reset is recorded.

[0042] Specifically, the continuous operating time of the coarse filter component after its last cleaning and reset is obtained by reading the operation record of the mobile air compressor control system in the mine. The cumulative operating time is calculated as follows: Cumulative runtime = current time - last coarse filter cleaning reset time.

[0043] For example, if a coarse filter component completes cleaning and reset at 12:00 on July 1, 2026, and the current detection time is 12:00 on July 3, 2026, then the cumulative running time = 12:00 on July 3, 2026 - 12:00 on July 1, 2026 = 48 hours.

[0044] The cumulative runtime is used to analyze the dust accumulation rate in conjunction with the effective dust holding capacity saturation of the coarse filter. When the effective dust holding capacity saturation of the coarse filter increases rapidly within the same operating time, it indicates that the dust concentration in the current mining operation area is high, and the subsequent maintenance cycle needs to be shortened.

[0045] In summary, this step, by collecting pressure difference values ​​before and after coarse filtration, pressure difference values ​​before and after fine filtration, self-cleaning device operation information, coarse filtration calibration parameters, and cumulative runtime, establishes a data foundation that reflects the operating status of the multi-stage intake filtration system of the mobile air compressor in the mine. Specifically, the pressure difference data between the coarse and fine filters describes changes in filtration resistance, the self-cleaning action data evaluates the effectiveness of the coarse filter's self-cleaning, the factory-calibrated dust holding capacity and the final pressure difference establish a benchmark for evaluating the coarse filter's dust holding capacity, and the cumulative runtime describes the long-term operating status of the filtration system. Through the fusion of these multi-dimensional data, a reliable basis can be provided for subsequently determining the failure status of the coarse filter self-cleaning device, predicting coarse filter maintenance time, and planning early warning times for fine filter replacement.

[0046] S200: Based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with real-time operating data, determine the working status of the coarse filter self-cleaning device; Step S200 in the method provided in this embodiment of the invention includes: A fixed time length is extracted forward from the current time to construct an analysis time window. The pressure difference values ​​before and after coarse filtration corresponding to each collection time within the analysis time window are extracted to form a coarse filtration pressure difference time series. Traverse the coarse filtration pressure difference time series, identify and analyze each local maximum point and the subsequent local minimum point adjacent to each local maximum point within the analysis time window, subtract the coarse filtration pressure difference value corresponding to the local minimum point from the coarse filtration pressure difference value corresponding to the local maximum point, and use it as the pressure difference recovery amount. Record the time corresponding to the local maximum point as the recovery start time and the time corresponding to the local minimum point as the recovery end time. Match the recovery start time with the action time of the coarse filter self-cleaning device. When the time deviation between the recovery start time and the action time is less than the allowable deviation, mark the corresponding differential pressure recovery amount as the effective recovery amount. The number of effective recovery amounts is counted, and the arithmetic mean of all effective recovery amounts is calculated as the average recovery rate. When the number of effective recovery quantities is less than the product of the total number of action moments within the analysis time window and the proportional coefficient, it is determined that the self-cleaning device has missing actions. When the average recovery amplitude is less than the product of the average historical effective recovery amplitude and the attenuation coefficient, it is determined that the self-cleaning device has an efficiency decay. The average historical effective recovery amplitude is determined based on the arithmetic mean of the average recovery amplitudes corresponding to multiple analysis time windows before the current time. When it is determined that there is a missing action or a decrease in execution efficiency, the working status of the self-cleaning device will be marked as an abnormal working status.

[0047] In this embodiment, based on the real-time operating data of the multi-stage intake filtration device of the mobile air compressor in the mine obtained by performing the aforementioned steps, the actual cleaning effect of the coarse filter self-cleaning device is further analyzed according to the pressure difference change process before and after the coarse filter, so as to determine whether the coarse filter self-cleaning device is in an abnormal working state.

[0048] Specifically, firstly, taking the current moment as the end point, a data interval of a preset time length is extracted to construct an analysis time window.

[0049] The length of the analysis time window is determined based on the operating conditions of the mobile air compressor in the mine, the operation cycle of the self-cleaning device, and the sampling cycle of the pressure detection sensor.

[0050] For example, for a certain model of mobile air compressor for mining, its coarse filter self-cleaning device usually performs a cleaning action every 3 hours, and the pressure detection sensor sampling period is 5 seconds. Therefore, the analysis time window is set to 24 hours to ensure that multiple self-cleaning actions are included in a single analysis time window.

[0051] Furthermore, the pressure difference values ​​before and after coarse filtration corresponding to each acquisition time within the analysis time window are extracted and arranged in chronological order of acquisition time to form a time series sequence of coarse filtration pressure difference.

[0052] Among them, the coarse filter pressure difference time series is used to characterize the change process of the coarse filter component's filtration resistance with operating time.

[0053] Furthermore, the coarse filter pressure difference time series is traversed to identify and analyze each local maximum point within the time window and the subsequent local minimum point corresponding to each local maximum point.

[0054] Among them, the local maximum point represents the state point where the dust accumulation of the coarse filter component causes the filtration resistance to reach the maximum value in a certain stage before a self-cleaning action; the local minimum point represents the state point where the coarse filter resistance recovers in a certain stage after the self-cleaning action is completed.

[0055] Specifically, the pressure difference between the local maximum point and the local minimum point is used to calculate the pressure difference before and after coarse filtration, thus obtaining the first pressure difference recovery amount. The pressure difference recovery amount is calculated as follows: Pressure difference recovery = Pressure difference before and after coarse filtration at the local maximum point - Pressure difference before and after coarse filtration at the local minimum point.

[0056] Meanwhile, the acquisition time corresponding to the local maximum point is recorded as the recovery start time, and the acquisition time corresponding to the local minimum point is recorded as the recovery end time.

[0057] For example, during a coarse filter self-cleaning process, the pressure difference before and after the coarse filter was detected to be 15 kPa before the self-cleaning action. After the self-cleaning action was completed, the pressure difference before and after the coarse filter decreased to 10 kPa. Then, the pressure difference recovery amount = 15 kPa - 10 kPa = 5 kPa. That is, the self-cleaning action produced a pressure difference recovery effect of 5 kPa.

[0058] Furthermore, to avoid misinterpreting pressure fluctuations during normal operation as a self-cleaning effect, this embodiment matches the recovery start time with the operation time of the coarse filter self-cleaning device.

[0059] Specifically, the time deviation between the recovery start time and the self-cleaning device activation time is calculated. The time deviation is calculated as follows: Time deviation = |Recovery start time - Self-cleaning device action time|.

[0060] When the time deviation is less than the preset deviation tolerance, the differential pressure recovery process is considered to be generated by the corresponding self-cleaning action, and the corresponding differential pressure recovery amount is marked as the effective recovery amount.

[0061] The allowable deviation is determined based on the self-cleaning device control response time, the backflush airflow transmission time, and the pressure sensor sampling delay time.

[0062] For example, statistics show that for a certain model of mobile air compressor used in mining, after the self-cleaning device is activated, the time for the backflushing airflow to reach the surface of the coarse filter media is less than 20 seconds, and the sampling delay time of the pressure detection sensor is less than 5 seconds. Therefore, the allowable deviation is set to 30 seconds. If the self-cleaning action occurs at 10:00:00, and the corresponding recovery start time is detected at 10:00:08, then the time deviation = 8 seconds < 30 seconds. Therefore, the pressure difference recovery amount is determined as the effective recovery amount.

[0063] Furthermore, all valid recovery values ​​within the current analysis time window are statistically analyzed, and the arithmetic mean of all valid recovery values ​​is calculated to obtain the average recovery magnitude corresponding to the current analysis time window. The average recovery magnitude is calculated as follows: Average recovery rate = Sum of all effective recovery amounts ÷ Number of effective recovery amounts

[0064] For example, if three effective recovery values ​​are obtained within a 24-hour analysis time window, namely 4 kPa, 5 kPa and 6 kPa, then the average recovery amplitude = (4 kPa + 5 kPa + 6 kPa) ÷ 3 = 5 kPa.

[0065] The average recovery amplitude is used to evaluate the average cleaning effect produced by a single operation of the coarse filter self-cleaning device within the current analysis time window.

[0066] Furthermore, in order to determine whether there are any missing actions in the coarse filter self-cleaning device, this embodiment performs a matching analysis on the number of self-cleaning actions and the amount of effective recovery within the analysis time window.

[0067] Specifically, the system obtains all action moments of the self-cleaning device within the current analysis time window and counts the total number of actions actually triggered by the self-cleaning device within the corresponding time window.

[0068] The total number of actions indicates the number of times the control system sends cleaning commands to the self-cleaning device.

[0069] Furthermore, the differential pressure recovery process after each self-cleaning action is time-matched with the corresponding action time. When the time deviation between the recovery start time and the self-cleaning action time meets the preset deviation tolerance, the recovery amount corresponding to the differential pressure recovery process is determined as the effective recovery amount.

[0070] The effective recovery quantity represents the number of self-cleaning actions whose actual cleaning effect can be detected within the current analysis time window.

[0071] Furthermore, due to potential influences from pressure sensor sampling errors, dust disturbances, and short-term operating condition changes during actual operation, some normal cleaning actions may not result in a significant pressure differential recovery process. Therefore, an effective action ratio coefficient is set to determine the relationship between the number of effective recovery actions and the total number of actions. Specifically, the criteria for determining missing actions are as follows: The effective recovery quantity is less than the total number of action moments within the analysis time window multiplied by the proportional coefficient.

[0072] The proportionality coefficient is used to characterize the minimum proportional requirement required for the self-cleaning action to generate an effective differential pressure recovery response under normal conditions. It is calibrated based on historical operating data of the equipment.

[0073] Specifically, the total number of self-cleaning actions and the corresponding effective recovery quantity are calculated within multiple analysis time windows under normal equipment operation. Within each time window: Effective percentage of movement = number of effective recovery units ÷ total number of movements.

[0074] Subsequently, the effective proportion of actions corresponding to multiple normal operating windows was statistically analyzed, and the lower quantile value was used as the proportion coefficient.

[0075] For example, data on the continuous 30-day normal operation of a certain type of mobile air compressor in a mine was collected, and the effective percentages of actions within each analysis time window were statistically obtained as follows: 100%, 100%, 92%, 95%, and 90%.

[0076] Statistical analysis shows that the effective rate of actions under normal operating conditions is no less than 90%, therefore the ratio is set to 0.90.

[0077] Furthermore, to adapt to the performance differences caused by different mining environments, different air compressor models, and different coarse filter structures, this embodiment uses the equipment's own historical operating data to dynamically determine the average historical effective recovery range, instead of using a fixed threshold.

[0078] Specifically, the effective recovery sets corresponding to multiple analysis time windows prior to the current time are extracted, and the arithmetic mean of the effective recovery within each analysis time window is calculated to obtain the average recovery magnitude across multiple windows. The average recovery magnitude across windows is calculated as follows: Average recovery margin within a window = Sum of all valid recovery values ​​within the corresponding analysis time window ÷ Number of valid recovery values ​​within the corresponding analysis time window.

[0079] Furthermore, the average recovery magnitudes of multiple windows are sorted, and the maximum and minimum values ​​are removed to reduce the impact of random anomalies on the evaluation results.

[0080] Then, the arithmetic mean of the remaining window's average recovery magnitude is calculated to obtain the historical average effective recovery magnitude. The historical average effective recovery magnitude is calculated as follows: Historical effective recovery average = Sum of average window recovery averages after removing the maximum and minimum values ​​÷ Number of remaining windows.

[0081] For example, the average recovery amplitude of the five analysis time windows preceding the current moment is obtained, which are 4.8 kPa, 5.2 kPa, 5.0 kPa, 5.5 kPa, and 7.0 kPa, respectively. After removing the maximum value of 7.0 kPa and the minimum value of 4.8 kPa, the average historical effective recovery amplitude is (5.2 kPa + 5.0 kPa + 5.5 kPa) ÷ 3 = 5.23 kPa.

[0082] The historical effective recovery average is continuously updated as the equipment operates, rather than being a fixed parameter.

[0083] In the application scenario of mobile air compressors in mines, the average historical effective recovery range is usually affected by the air compressor model, coarse filter model and working air volume, with a typical range of 0.5kPa to 5.0kPa.

[0084] Furthermore, based on the principle that the cleaning efficiency of coarse filter media naturally decreases with increasing operating time, an efficiency attenuation coefficient is introduced to evaluate the current average recovery rate.

[0085] The efficiency decay coefficient characterizes the degree of decrease in the allowable recovery range of the self-cleaning device as the coarse filter's dust holding capacity increases and the filter media porosity decreases. Determined based on historical maintenance data and operational experience with similar equipment, its value ranges from 0.70 to 0.95.

[0086] For example, a lower efficiency attenuation coefficient can be set for mining environments with high dust concentration and rapid filter material contamination; a higher efficiency attenuation coefficient can be set for mining environments with stable operation and low dust concentration.

[0087] Furthermore, the average recovery magnitude corresponding to the current analysis time window is compared with the product of the historical average effective recovery magnitude and the execution efficiency decay coefficient. The execution efficiency decay judgment criteria are determined as follows: The current average recovery rate is less than the historical average effective recovery rate multiplied by the execution efficiency decay coefficient.

[0088] When the above judgment conditions are met, it is determined that the coarse filter self-cleaning device has a decrease in execution efficiency.

[0089] For example, if the historical effective recovery amplitude is averaged at 5.0 kPa and the execution efficiency attenuation coefficient is set to 0.8, then the historical effective recovery amplitude average × execution efficiency attenuation coefficient = 5.0 kPa × 0.8 = 4.0 kPa. If the average recovery amplitude calculated in the current analysis time window is 3.2 kPa, then 3.2 kPa < 4.0 kPa. Therefore, it is determined that the coarse filter self-cleaning device exhibits execution efficiency attenuation.

[0090] Furthermore, when it is determined that the coarse filter self-cleaning device is experiencing a decline in operating efficiency, the operating status of the coarse filter self-cleaning device is marked as an abnormal operating status.

[0091] For example, during normal operation, the self-cleaning device of a mobile air compressor in a mine can reduce the pressure difference before and after the coarse filter by an average of 5 kPa with a single operation. However, the average recovery rate has dropped to 3 kPa over several consecutive analysis time windows, indicating that although the self-cleaning device can perform the operation, its actual cleaning capacity has decreased and further maintenance and inspection are required.

[0092] Step S200 in the method provided in this embodiment of the invention further includes: The difference between the total number of self-cleaning device action moments and the number of effective recovery moments within the analysis time window is used as the invalid action count; The missing action rate is obtained by dividing the number of invalid actions by the total number of action moments. When the missing action rate exceeds the missing action rate tolerance threshold, the self-cleaning device's working status is marked as an action failure state.

[0093] Based on the effective recovery quantity set and coarse filter self-cleaning device action record obtained in the above embodiment, this embodiment further determines whether the coarse filter self-cleaning device has an action failure based on the correspondence between the number of self-cleaning device actions and the number of effective recovery quantities.

[0094] Specifically, the total number of times the coarse filter self-cleaning device operates within the current analysis time window is counted, and the number of effective recovery quantities determined in the aforementioned steps is obtained.

[0095] The effective recovery quantity refers to the number of self-cleaning actions that successfully match the operation time of the coarse filter self-cleaning device within the current analysis time window and can produce an effective differential pressure recovery effect.

[0096] Furthermore, based on the difference between the total number of self-cleaning device operation moments and the effective recovery quantity, the number of invalid operations that did not produce an effective cleaning effect is determined. The invalid operation count is calculated as follows: Invalid action count = Total number of self-cleaning device operation times - Effective recovery quantity.

[0097] For example, within a certain analysis time window, the total number of times the coarse filter self-cleaning device operated was recorded as 8. After differential pressure recovery analysis, the number of effective recovery actions was determined to be 6. Therefore, the count of invalid actions = 8 - 6 = 2. This indicates that within the current analysis time window, there were 2 self-cleaning actions that did not produce the corresponding effective differential pressure recovery effect.

[0098] Furthermore, to eliminate the impact of differences in the number of self-cleaning actions within different operating cycles, the number of invalid actions is normalized to obtain the action missing rate. The action missing rate is calculated as follows: Missing action rate = Invalid action count ÷ Total number of self-cleaning device action times.

[0099] For example, using the above example, the action missing rate = 2 times ÷ 8 times = 25%, which means that 25% of the self-cleaning actions did not form an effective cleanup response within the current analysis time window.

[0100] Furthermore, to accommodate the differences in structure between different models of mobile air compressors for mining and different coarse filter components, the tolerance threshold for the action loss rate in this embodiment is determined based on historical data from the normal operation phase of the equipment itself.

[0101] Specifically, during the first cleaning cycle after the coarse filter self-cleaning device is installed, multiple analysis time windows are selected when the equipment is in normal operation, and the action missing rate corresponding to each analysis time window is calculated.

[0102] The normal operating status is determined based on the equipment operation records, including that the air compressor operating parameters are within the normal range, the pressure difference before and after the coarse filter changes in a stable trend, and no manual maintenance, filter element replacement, or fault alarm occurs.

[0103] Furthermore, statistical analysis was performed on the missing action rate corresponding to each analysis time window during normal operation, and the tolerance threshold for the missing action rate was determined based on the statistical results. Specifically, it was determined as follows: The tolerance threshold for missing actions is the lower confidence level of the statistical distribution of missing actions during normal operation.

[0104] In this embodiment, the statistical boundary corresponding to the 95% confidence level is taken as the tolerance threshold for the action missing rate.

[0105] For example, during the initial normal operation phase of a mobile air compressor in a mine, data was collected over 30 analysis time windows. Statistical analysis showed that the missing action rate was mainly distributed in the range of 5% to 12%. Therefore, based on a 95% confidence level, the tolerance threshold for the missing action rate was determined to be 15%.

[0106] This threshold is recalibrated based on changes in equipment model, coarse filter structure, and operating environment, rather than using a fixed empirical value.

[0107] Furthermore, the action missing rate corresponding to the current analysis time window is compared with the action missing rate tolerance threshold. Specifically, when the action missing rate is greater than the action missing rate tolerance threshold, it is determined that the coarse filter self-cleaning device has an action failure, and the working status of the self-cleaning device is marked as an action failure state.

[0108] For example, the current analysis time window calculates an action missing rate of 25%, while the action missing rate tolerance threshold determined based on the equipment's historical normal operation data is 15%. Therefore, 25% > 15%. Thus, it is determined that the current coarse filter self-cleaning device is in an action failure state.

[0109] At this point, it indicates that although the self-cleaning device has performed control actions, some actions have not generated corresponding differential pressure recovery responses. There may be problems such as insufficient backflushing air source pressure, abnormal actuator, or blocked spray channel, requiring further maintenance and inspection.

[0110] In summary, this step, through time-series analysis of the pressure difference changes before and after coarse filtration, and combining the correspondence between the self-cleaning action timing and the pressure difference recovery response, evaluates the working status of the coarse filter self-cleaning device from two dimensions: the effectiveness of action execution and the ability to maintain cleaning effect. Specifically, the number of effective recovery values ​​determines whether there are missing self-cleaning actions, the comparison between the average recovery amplitude and historical effective recovery amplitude determines whether the cleaning efficiency has decreased, and the action failure rate is further analyzed to determine action failure. This allows for accurate identification of abnormal states of the coarse filter self-cleaning device, providing a basis for subsequent fault warning and maintenance decisions.

[0111] S300: When the rate of increase of the pressure difference before and after fine filtration exceeds the first threshold and the rate of decrease of the pressure difference before and after coarse filtration does not reach the second threshold within a preset time window, the coarse filtration self-cleaning device is determined to be faulty. Step S300 in the method provided in this embodiment of the invention includes: Within a preset time window, a sampling sequence of pressure difference values ​​before and after fine filtration is extracted. Based on the difference between the pressure difference values ​​before and after fine filtration corresponding to the start and end times of the preset time window, combined with the duration of the preset time window, the rising rate of the pressure difference before and after fine filtration is used. Acquire all action times of the coarse filter self-cleaning device within a preset time window, extract the pressure difference value before and after the coarse filter corresponding to each action time as the pressure difference value before each cleaning, and extract the pressure difference value before and after the sampling time after each action time when the pressure difference drops back to the stable platform area as the pressure difference value after each cleaning. Calculate the difference between the pressure difference before each cleaning and the corresponding pressure difference after each cleaning to obtain the single drop pressure difference for each cleaning. Take the arithmetic mean of all single drop pressure differences as the decrease in pressure difference before and after coarse filtration. When the rate of increase of the pressure difference before and after the fine filter exceeds the first threshold, and the rate of decrease of the pressure difference before and after the coarse filter is less than the second threshold, the coarse filter self-cleaning device is determined to have failed.

[0112] Furthermore, to further distinguish between reduced efficiency and complete failure of the coarse filter self-cleaning device, this embodiment combines the pressure difference growth trend before and after fine filtration with the pressure difference recovery capability after coarse filter self-cleaning to determine the failure of the coarse filter self-cleaning device.

[0113] Specifically, within a preset time window, differential pressure data continuously collected by differential pressure sensors before and after fine filtration are acquired, and a sequence of differential pressure changes during fine filtration is constructed according to the time sequence of acquisition.

[0114] The pressure difference before and after fine filtration is used to characterize the change in filtration resistance after the air filtered by coarse filtration enters the fine filtration component.

[0115] Furthermore, based on the pressure difference before and after fine filtration at the start and end times of the preset time window, the changing trend of the fine filtration resistance within that time window is calculated. Specifically, the rate of increase of the pressure difference before and after fine filtration is calculated as follows: Rate of increase of pressure difference before and after fine filtration = (Pressure difference value before and after fine filtration at the end of the preset time window - Pressure difference value before and after fine filtration at the beginning of the preset time window) ÷ Duration of preset time window.

[0116] The rate of increase in pressure difference before and after fine filtration is used to characterize the degree of increase in fine filtration resistance per unit time.

[0117] When the coarse filter self-cleaning device is working normally, most of the dust can be intercepted by the coarse filter and removed through the self-cleaning process before entering the fine filter component. Therefore, the dust load on the fine filter component is low, and the rate of increase of the pressure difference before and after the fine filter is within the normal range.

[0118] When the coarse filter self-cleaning device fails, the coarse filter cannot effectively remove accumulated dust, causing a large amount of dust to continuously enter the fine filter component, which rapidly increases the filtration resistance of the fine filter. As a result, the rate of increase in pressure difference before and after the fine filter increases significantly.

[0119] Furthermore, obtain all the action times of the coarse filter self-cleaning device within the current preset time window.

[0120] Specifically, based on the control records of the self-cleaning device, the start time corresponding to each backflushing action is determined and used as the time for the coarse filter cleaning action.

[0121] For each coarse filtration cleaning action, the pressure difference value before and after the coarse filtration is extracted at the time the action occurs, and used as the pressure difference value before the cleaning action is performed.

[0122] Meanwhile, after this cleaning action, the pressure difference change process before and after coarse filtration is analyzed. When the pressure difference decreases and enters a stable change stage, the pressure difference value at the sampling time of the corresponding stable platform area is taken as the post-cleaning pressure difference value after this cleaning action.

[0123] The stable platform area represents the stable operating state after the coarse filter resistance is restored following the completion of the self-cleaning process.

[0124] Furthermore, based on the pressure difference before and after each cleaning action, the pressure difference recovery effect generated by a single cleaning action is calculated.

[0125] Specifically, the pressure difference during a single pullback is calculated as follows: Single drop pressure difference = Pressure difference before cleaning - Pressure difference after cleaning.

[0126] Subsequently, the pressure difference of all single drops within the preset time window was statistically analyzed, and their arithmetic mean was calculated to obtain the decrease in coarse filtration pressure difference. Specifically: The decrease in coarse filter pressure differential = the sum of all single-time pressure differential drops ÷ the number of effective cleaning actions.

[0127] The decrease in coarse filter pressure differential is used to characterize the average dust removal effect actually produced by the coarse filter self-cleaning device within the current time window.

[0128] Furthermore, to avoid misjudgment based solely on the rate of increase in fine filtration pressure differential, this embodiment combines the fine filtration load growth status and the coarse filtration cleaning and recovery capability for joint judgment.

[0129] Specifically, when the rate of increase of the pressure difference before and after fine filtration is greater than the first threshold and the magnitude of decrease of the pressure difference before and after coarse filtration is less than the second threshold, the coarse filtration self-cleaning device is determined to be faulty.

[0130] Among them, if the rate of increase of the pressure difference before and after fine filtration is greater than the first threshold, it means that the dust load on the fine filtration component increases at a rate exceeding the normal operating range; if the rate of decrease of the pressure difference before and after coarse filtration is less than the second threshold, it means that the self-cleaning action of the coarse filtration cannot effectively reduce the filtration resistance.

[0131] Therefore, when both of the above conditions are met simultaneously, it can be determined that the coarse filter has failed to perform its normal filtration and protection function, that is, the coarse filter self-cleaning device is in a malfunctioning state.

[0132] Furthermore, the first and second thresholds are obtained by calibration based on historical normal operation data.

[0133] For example, during the operation of a mobile air compressor in a mine, an analysis time window was set to 24 hours. Within this time window, the pressure difference before and after fine filtration increased from 2 kPa to 8 kPa; therefore, the rate of increase in pressure difference before and after fine filtration = (8-2) ÷ 24 = 0.25 kPa / h. Simultaneously, three coarse filter self-cleaning actions were detected within this time window, with corresponding single pressure difference drops of 0.5 kPa, 0.4 kPa, and 0.3 kPa, respectively. Therefore, the coarse filter pressure difference decrease = (0.5 + 0.4 + 0.3) ÷ 3 = 0.4 kPa. If the current calibration yields a first threshold of 0.20 kPa / h and a second threshold of 0.50 kPa, then 0.25 kPa / h > 0.20 kPa / h and 0.4 kPa < 0.50 kPa. Therefore, it is determined that the fine filter load increase is abnormal, and the coarse filter cleaning recovery capacity is insufficient, confirming that the coarse filter self-cleaning device has failed.

[0134] Step S300 in the method provided in this embodiment of the invention further includes: The rising rate of the pressure difference before and after fine filtration for each of the multiple historical analysis time windows before the current moment is obtained, forming a set of historical rising rates. After removing the maximum and minimum values ​​in the set of historical rising rates, the upper quartile of the remaining values ​​is taken as the first threshold. The saturation correction factor is calculated based on the ratio of the current pressure difference before and after the coarse filter to the final pressure difference. The saturation correction factor is directly proportional to the difference between the final pressure difference and the current pressure difference before and after the coarse filter, and inversely proportional to the final pressure difference. The second threshold is obtained by multiplying the average historical effective recovery magnitude by the saturation correction factor.

[0135] Furthermore, to avoid the problem of large deviations in judgment results under different mining environments, different dust concentrations, and different equipment loads due to the use of fixed thresholds, this embodiment dynamically corrects the first threshold and the second threshold based on the equipment's own historical operating status and the current coarse filter load.

[0136] Specifically, regarding the determination of the first threshold, this embodiment uses statistical analysis based on the variation pattern of fine filtration pressure difference during historical normal operation.

[0137] First, obtain the rate of increase of the pressure difference before and after fine filtration within multiple historical analysis time windows before the current moment, and construct a set of historical rate of increase in chronological order.

[0138] The historical rate of increase set represents the historical range of changes in the rate of increase of the fine filter resistance at different time stages under normal operating conditions. Specifically, the historical rate of increase set is represented as: The set of historical rate of increase = {first historical rate of increase, second historical rate of increase, ..., nth historical rate of increase}.

[0139] Furthermore, in order to reduce the impact of abnormal factors such as accidental dust impacts and instantaneous fluctuations of sensors in historical data on the threshold calculation results, this embodiment first removes outliers from the historical rise rate set.

[0140] Specifically, the maximum and minimum values ​​in the historical rate of rise set are removed to obtain the effective historical rate of rise set.

[0141] Subsequently, the effective historical rate of increase set is statistically sorted, and its upper quartile is extracted as the first threshold.

[0142] The upper quartile represents the data distribution boundary at the 75th percentile after sorting. It can cover most of the differential pressure growth under normal equipment conditions, while retaining sensitivity to abnormally rapid growth trends.

[0143] Specifically, the first threshold is determined as follows: The first threshold is the value corresponding to the 75th percentile of the effective historical rate of ascent set.

[0144] For example, the rate of increase of the fine filter differential pressure corresponding to 10 historical analysis time windows prior to the current moment is obtained as follows: 0.08 kPa / h, 0.10 kPa / h, 0.09 kPa / h, 0.12 kPa / h, 0.11 kPa / h, 0.13 kPa / h, 0.10 kPa / h, 0.15 kPa / h, 0.09 kPa / h, and 0.30 kPa / h. Among these, 0.30 kPa / h represents a sudden change caused by abnormal dust impact, so the maximum value of 0.30 kPa / h is removed; the minimum value of 0.08 kPa / h is also removed. The remaining data, after sorting, are: 0.09 kPa / h, 0.09 kPa / h, 0.10 kPa / h, 0.10 kPa / h, 0.11 kPa / h, 0.12 kPa / h, 0.13 kPa / h, and 0.15 kPa / h. The corresponding upper quartile position is between 0.13 kPa / h and 0.15 kPa / h, and the first threshold is approximately 0.135 kPa / h obtained through linear interpolation. Therefore, when the rate of increase of the fine filtration pressure difference is greater than 0.135 kPa / h in the subsequent analysis window, it is considered that the rate of increase of the fine filtration load exceeds the normal variation range of the equipment.

[0145] Furthermore, in order to enable the second threshold to adapt to changes in the current level of contamination of the coarse filter, this embodiment dynamically corrects the average historical effective recovery range based on the ratio of the current pressure difference of the coarse filter to the termination pressure difference.

[0146] Specifically, obtain the current pressure difference before and after the coarse filter, as well as the termination pressure difference value at which the coarse filter is allowed to operate.

[0147] The termination pressure difference represents the maximum allowable pressure difference value when the coarse filter component reaches the maintenance or replacement condition, which is determined based on the rated resistance range of the coarse filter media and the equipment maintenance standards.

[0148] Furthermore, a saturation correction coefficient is constructed based on the current level of pollution in the coarse filter.

[0149] The saturation correction coefficient is used to represent the impact of the remaining recoverable space of the current coarse filter on the self-cleaning effect evaluation threshold. Specifically, the saturation correction coefficient is calculated as follows: Saturation correction factor = (Termination pressure difference - Current pressure difference before and after coarse filtration) ÷ Termination pressure difference.

[0150] Among them, the termination pressure difference - the current pressure difference before and after the coarse filter represents the remaining margin of the coarse filter at the current distance from the maximum allowable resistance state.

[0151] Therefore, when the current pressure difference of the coarse filter is low, the remaining margin is large and the saturation correction coefficient is large; when the coarse filter is close to being clogged, the remaining margin decreases and the saturation correction coefficient decreases.

[0152] Furthermore, based on the historical average effective recovery amplitude and the saturation correction coefficient, the second threshold corresponding to the current state is obtained. Specifically: Second threshold = average historical effective recovery magnitude × saturation correction coefficient.

[0153] The average historical effective recovery amplitude is obtained by statistically analyzing the effective recovery amount in the aforementioned multiple historical analysis time windows. It is used to represent the average differential pressure recovery capability that a single self-cleaning action can generate under normal equipment conditions.

[0154] For example, the historical effective recovery average of a certain model of mobile air compressor for mining is 5 kPa, and the final pressure difference of the coarse filter assembly is 20 kPa. When the current pressure difference before and after the coarse filter is 8 kPa, the saturation correction coefficient = (20-8) ÷ 20 = 0.6. Therefore, the second threshold = 5 kPa × 0.6 = 3 kPa. That is, under the current condition, if the average pressure difference decrease caused by one self-cleaning action is less than 3 kPa, the self-cleaning effect of the coarse filter is considered insufficient.

[0155] In summary, this step combines the pressure difference growth trend before and after fine filtration with the pressure difference recovery capability after coarse filtration self-cleaning to jointly determine the state of the coarse filtration self-cleaning device. Specifically, a first threshold is determined by dynamically statistically analyzing the rate of increase of the fine filtration pressure difference under historical normal operating conditions, and the historical effective recovery range is corrected by combining the current level of coarse filtration contamination to obtain a second threshold adapted to the current operating state. When the rate of increase of the fine filtration pressure difference exceeds the normal range, and the pressure difference recovery range generated by the coarse filtration self-cleaning action is lower than the allowable range for the current state, it indicates that dust has not been effectively intercepted and continues to enter the fine filtration component, and the coarse filtration self-cleaning function cannot restore filtration performance. Therefore, the coarse filtration self-cleaning device is determined to be in a failed state. Through the above-mentioned dynamic threshold and dual-parameter joint judgment method, the misjudgment caused by the influence of changes in operating conditions on a single pressure difference parameter can be reduced, and the accuracy of identifying the failure state of the coarse filtration self-cleaning device can be improved.

[0156] S400: When it is determined that the coarse filter self-cleaning device has failed, calculate the current effective dust holding capacity saturation of the coarse filter and plan the cleaning and maintenance sequence of the coarse filter according to the current effective dust holding capacity saturation. Step S400 in the method provided in this embodiment of the invention includes: Taking the current moment as the endpoint, a fixed time period is extracted to extract the pressure difference value before and after coarse filtration at each collection moment within the preset time window, and the slope of the change of the pressure difference value before and after coarse filtration with time is fitted as the rate of increase of the pressure difference of coarse filtration. Based on the relative degree between the current pressure difference before and after the coarse filter and the final pressure difference, combined with the factory-calibrated dust holding capacity, the current dust holding capacity of the coarse filter is determined, and the current effective dust holding saturation is obtained based on the comparison between the current dust holding capacity and the factory-calibrated dust holding capacity. Based on the fluctuation of the coarse filtration pressure differential rise rate in each adjacent sub-interval within a preset time window, the stability coefficient of the coarse filtration pressure differential rise rate is determined. The stability coefficient decreases as the dispersion of the rise rate in each sub-interval increases. Based on the correlation between the current effective dust holding capacity saturation and the cumulative running time, the running time required per unit saturation is deduced; Based on the difference between the termination pressure differential and the current pressure differential before and after, and combined with the rate of increase of the coarse filtration pressure differential, the remaining duration of the pressure differential estimation is calculated. The cleaning saturation threshold is determined based on the termination pressure difference of the coarse filter and the factory-calibrated dust holding capacity. The remaining time for saturation estimation is then calculated based on the difference between the cleaning saturation threshold and the current effective dust holding capacity, combined with the cumulative running time. Based on the stability coefficient, the contribution of the remaining time of differential pressure projection in the fusion is determined, and the remaining time of differential pressure projection and the remaining time of saturation projection are fused according to the contribution to obtain the coarse filter can continue to run. The current time plus the remaining time of coarse filter can be used as the coarse filter cleaning and maintenance time.

[0157] Furthermore, based on the judgment of the working status of the coarse filter self-cleaning device, this embodiment further predicts the subsequent sustainable operating time of the coarse filter according to the current filtration resistance change trend and dust holding state change law of the coarse filter component, so as to determine the reasonable cleaning and maintenance time.

[0158] Specifically, firstly, taking the current moment as the end point, a data interval of a preset time length is extracted as the remaining running time analysis window, and the pressure difference values ​​before and after coarse filtration corresponding to each sampling moment within the analysis time window are extracted to form a coarse filtration pressure difference change sequence in chronological order.

[0159] The pressure difference before and after the coarse filter is used to characterize the change in filtration resistance generated when air passes through the coarse filter assembly, and its trend can reflect the degree of dust accumulation inside the coarse filter assembly.

[0160] Furthermore, to obtain the current trend of coarse filtration resistance growth, this embodiment performs linear fitting on the coarse filtration pressure difference change sequence within the analysis time window, and uses the slope of the fitted line as the rate of increase of the coarse filtration pressure difference. Specifically, the rate of increase of the coarse filtration pressure difference is calculated as follows: The rate of increase of coarse filtration pressure difference = the slope of the fitted straight line.

[0161] The fitted straight line uses sampling time as the independent variable and the pressure difference before and after coarse filtration at the corresponding time as the dependent variable. The slope of the pressure difference changing with time is obtained using the least squares method. The specific calculation process is as follows: The rate of increase of coarse filter pressure difference = the sum of the products of sampling time deviation and pressure difference deviation ÷ the sum of the squares of sampling time deviation.

[0162] Among them, sampling time deviation represents the time difference between each sampling time and the average sampling time of the analysis time window, and pressure difference deviation represents the difference between the pressure difference value at each sampling time and the average pressure difference value.

[0163] For example, within a certain analysis time window, the pressure difference before and after coarse filtration gradually increases from 8 kPa to 10 kPa. After linear fitting, the pressure difference change trend is found to be 0.10 kPa / h. Therefore, the current coarse filtration pressure difference increase rate is determined to be 0.10 kPa / h.

[0164] Furthermore, to avoid the influence of the initial pressure difference level on the prediction of maintenance time based solely on the rate of change of pressure difference, this embodiment further calculates the effective dust holding saturation of the coarse filter by combining the current dust holding state of the coarse filter.

[0165] Specifically, the current dust holding capacity of the coarse filter is determined based on the proportional relationship between the current pressure difference before and after the coarse filter and the final pressure difference.

[0166] The termination pressure difference represents the maximum allowable filtration resistance when the coarse filter reaches the maintenance and cleaning conditions. It is determined based on the structural parameters of the coarse filter media, the manufacturer's rated performance parameters, and the equipment maintenance standards.

[0167] Specifically, the current dust holding capacity is calculated as follows: Current dust holding capacity = factory rated dust holding capacity × (current pressure difference before and after ÷ final pressure difference).

[0168] Among them, the factory-calibrated dust holding capacity represents the maximum mass of dust that the coarse filter material can hold under the condition of reaching the final differential pressure. It is obtained from the filter material performance parameters provided by the manufacturer or experimental calibration.

[0169] Furthermore, based on the ratio between the current dust holding capacity and the factory-specified dust holding capacity, the current effective dust holding saturation is determined. Specifically: Current effective dust holding capacity saturation = Current dust holding capacity ÷ Factory rated dust holding capacity.

[0170] Substituting the above relationships: Current effective dust saturation = Current pressure difference before and after ÷ Termination pressure difference.

[0171] For example, a coarse filter module has a factory-calibrated dust holding capacity of 500g and a final pressure difference of 20kPa. When the current pressure difference before and after the coarse filter is 8kPa, the current dust holding capacity = 500g × (8 ÷ 20) = 200g, and the current effective dust holding saturation = 200 ÷ 500 = 0.4. That is, the current coarse filter has reached 40% of its effective dust holding capacity.

[0172] Furthermore, to reduce the deviation in differential pressure prediction caused by dust impact and transient load changes, this embodiment introduces a stability coefficient based on the stability of the coarse filter differential pressure rise rate within the analysis time window.

[0173] Specifically, the analysis time window is divided into multiple equal-length sub-intervals, and the coarse filtration pressure difference rise rate corresponding to each sub-interval is calculated to obtain the sub-interval pressure difference rise rate set.

[0174] The stability coefficient characterizes the reliability of the current pressure differential growth trend. Specifically, the stability coefficient is calculated as follows: Stability coefficient = 1 - (standard deviation of rise rate of each sub-interval ÷ average rise rate of each sub-interval).

[0175] Among them, when the rate of increase in each sub-interval changes little, the standard deviation is small and the stability coefficient is close to 1, indicating that the current pressure difference growth trend is stable; when the rate of increase in each sub-interval fluctuates greatly, the standard deviation increases and the stability coefficient decreases, indicating that the current pressure difference growth trend is subject to strong disturbance.

[0176] For example, the 24-hour analysis time window is divided into four 6-hour sub-intervals, and the corresponding differential pressure rise rates are 0.08 kPa / h, 0.10 kPa / h, 0.09 kPa / h and 0.11 kPa / h, respectively. The average value is 0.095 kPa / h and the standard deviation is approximately 0.011 kPa / h. Then the stability coefficient is approximately 1 - (0.011 ÷ 0.095) ≈ 0.884.

[0177] Furthermore, in order to simultaneously consider the current rate of change of pressure difference and the rate of increase of dust holding capacity, this embodiment establishes a remaining time prediction result based on the pressure difference trend and a remaining time prediction result based on dust holding capacity saturation, respectively.

[0178] Specifically, based on the correspondence between the current effective dust saturation and the cumulative running time, the running time corresponding to a unit saturation is calculated. The running time per unit saturation is calculated as follows: Unit saturation running time = cumulative running time ÷ current effective dust saturation.

[0179] For example, when the cumulative operating time of the equipment is 1200h and the current effective dust holding capacity saturation is 0.4, the operating time per unit saturation is 1200 ÷ 0.4 = 3000h.

[0180] Furthermore, based on the difference between the termination pressure differential and the current pressure differential before and after coarse filtration, and combined with the current rate of increase of the coarse filtration pressure differential, the remaining operating time based on the pressure differential trend is calculated. Specifically: The remaining duration of the differential pressure calculation = (termination differential pressure - current differential pressure value before and after) ÷ rate of increase of the coarse filtration differential pressure.

[0181] For example, if the termination pressure difference is 20 kPa, the current pressure difference is 8 kPa, and the coarse filtration pressure difference rise rate is 0.10 kPa / h, then the remaining time for pressure difference projection = (20-8) ÷ 0.10 = 120 h.

[0182] Furthermore, in this embodiment, the cleaning saturation threshold is determined based on the dust holding state corresponding to the termination pressure difference.

[0183] Specifically, the cleaning saturation threshold is determined based on the dust holding ratio corresponding to the optimal cleaning cycle of the coarse filter in historical maintenance records.

[0184] For example, by analyzing data from 50 maintenance operations, it was found that when the effective dust holding capacity saturation of the coarse filter reaches 0.85, continued operation will lead to a rapid increase in pressure differential. Therefore, the cleaning saturation threshold is set to 0.85.

[0185] Furthermore, based on the difference between the cleaning saturation threshold and the current effective dust holding saturation, and combined with the unit saturation running time, the remaining time for saturation estimation is calculated. Specifically: Saturation estimation remaining time = (cleaning saturation threshold - current effective dust holding saturation) × unit saturation running time.

[0186] For example, if the current effective dust holding saturation is 0.4, the cleaning saturation threshold is 0.85, and the unit saturation running time is 3000h, then the remaining time for saturation estimation = (0.85-0.4)×3000 = 1350h.

[0187] Furthermore, to improve the adaptability of the prediction results to changes in the current operating status, this embodiment fuses the two remaining operating time prediction results based on the stability coefficient.

[0188] Specifically, the stability coefficient is used to determine the contribution ratio of the remaining time of the pressure difference simulation.

[0189] Among them, the contribution weight of differential pressure prediction is equal to the stability coefficient.

[0190] Dust concentration prediction contribution weight = 1 - stability coefficient.

[0191] Therefore, the remaining runtime of the coarse filter is calculated as follows: The remaining operating time of the coarse filter = stability coefficient × remaining time estimated by pressure difference + (1 - stability coefficient) × remaining time estimated by saturation.

[0192] For example, when the stability coefficient is 0.884, the remaining time for differential pressure estimation is 120 hours, and the remaining time for saturation estimation is 1350 hours, the remaining operating time of the coarse filter is approximately 0.884 × 120 + (1 - 0.884) × 1350 ≈ 262.7 hours. Therefore, it is predicted that the current coarse filter assembly can still operate for approximately 263 hours.

[0193] Furthermore, the current time plus the time the coarse filter can continue to run is taken as the time for coarse filter cleaning and maintenance, and output to the equipment maintenance management module for advance scheduling of coarse filter component cleaning or replacement operations.

[0194] In summary, this step, based on the status monitoring of the coarse filter self-cleaning device, further predicts the remaining operating time of the coarse filter assembly by combining the coarse filter pressure difference change trend, the current dust saturation level, and the stability of pressure difference growth. The pressure difference rise rate is obtained by fitting the pressure difference change trend before and after coarse filtration, and the current effective dust saturation is estimated by combining the termination pressure difference. Remaining time prediction models based on pressure difference trend and dust saturation are established respectively. Simultaneously, the fusion weights of the two prediction results are dynamically adjusted according to the stability of pressure difference change to obtain a coarse filter operating time that better reflects the current operating state. Based on this, the cleaning and maintenance timing is determined, avoiding resource waste or over-operation caused by fixed-cycle maintenance and improving equipment reliability.

[0195] S500: Predicts the remaining dust holding capacity of the fine filter, and determines the early warning time for fine filter replacement based on the remaining dust holding capacity and the current effective dust saturation, and outputs an operation and maintenance management plan.

[0196] Step S500 in the method provided in this embodiment of the invention includes: The current effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after the fine filter, and the cumulative running time are used as inputs to the dust holding capacity prediction model of the fine filter, and the remaining dust holding capacity of the fine filter is obtained as the output. The steps for constructing the dust holding capacity prediction model for fine filters include: Collect historical dust holding data of the same type of fine filter under different operating conditions. For each historical record, obtain the corresponding effective dust holding saturation of the coarse filter, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time, as input feature triplet. The remaining dust holding capacity of the fine filter at the corresponding time is obtained by offline weighing and used as the output label to construct a sample pair set; A three-input single-output mapping network is constructed with the effective dust holding capacity saturation of coarse filtration, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time as inputs, and the remaining dust holding capacity of fine filtration as output. The input layer has 3 nodes, the hidden layer uses a nonlinear activation function, and the output layer has 1 node. The output value represents the remaining dust holding capacity of fine filtration corresponding to the current combination of input features. The sample set is divided into a fitting set and a validation set, and the mean square error between the output layer prediction value and the remaining dust capacity measured by the offline weighing method is used as the loss function. The network weights of the mapping network are updated by backpropagation of error. The process is iteratively fitted until the error of the loss function on the validation set no longer decreases. The fitted mapping network is then used as the prediction model for the dust holding capacity of the fine filter.

[0197] Furthermore, based on the status monitoring of the coarse filter self-cleaning device and the prediction of the remaining operating time of the coarse filter component, this embodiment further addresses the problem of the reduced filtration capacity of the fine filter component due to dust accumulation by constructing a dust holding capacity prediction model for the fine filter. By integrating the current effective dust holding status of the coarse filter, the growth trend of the fine filter pressure difference, and the cumulative operating time of the equipment, the current remaining dust holding capacity of the fine filter is predicted, so as to achieve advance assessment of the fine filter maintenance cycle.

[0198] Specifically, the effective dust holding capacity saturation of the coarse filter at the current moment, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time are used as input features for the dust holding capacity prediction model of the fine filter.

[0199] Among them, the effective dust holding capacity saturation of the coarse filter is used to characterize the filtration status of the dust load in the air before entering the fine filter component; the rate of increase of the pressure difference before and after the fine filter is used to characterize the speed at which the filtration resistance increases due to the accumulation of dust inside the fine filter component; and the cumulative running time is used to reflect the basic pollution level of the filter material during long-term operation.

[0200] Furthermore, the above three operating parameters are combined into an input feature triplet, specifically represented as follows: Input feature triplet = (coarse filter effective dust holding capacity saturation, pressure difference rise rate before and after fine filtration, cumulative running time).

[0201] Among them, the effective dust holding capacity saturation of the coarse filter is a dimensionless parameter, the rate of increase of the pressure difference before and after the fine filter is in kPa / h, and the cumulative running time is in h.

[0202] Furthermore, before constructing a prediction model for the dust holding capacity of fine filters, historical operating data of the same model of mobile air compressors used in mines under different operating conditions were first collected.

[0203] Specifically, during the actual operation of the fine filter component, the operating status of the fine filter is recorded according to the preset sampling period, and when the equipment is shut down for maintenance or the fine filter component is replaced, the actual remaining dust holding capacity of the fine filter component at the corresponding moment is obtained by offline weighing.

[0204] The offline weighing method specifically involves: disassembling the fine filter component to be tested, removing non-fixed dust adhering to the surface of the filter material, weighing the fine filter component, and obtaining the dust mass that the fine filter component has adsorbed based on the difference between the mass of the fine filter component in the clean state and the current tested mass; then, combining this with the maximum dust holding capacity calibrated for the fine filter component, calculating the corresponding remaining dust holding capacity.

[0205] Specifically, the remaining dust holding capacity is calculated as follows: Remaining dust holding capacity = factory rated dust holding capacity - current mass of adsorbed dust.

[0206] For example, a certain model of fine filter module has a factory-calibrated maximum dust holding capacity of 1000g. When it is disassembled for testing, the mass of dust that has been adsorbed is found to be 650g. Therefore, the remaining dust holding capacity of the fine filter at that moment is 1000g - 650g = 350g.

[0207] Furthermore, for each historical detection record, the effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time corresponding to that detection moment are extracted as historical input features; at the same time, the remaining dust holding capacity of the fine filter obtained by the offline weighing method is used as the corresponding output label.

[0208] This leads to the construction of a set of historical sample pairs: Sample pair set = {(input feature triplet, remaining dust capacity)}.

[0209] For example, in a certain historical operation record, the effective dust holding capacity saturation of the coarse filter was detected to be 0.45, the rate of increase of the pressure difference before and after the fine filter was 0.12 kPa / h, the cumulative running time was 800h, and the corresponding offline weighing obtained the remaining dust holding capacity of the fine filter was 420g. Thus, a set of sample data (0.45, 0.12 kPa / h, 800h) → 420g is formed.

[0210] Furthermore, based on the aforementioned set of historical sample pairs, a three-input single-output mapping network is constructed as a model for predicting the dust holding capacity of fine filters.

[0211] Specifically, the mapping network includes an input layer, a hidden layer, and an output layer.

[0212] The input layer has three input nodes, which are used to receive three input features: the effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after the fine filter, and the cumulative running time. The hidden layer is used to perform feature mapping on the nonlinear relationship between the input features. The output layer has one output node, which is used to output the predicted value of the remaining dust holding capacity of the fine filter in the current state.

[0213] Furthermore, the hidden layer employs a nonlinear activation function to enhance the model's ability to fit complex dust accumulation processes.

[0214] Specifically, the output of the j-th node in the hidden layer is calculated as follows: Hidden layer output value = activation function (weighted sum of input features + bias).

[0215] Among them, the input feature weighted sum represents the cumulative result after multiplying the three input features by their corresponding network weights; the network weights represent the degree of influence of different input factors on the prediction result of the remaining dust holding capacity of the fine filter; and the bias is used to adjust the network output benchmark.

[0216] The output layer prediction results are calculated as follows: Predicted remaining dust holding capacity of fine filter = Output layer weight × sum of hidden layer output values ​​+ output layer bias.

[0217] Furthermore, in order to adapt the prediction model to the actual operating patterns of the equipment, this embodiment uses historical sample data to train the network parameters.

[0218] Specifically, the historical sample set is divided into a fitting set and a validation set according to a preset ratio.

[0219] The fitted set is used to update the network weight parameters, and the validation set is used to evaluate the model's generalization ability.

[0220] For example, 1000 sets of historical running samples are divided in an 8:2 ratio, with 800 sets as the fitting set and 200 sets as the validation set.

[0221] Furthermore, the mean square error between the predicted remaining dust holding capacity of the fine filter output by the model and the actual remaining dust holding capacity obtained by offline weighing is used as the training loss function. Specifically, the loss function is calculated as follows: Loss function = Sum of squares of the difference between predicted remaining dust capacity and actual remaining dust capacity ÷ Sample size.

[0222] The loss function is used to evaluate the degree of deviation between the model's prediction results and the actual detection results. The smaller the loss function, the higher the model's prediction accuracy.

[0223] Furthermore, the backpropagation algorithm is used to iteratively update the network weights and biases.

[0224] Specifically, based on the partial derivative of the loss function with respect to each network parameter, the adjustment direction corresponding to each network parameter is calculated, and the network weights are updated according to the preset learning step size.

[0225] The learning step size is used to control the magnitude of each parameter update.

[0226] Furthermore, the learning step size is determined based on the convergence speed during the historical training process.

[0227] For example, when training comparisons were performed using different learning step sizes in the range of 0.001 to 0.01, it was found that when the learning step size was 0.005, the error of the model on the validation set decreased faster and there was no obvious oscillation. Therefore, the learning step size was set to 0.005.

[0228] Furthermore, the network training process is continuously executed. When the mean square error on the validation set no longer decreases for several consecutive training cycles, the training is stopped, and the mapping network corresponding to the current network parameters is used as the prediction model for the dust holding capacity of the fine filter.

[0229] The stopping conditions are determined based on historical model training experience.

[0230] For example, when the mean squared error of the validation set decreases by less than 0.1% over 50 consecutive training cycles, the model is considered to have reached a stable state, and network parameter updates are stopped.

[0231] Furthermore, during actual operation, the currently collected effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time are input into the trained fine filtration dust holding capacity prediction model. The model outputs the corresponding remaining dust holding capacity of the fine filter based on the learned historical operating patterns.

[0232] For example, under the current operating conditions, the effective dust holding capacity saturation of the coarse filter is detected to be 0.60, the rate of increase of the pressure difference before and after the fine filter is 0.15 kPa / h, and the cumulative running time is 1500h. After inputting the above parameters into the fine filter dust holding capacity prediction model, the predicted value of the remaining dust holding capacity of the fine filter is 220g.

[0233] At this point, the equipment maintenance management module can schedule fine filter cleaning or replacement operations in advance based on the predicted remaining dust holding capacity and the set maintenance threshold.

[0234] Step S500 in the method provided in this embodiment of the invention further includes: Based on the rate of increase of the pressure difference in the coarse filtration and the rate of increase of the pressure difference before and after the fine filtration, the correction coefficient for the dust consumption rate of the fine filtration is determined. Based on the rate of increase of the pressure difference before and after fine filtration and the correction coefficient of the dust consumption rate, the dust consumption of fine filtration per unit time is determined, and combined with the remaining dust capacity of fine filtration, the estimated remaining usable time of fine filtration is determined. Add the estimated remaining time of the fine filter to the current time to obtain the estimated replacement time of the fine filter; Obtain the coarse filter cleaning and maintenance time. When the estimated replacement time of the fine filter is earlier than the coarse filter cleaning and maintenance time, use the estimated replacement time of the fine filter as the fine filter replacement warning time. When the estimated replacement time of the fine filter is later than or equal to the cleaning and maintenance time of the coarse filter, the cleaning and maintenance time of the coarse filter will be used as the early warning time for the replacement of the fine filter. By linking the early warning time for fine filter replacement with the time for coarse filter cleaning and maintenance, a joint shutdown maintenance window is generated and output as an operation and maintenance management plan.

[0235] Furthermore, based on the prediction of the coarse filter cleaning and maintenance time and the prediction of the remaining dust holding capacity of the fine filter, this embodiment further considers the impact of the coarse filter maintenance process on the fine filter operation status. By integrating the coarse filter pressure difference growth trend and the fine filter pressure difference growth trend, the remaining available time of the fine filter is dynamically corrected, and a joint shutdown maintenance window is generated in conjunction with the coarse filter cleaning and maintenance plan to avoid repeated shutdowns caused by separate maintenance of the coarse filter and the fine filter, thereby improving equipment maintenance efficiency.

[0236] Specifically, firstly, based on the relative relationship between the rate of increase of the coarse filtration pressure difference and the rate of increase of the pressure difference before and after fine filtration, the correction coefficient for the dust holding capacity consumption rate of fine filtration is determined.

[0237] The rate of increase of the coarse filter pressure differential is used to characterize the current dust accumulation rate of the coarse filter module, while the rate of increase of the pressure differential before and after the fine filter is used to characterize the rate of increase of the filtration resistance of the fine filter module. When the coarse filter pressure differential increases rapidly, it indicates that the coarse filter's dust interception capacity decreases, the dust load entering the fine filter module increases, and the dust consumption rate of the fine filter accelerates. When the coarse filter pressure differential increases at a low rate, it indicates that the coarse filter still has a good dust interception capacity, and the dust load on the fine filter is low.

[0238] Furthermore, a correction coefficient for the dust consumption rate of the fine filter is constructed. Specifically, the correction coefficient for the dust consumption rate of the fine filter is calculated as follows: Correction factor for the dust holding capacity consumption rate of fine filter = 1 + coarse filter pressure difference rise rate ÷ fine filter pressure difference rise rate.

[0239] The correction factor is used to represent the degree of amplification of the coarse filter contamination state on the rate of dust consumption in the fine filter.

[0240] When the rate of increase of the coarse filter pressure difference is small, and the increase of the fine filter pressure difference is mainly caused by its own cumulative operation, the correction coefficient is close to 1; when the rate of increase of the coarse filter pressure difference increases, the correction coefficient increases, indicating that the dust load entering the fine filter increases, and the estimated value of the dust consumption rate of the fine filter needs to be increased.

[0241] Furthermore, to avoid abnormal changes in the correction coefficient caused by a single pressure difference fluctuation, this embodiment sets a reasonable range limit for the correction coefficient.

[0242] The upper and lower limits of the correction coefficient are obtained by calibration based on historical operating data of the same type of equipment.

[0243] Specifically, the coarse filter pressure difference growth and fine filter dust consumption changes during multiple maintenance cycles in the normal operation phase of the equipment are collected. The corresponding correction coefficient ranges under different operating conditions are statistically analyzed, and the data range covering 95% of the normal operation conditions is selected as the correction coefficient value range.

[0244] For example, by collecting and analyzing data from 50 maintenance cycles, it was found that the correction coefficient for the fine filter dust consumption rate mainly falls within the range of 1.0 to 2.5. Therefore, the lower limit of the fine filter dust consumption rate correction coefficient is set to 1.0, and the upper limit is set to 2.5. When the calculated correction coefficient exceeds the above range, it is restricted to the corresponding boundary range.

[0245] Furthermore, the amount of dust consumed per unit time is determined based on the rate of increase of the pressure difference before and after fine filtration and the correction coefficient for the rate of dust consumption in the fine filtration.

[0246] Specifically, the relationship between the rate of increase in fine filtration pressure differential and the rate of dust consumption is first established based on historical operating data. The basic dust consumption rate is determined as follows: Basic dust consumption rate = Change in fine filter dust capacity ÷ Corresponding operating time.

[0247] Furthermore, the rate of increase in pressure difference before and after fine filtration is used as a parameter characterizing the current pollution growth state to correct the basic dust consumption rate. Specifically: Current fine filter dust consumption rate = basic dust consumption rate × fine filter dust consumption rate correction factor.

[0248] The current fine filter dust consumption rate represents the amount of remaining fine filter dust capacity consumed per unit time under the current operating conditions.

[0249] For example, based on historical maintenance data, the basic dust consumption rate of a certain type of fine filter module is 2g / h; the current calculated dust consumption rate correction factor is 1.5, then the current dust consumption rate of the fine filter = 2g / h × 1.5 = 3g / h. This means that under the current operating condition, the fine filter consumes approximately 3g of remaining dust capacity every hour of operation.

[0250] Furthermore, based on the current remaining dust holding capacity of the fine filter and the current dust consumption rate of the fine filter, the estimated remaining usable time of the fine filter is calculated. Specifically: Estimated remaining usable time for fine filter = remaining dust holding capacity of fine filter ÷ current dust consumption rate of fine filter.

[0251] The remaining dust holding capacity of the fine filter is obtained from the output of the aforementioned fine filter dust holding capacity prediction model.

[0252] For example, if the current predicted remaining dust holding capacity of the fine filter is 300g and the current dust consumption rate of the fine filter is 3g / h, then the estimated remaining usable time of the fine filter = 300 ÷ 3 = 100h.

[0253] Furthermore, using the current time as the time base, the current time is added to the estimated remaining available time for fine filtration to obtain the estimated replacement time for fine filtration. Specifically: Estimated replacement time for fine filter = current time + estimated remaining usable time for fine filter.

[0254] For example, if the current time is 8:00 on August 3, 2026, and the estimated remaining available time for the filter is 100 hours, then the estimated replacement time for the filter is 12:00 on August 7, 2026.

[0255] Furthermore, to reduce repeated downtime caused by separate maintenance of the coarse filter and the fine filter, this embodiment further obtains the coarse filter cleaning and maintenance time calculated in the aforementioned steps, and generates a joint maintenance plan based on the order of maintenance times for the coarse filter and the fine filter.

[0256] Specifically, the estimated replacement time of the fine filter is compared with the cleaning and maintenance time of the coarse filter.

[0257] If the estimated replacement time of the fine filter is less than the cleaning and maintenance time of the coarse filter, it means that the fine filter will reach the replacement condition before the coarse filter maintenance. In this case, the estimated replacement time of the fine filter will be used as the early warning time for the fine filter replacement.

[0258] Among them, the fine filter replacement early warning is used to remind maintenance personnel in advance to prepare for the replacement of the fine filter components.

[0259] For example, if the estimated replacement time for the fine filter is 12:00 on August 7, 2026, and the estimated time for the coarse filter cleaning and maintenance is 8:00 on August 10, 2026, then 12:00 on August 7, 2026 < 8:00 on August 10, 2026. Therefore, 12:00 on August 7, 2026 will be used as the early warning time for the fine filter replacement.

[0260] Furthermore, if the estimated replacement time of the fine filter is greater than or equal to the cleaning and maintenance time of the coarse filter, it means that the coarse filter will reach the maintenance condition before the fine filter. In this case, there is no need to arrange fine filter maintenance separately in advance, but the cleaning and maintenance time of the coarse filter is used as the early warning reference time for fine filter replacement.

[0261] Specifically: Fine filter replacement warning time = Coarse filter cleaning and maintenance time.

[0262] For example, if the estimated replacement time for the fine filter is 10:00 AM on August 15, 2026, while the estimated time for the coarse filter cleaning and maintenance is 8:00 AM on August 10, 2026, then 10:00 AM on August 15, 2026 ≥ 8:00 AM on August 10, 2026. Therefore, 8:00 AM on August 10, 2026, is used as the early warning time for fine filter replacement, so that fine filter maintenance and coarse filter cleaning are scheduled simultaneously.

[0263] Furthermore, the early warning time for fine filter replacement is time-bound with the time for coarse filter cleaning and maintenance, forming a joint shutdown maintenance window. Specifically, the joint shutdown maintenance window is determined as follows: The combined shutdown maintenance window is calculated from min (estimated replacement time of fine filter and cleaning and maintenance time of coarse filter) to max (estimated replacement time of fine filter and cleaning and maintenance time of coarse filter).

[0264] The joint shutdown maintenance window indicates the time range within which coarse filter cleaning and fine filter replacement operations need to be coordinated.

[0265] For example, if the fine filter replacement warning time is 12:00 on August 7, 2026, and the coarse filter cleaning and maintenance time is 8:00 on August 10, 2026, then a joint shutdown maintenance window will be generated from 12:00 on August 7, 2026 to 8:00 on August 10, 2026. The equipment maintenance management module can arrange the shutdown plan in advance according to this time range, and simultaneously complete the coarse filter cleaning and fine filter replacement during the shutdown period.

[0266] In summary, this step, based on the prediction of the remaining dust holding capacity of the fine filter, further integrates the growth trend of the coarse filter differential pressure and the variation characteristics of the fine filter differential pressure to dynamically correct the dust consumption rate of the fine filter and predict the remaining usable time of the fine filter. By combining the estimated replacement time of the fine filter with the cleaning and maintenance time of the coarse filter, an early warning time for fine filter replacement is determined, and a joint shutdown maintenance window is generated. This enables coordinated planning of coarse filter cleaning and fine filter replacement, reduces repeated shutdowns, and improves the maintenance efficiency and operational reliability of the filtration system.

[0267] Example 2, as Figure 3 As shown, this invention provides an intelligent operation and maintenance management planning system for air compressor units, comprising: The operation data acquisition module 11 is used to acquire real-time operation data of the multi-stage air intake filtration device of the mobile air compressor in the mine, wherein the multi-stage air intake filtration device includes coarse filter and fine filter; The self-cleaning status analysis module 12 is used to determine the working status of the coarse filter self-cleaning device based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with real-time operating data. The coarse filter maintenance planning module 13 is used to determine that the coarse filter self-cleaning device has failed when the rate of increase of the pressure difference before and after the fine filter exceeds the first threshold and the rate of decrease of the pressure difference before and after the coarse filter does not reach the second threshold within a preset time window. The fine filter capacity prediction module 14 is used to calculate the current effective dust holding capacity saturation of the coarse filter when it is determined that the coarse filter self-cleaning device has failed, and to plan the cleaning and maintenance sequence of the coarse filter based on the current effective dust holding capacity saturation. The joint operation and maintenance planning module 15 is used to predict the remaining dust holding capacity of the fine filter, and determine the early warning time for the fine filter replacement based on the remaining dust holding capacity and the current effective dust holding saturation, and output the operation and maintenance management plan.

[0268] In this embodiment of the invention, the data acquisition module 11 is further configured to: Collect the upstream and downstream pressures of the coarse filter and calculate the pressure difference before and after the coarse filter. Collect the upstream and downstream pressures of the fine filter and calculate the pressure difference before and after the fine filter. Record the cumulative number of actions of the coarse filter self-cleaning device since the last coarse filter cleaning and reset, as well as the time of each action. Read the factory-calibrated dust holding capacity and the termination pressure difference of the coarse filter. The factory-calibrated dust holding capacity is the total mass of test dust that the coarse filter can hold when it reaches the termination pressure difference under standard test conditions. The termination pressure difference is used to calculate the current effective dust holding capacity saturation of the coarse filter by combining the current pressure difference value before and after the coarse filter. Record the cumulative runtime since the last coarse filter cleaning and reset.

[0269] In this embodiment of the invention, the self-cleaning status analysis module 12 is further used for: A fixed time length is extracted forward from the current time to construct an analysis time window. The pressure difference values ​​before and after coarse filtration corresponding to each collection time within the analysis time window are extracted to form a coarse filtration pressure difference time series. Traverse the coarse filtration pressure difference time series, identify and analyze each local maximum point and the subsequent local minimum point adjacent to each local maximum point within the analysis time window, subtract the coarse filtration pressure difference value corresponding to the local minimum point from the coarse filtration pressure difference value corresponding to the local maximum point, and use it as the pressure difference recovery amount. Record the time corresponding to the local maximum point as the recovery start time and the time corresponding to the local minimum point as the recovery end time. Match the recovery start time with the action time of the coarse filter self-cleaning device. When the time deviation between the recovery start time and the action time is less than the allowable deviation, mark the corresponding differential pressure recovery amount as the effective recovery amount. The number of effective recovery amounts is counted, and the arithmetic mean of all effective recovery amounts is calculated as the average recovery rate. When the number of effective recovery quantities is less than the product of the total number of action moments within the analysis time window and the proportional coefficient, it is determined that the self-cleaning device has missing actions. When the average recovery amplitude is less than the product of the average historical effective recovery amplitude and the attenuation coefficient, it is determined that the self-cleaning device has an efficiency decay. The average historical effective recovery amplitude is determined based on the arithmetic mean of the average recovery amplitudes corresponding to multiple analysis time windows before the current time. When it is determined that there is a missing action or a decrease in execution efficiency, the working status of the self-cleaning device will be marked as an abnormal working status.

[0270] Among them, when the number of effective recovery quantities is less than the product of the total number of action moments within the analysis time window and the proportional coefficient, it is determined that the self-cleaning device has missing actions, which also includes: The difference between the total number of self-cleaning device action moments and the number of effective recovery moments within the analysis time window is used as the invalid action count; The missing action rate is obtained by dividing the number of invalid actions by the total number of action moments. When the missing action rate exceeds the missing action rate tolerance threshold, the self-cleaning device is marked as having failed.

[0271] In this embodiment of the invention, the coarse filter maintenance planning module 13 is further used for: Within a preset time window, a sampling sequence of pressure difference values ​​before and after fine filtration is extracted. Based on the difference between the pressure difference values ​​before and after fine filtration corresponding to the start and end times of the preset time window, combined with the duration of the preset time window, the rising rate of the pressure difference before and after fine filtration is used. Acquire all action moments of the coarse filter self-cleaning device within a preset time window, extract the pressure difference value before and after the coarse filter corresponding to each action moment as the pressure difference value before each cleaning, and extract the pressure difference value before and after the sampling moment after each action moment when the pressure difference drops back to the stable platform area as the pressure difference value after each cleaning. Calculate the difference between the pressure difference before each cleaning and the corresponding pressure difference after each cleaning to obtain the single drop pressure difference for each cleaning. Take the arithmetic mean of all single drop pressure differences as the decrease in pressure difference before and after coarse filtration. When the rate of increase of the pressure difference before and after the fine filter exceeds the first threshold, and the rate of decrease of the pressure difference before and after the coarse filter is less than the second threshold, the coarse filter self-cleaning device is determined to have failed.

[0272] The steps for determining the first threshold and the second threshold include: The rising rate of the pressure difference before and after fine filtration for each of the multiple historical analysis time windows before the current moment is obtained, forming a set of historical rising rates. After removing the maximum and minimum values ​​in the set of historical rising rates, the upper quartile of the remaining values ​​is taken as the first threshold. The saturation correction factor is calculated based on the ratio of the current pressure difference before and after the coarse filter to the final pressure difference. The saturation correction factor is directly proportional to the difference between the final pressure difference and the current pressure difference before and after the coarse filter, and inversely proportional to the final pressure difference. The second threshold is obtained by multiplying the average historical effective recovery magnitude by the saturation correction factor.

[0273] In this embodiment of the invention, the fine filtration capacity prediction module 14 is further configured to: Taking the current moment as the endpoint, a fixed time period is extracted forward. The pressure difference value before and after coarse filtration is extracted at each collection moment within the preset time window. The slope of the change of the pressure difference value before and after coarse filtration with time is fitted as the rate of increase of the coarse filtration pressure difference. Based on the relative degree between the current pressure difference before and after the coarse filter and the final pressure difference, and in conjunction with the factory-calibrated dust holding capacity, the current dust holding capacity of the coarse filter is determined, and the current effective dust holding saturation is obtained based on the comparison between the current dust holding capacity and the factory-calibrated dust holding capacity. Based on the fluctuation of the coarse filtration differential pressure rise rate in each adjacent sub-interval within a preset time window, the stability coefficient of the coarse filtration differential pressure rise rate is determined. The stability coefficient decreases as the dispersion of the rise rate in each sub-interval increases. Based on the correlation between the current effective dust holding capacity saturation and the cumulative running time, the running time required per unit saturation is deduced; Based on the difference between the termination pressure differential and the current pressure differential before and after, and combined with the rate of increase of the coarse filtration pressure differential, the remaining duration of the pressure differential estimation is calculated. The cleaning saturation threshold is determined based on the termination pressure difference of the coarse filter and the factory-calibrated dust holding capacity. The remaining time for saturation estimation is then calculated based on the difference between the cleaning saturation threshold and the current effective dust holding capacity, combined with the cumulative running time. Based on the stability coefficient, the contribution of the remaining time of differential pressure projection in the fusion is determined, and the remaining time of differential pressure projection and the remaining time of saturation projection are fused according to the contribution to obtain the coarse filter can continue to run. The current time plus the remaining time of coarse filter can be used as the coarse filter cleaning and maintenance time.

[0274] In this embodiment of the invention, the joint operation and maintenance planning module 15 is further configured to: The current effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after the fine filter, and the cumulative running time are used as inputs to the dust holding capacity prediction model of the fine filter, and the remaining dust holding capacity of the fine filter is obtained as the output. The steps for constructing the dust holding capacity prediction model for fine filters include: Collect historical dust holding data of the same type of fine filter under different operating conditions. For each historical record, obtain the corresponding effective dust holding saturation of the coarse filter, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time, as input feature triplet. The remaining dust holding capacity of the fine filter at the corresponding time is obtained by offline weighing and used as the output label to construct a sample pair set; A three-input single-output mapping network is constructed with the effective dust holding capacity saturation of coarse filtration, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time as inputs, and the remaining dust holding capacity of fine filtration as output. The input layer has 3 nodes, the hidden layer uses a nonlinear activation function, and the output layer has 1 node. The output value represents the remaining dust holding capacity of fine filtration corresponding to the current combination of input features. The sample set is divided into a fitting set and a validation set, and the mean square error between the output layer prediction value and the remaining dust capacity measured by the offline weighing method is used as the loss function. The network weights of the mapping network are updated by backpropagation of error. The process is iteratively fitted until the error of the loss function on the validation set no longer decreases. The fitted mapping network is then used as the prediction model for the dust holding capacity of the fine filter.

[0275] Based on the remaining dust holding capacity and current effective dust saturation of the fine filter, the system determines the early warning time for fine filter replacement and outputs an operation and maintenance management plan, including: Based on the rate of increase of the pressure difference in the coarse filtration and the rate of increase of the pressure difference before and after the fine filtration, the correction coefficient for the dust consumption rate of the fine filtration is determined. Based on the rate of increase of the pressure difference before and after fine filtration and the correction coefficient of the dust consumption rate, the dust consumption of fine filtration per unit time is determined, and combined with the remaining dust capacity of fine filtration, the estimated remaining usable time of fine filtration is determined. Add the estimated remaining time of the fine filter to the current time to obtain the estimated replacement time of the fine filter; Obtain the coarse filter cleaning and maintenance time. When the estimated replacement time of the fine filter is earlier than the coarse filter cleaning and maintenance time, use the estimated replacement time of the fine filter as the fine filter replacement warning time. When the estimated replacement time of the fine filter is later than or equal to the cleaning and maintenance time of the coarse filter, the cleaning and maintenance time of the coarse filter will be used as the early warning time for the replacement of the fine filter. By linking the early warning time for fine filter replacement with the time for coarse filter cleaning and maintenance, a joint shutdown maintenance window is generated and output as an operation and maintenance management plan.

[0276] In summary, the intelligent operation and maintenance management planning system for air compressor units provided in this embodiment of the invention collects operating status parameters and self-cleaning action information of the coarse and fine filters through the operation data acquisition module, and comprehensively judges the effectiveness and cleaning recovery capability of the coarse filter self-cleaning device using the self-cleaning status analysis module. Furthermore, the coarse filter maintenance planning module combines the fine filter differential pressure growth trend with the coarse filter differential pressure recovery capability to achieve coarse filter failure identification and cleaning / maintenance timing prediction. Simultaneously, the fine filter capacity prediction module constructs a fine filter dust holding capacity prediction model based on historical operating data to predict the remaining dust holding capacity of the fine filter, and dynamically corrects the fine filter consumption rate based on the coarse filter contamination status. Finally, the joint operation and maintenance planning module generates a collaborative maintenance plan for coarse filter cleaning and fine filter replacement. This invention, through multi-level filtration status correlation analysis, realizes the prediction of the remaining lifespan of the filtration system and optimizes maintenance timing, reducing resource waste caused by manual periodic maintenance and improving the operational reliability and intelligent operation and maintenance level of mobile air compressors in mines.

[0277] It should be noted that the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0278] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0279] This specification and accompanying drawings are merely illustrative examples of the invention and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its scope. Therefore, if such modifications and modifications fall within the scope of the invention and its equivalents, the invention is intended to include these modifications and modifications.

Claims

1. A method for intelligent operation and maintenance management planning of air compressor units, characterized in that, The method includes: Real-time operating data of a multi-stage air intake filtration device for a mobile air compressor in a mine is obtained, wherein the multi-stage air intake filtration device includes a coarse filter and a fine filter; Based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with the real-time operating data, the working status of the coarse filter self-cleaning device is determined. When the rate of increase of the pressure difference before and after the fine filter exceeds the first threshold and the rate of decrease of the pressure difference before and after the coarse filter does not reach the second threshold within a preset time window, the coarse filter self-cleaning device is determined to be faulty. When it is determined that the coarse filter self-cleaning device has failed, the current effective dust holding capacity saturation of the coarse filter is calculated, and the cleaning and maintenance sequence of the coarse filter is planned according to the current effective dust holding capacity saturation. Predict the remaining dust holding capacity of the fine filter, and determine the early warning time for replacing the fine filter based on the remaining dust holding capacity of the fine filter and the current effective dust holding saturation, and output the operation and maintenance management plan.

2. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, Acquire real-time operating data of the multi-stage air intake filtration device of the mobile air compressor in the mine, including: Collect the upstream and downstream pressures of the coarse filter and calculate the pressure difference before and after the coarse filter. Collect the upstream and downstream pressures of the fine filter and calculate the pressure difference before and after the fine filter. Record the cumulative number of actions of the coarse filter self-cleaning device since the last coarse filter cleaning and reset, as well as the time of each action. Read the factory-calibrated dust holding capacity and the termination pressure difference of the coarse filter. The factory-calibrated dust holding capacity is the total mass of test dust that the coarse filter can hold when it reaches the termination pressure difference under standard test conditions. The termination pressure difference is used to calculate the current effective dust holding capacity saturation of the coarse filter by combining the current pressure difference value before and after the coarse filter. Record the cumulative runtime since the last coarse filter cleaning and reset.

3. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, Based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, and in conjunction with the real-time operating data, the working status of the coarse filter self-cleaning device is determined, including: A fixed time length is extracted forward from the current time to construct an analysis time window. The pressure difference values ​​before and after coarse filtration corresponding to each collection time within the analysis time window are extracted to form a coarse filtration pressure difference time series. Traverse the coarse filtration pressure difference time series, identify each local maximum point and the subsequent local minimum point adjacent to each local maximum point within the analysis time window, subtract the coarse filtration pressure difference value corresponding to the local minimum point from the coarse filtration pressure difference value corresponding to the local maximum point, and use this as the pressure difference recovery amount. Record the time corresponding to the local maximum point as the recovery start time and the time corresponding to the local minimum point as the recovery end time. Match the recovery start time with the operation time of the coarse filter self-cleaning device. When the time deviation between the recovery start time and the operation time is less than the allowable deviation, mark the corresponding differential pressure recovery amount as the effective recovery amount. The number of effective recovery amounts is counted, and the arithmetic mean of all effective recovery amounts is calculated as the average recovery magnitude. When the number of effective recovery quantities is less than the product of the total number of action moments within the analysis time window and the proportional coefficient, it is determined that the self-cleaning device has missing actions. When the average recovery amplitude is less than the product of the average historical effective recovery amplitude and the attenuation coefficient, it is determined that the self-cleaning device has an efficiency decay. The average historical effective recovery amplitude is determined based on the arithmetic mean of the average recovery amplitudes corresponding to multiple analysis time windows before the current time. When it is determined that the action is missing or the execution efficiency is reduced, the working state of the self-cleaning device is marked as an abnormal working state.

4. The intelligent operation and maintenance management planning method for air compressor units according to claim 3, characterized in that, When the number of effective recovery quantities is less than the product of the total number of action moments within the analysis time window and the proportional coefficient, it is determined that the self-cleaning device has a missing action, further including: The difference between the total number of times the self-cleaning device operates within the analysis time window and the number of effective recovery times is calculated and used as the invalid operation count. The invalid action count is divided by the total number of action moments to obtain the action missing rate; When the missing action rate exceeds the missing action rate tolerance threshold, the working state of the self-cleaning device is marked as an action failure state.

5. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, When the rate of increase of the pressure difference before and after the fine filtration exceeds a first threshold and the rate of decrease of the pressure difference before and after the coarse filtration does not reach a second threshold within a preset time window, the coarse filtration self-cleaning device is determined to have failed, including: Within a preset time window, a sampling sequence of pressure difference values ​​before and after fine filtration is extracted. Based on the difference between the pressure difference values ​​before and after fine filtration corresponding to the start and end times of the preset time window, combined with the duration of the preset time window, the rate of increase of the pressure difference before and after fine filtration is used. All action times of the coarse filter self-cleaning device within the preset time window are obtained. The pressure difference value before and after coarse filtration corresponding to each action time is extracted as the pressure difference value before each cleaning. The pressure difference value before and after coarse filtration corresponding to the sampling time after each action time when the pressure difference drops back to the stable platform area is extracted as the pressure difference value after each cleaning. Calculate the difference between the pressure difference before each cleaning and the corresponding pressure difference after each cleaning to obtain the single drop pressure difference for each cleaning. Take the arithmetic mean of all single drop pressure differences as the decrease in pressure difference before and after coarse filtration. When the rate of increase of the pressure difference before and after the fine filter exceeds a first threshold, and the rate of decrease of the pressure difference before and after the coarse filter is less than a second threshold, the coarse filter self-cleaning device is determined to have failed.

6. The intelligent operation and maintenance management planning method for air compressor units according to claim 5, characterized in that, The steps for determining the first threshold and the second threshold include: The rising rate of the pressure difference before and after fine filtration corresponding to each of the multiple historical analysis time windows before the current moment is obtained, forming a set of historical rising rates. After removing the maximum and minimum values ​​in the set of historical rising rates, the upper quartile of the remaining values ​​is taken as the first threshold. Based on the proportion of the current pressure difference before and after the coarse filter to the final pressure difference, a saturation correction coefficient is calculated, wherein the saturation correction coefficient is directly proportional to the difference between the final pressure difference and the current pressure difference before and after the coarse filter, and inversely proportional to the final pressure difference. The second threshold is obtained by multiplying the average historical effective recovery amplitude by the saturation correction coefficient.

7. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, When it is determined that the coarse filter self-cleaning device has failed, the current effective dust holding capacity saturation of the coarse filter is calculated, and the cleaning and maintenance sequence of the coarse filter is planned according to the current effective dust holding capacity saturation, including: Taking the current moment as the endpoint, a fixed time period is extracted forward. The pressure difference value before and after coarse filtration corresponding to each collection moment within the preset time window is extracted, and the slope of the change of the pressure difference value before and after coarse filtration with time is fitted as the rate of increase of the pressure difference of coarse filtration. Based on the relative degree between the current pressure difference before and after the coarse filter and the final pressure difference, and in conjunction with the factory-calibrated dust holding capacity, the current dust holding capacity of the coarse filter is determined, and the current effective dust holding saturation is obtained based on the comparison between the current dust holding capacity and the factory-calibrated dust holding capacity. Based on the fluctuation of the coarse filtration pressure differential rise rate in each adjacent sub-interval within the preset time window, a stability coefficient of the coarse filtration pressure differential rise rate is determined, wherein the stability coefficient decreases as the dispersion of the rise rate in each sub-interval increases. Based on the relationship between the current effective dust holding capacity saturation and the cumulative running time, the running time required per unit saturation is deduced. Based on the difference between the termination pressure difference and the current pressure difference before and after, and combined with the rate of increase of the coarse filtration pressure difference, the remaining duration of the pressure difference estimation is deduced. The cleaning saturation threshold is determined based on the termination pressure difference of the coarse filter and the factory-calibrated dust holding capacity. The remaining time for saturation estimation is then calculated based on the difference between the cleaning saturation threshold and the current effective dust holding capacity, combined with the cumulative running time. Based on the stability coefficient, the contribution of the remaining time of the differential pressure projection in the fusion is determined, and the remaining time of the differential pressure projection and the remaining time of the saturation projection are fused with the contribution to obtain the coarse filter can continue to run. The current time plus the remaining time of the coarse filter is taken as the coarse filter cleaning and maintenance time.

8. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, The remaining dust holding capacity prediction step of the fine filter includes: The current effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after the fine filter, and the cumulative running time are used as inputs to the dust holding capacity prediction model of the fine filter, and the remaining dust holding capacity of the fine filter is obtained as the output. The construction steps of the fine filter dust holding capacity prediction model include: Collect historical dust holding data of the same type of fine filter under different operating conditions. For each historical record, obtain the corresponding effective dust holding saturation of the coarse filter, the rate of increase of the pressure difference before and after fine filtration, and the cumulative running time, as input feature triplet. The remaining dust holding capacity of the fine filter at the corresponding time is obtained by offline weighing and used as the output label to construct a sample pair set; A three-input single-output mapping network is constructed with the effective dust holding capacity saturation of the coarse filter, the rate of increase of the pressure difference before and after the fine filter, and the cumulative running time as inputs, and the remaining dust holding capacity of the fine filter as outputs. The input layer has 3 nodes, the hidden layer uses a nonlinear activation function, and the output layer has 1 node. The output value represents the remaining dust holding capacity of the fine filter corresponding to the current combination of input features. The sample set is divided into a fitting set and a validation set, and the mean square error between the output layer prediction value and the remaining dust capacity measured by the offline weighing method is used as the loss function. The network weights of the mapping network are updated using backpropagation of the error. The process is iteratively fitted until the error of the loss function on the validation set no longer decreases. The fitted mapping network is then used as the prediction model for the dust holding capacity of the fine filter.

9. The intelligent operation and maintenance management planning method for air compressor units according to claim 1, characterized in that, Based on the remaining dust holding capacity of the fine filter and the current effective dust holding saturation, determine the early warning time for fine filter replacement and output an operation and maintenance management plan, including: Based on the rate of increase of the pressure difference in the coarse filtration and the rate of increase of the pressure difference before and after the fine filtration, the dust consumption rate correction coefficient of the fine filtration is determined. Based on the rate of increase of the pressure difference before and after fine filtration and the correction coefficient of the dust consumption rate, the dust consumption of fine filtration per unit time is determined, and combined with the remaining dust capacity of fine filtration, the estimated remaining usable time of fine filtration is determined. Add the estimated remaining available time of the fine filter to the current time to obtain the estimated replacement time of the fine filter; Obtain the coarse filter cleaning and maintenance time. When the estimated replacement time of the fine filter is earlier than the coarse filter cleaning and maintenance time, use the estimated replacement time of the fine filter as the fine filter replacement warning time. When the estimated replacement time of the fine filter is later than or equal to the cleaning and maintenance time of the coarse filter, the cleaning and maintenance time of the coarse filter will be used as the early warning time for the replacement of the fine filter. The fine filter replacement early warning time is bound to the coarse filter cleaning and maintenance time to generate a joint shutdown maintenance time window, which is output as an operation and maintenance management plan.

10. An intelligent operation and maintenance management planning system for air compressor units, characterized in that, A method for implementing an intelligent operation and maintenance management planning method for an air compressor unit as described in any one of claims 1 to 9 includes: The operation data acquisition module is used to acquire real-time operation data of the multi-stage air intake filtration device of the mobile air compressor in the mine, wherein the multi-stage air intake filtration device includes a coarse filter and a fine filter; The self-cleaning status analysis module is used to determine the working status of the coarse filter self-cleaning device based on the pressure difference before and after the coarse filter and the pressure difference before and after the fine filter, combined with the real-time operating data. The coarse filter maintenance planning module is used to determine that the coarse filter self-cleaning device has failed when the rate of increase of the pressure difference before and after the fine filter exceeds a first threshold and the rate of decrease of the pressure difference before and after the coarse filter does not reach a second threshold within a preset time window. The fine filter capacity prediction module is used to calculate the current effective dust holding capacity saturation of the coarse filter when it is determined that the coarse filter self-cleaning device has failed, and to plan the cleaning and maintenance sequence of the coarse filter based on the current effective dust holding capacity saturation. The joint operation and maintenance planning module is used to predict the remaining dust holding capacity of the fine filter, and determine the fine filter replacement warning time based on the remaining dust holding capacity of the fine filter and the current effective dust holding saturation, and output the operation and maintenance management plan.

Images