An integrated air cannon equipment condition monitoring system and method

Abstract

This invention discloses an integrated air cannon equipment status monitoring system and method, belonging to the field of intelligent equipment monitoring technology. The invention utilizes sensors to collect operating data of the integrated air cannon; constructs a jetting demand assessment model, using different operating data to calculate blockage signal components, load change components, time factor components, and historical pattern components as inputs to the jetting demand assessment model, weighted and fused to obtain a demand index, calculates the theoretically optimal number of groups for the integrated air cannon, optimizes the theoretically optimal number of groups by judging the real-time pressure of the air tank, and constructs a time-constrained dynamic calculation interval; constructs an air source availability assessment model to calculate the maximum number of concurrent actions, and dynamically schedules the air cannon according to the maximum number of concurrent actions; optimizes and adjusts the weights of the jetting demand assessment model; and sets up a Chinese and English operating interface to construct an intelligent integrated air cannon operation panel.

Description

Technical Field

[0001] This invention relates to the field of intelligent device monitoring technology, specifically to an integrated air cannon device status monitoring system and method. Background Technology

[0002] Integrated air cannons typically operate in harsh industrial environments characterized by high temperatures, high dust levels, and high vibrations. These devices integrate multiple components, including a compressed air system, solenoid valves, the cannon body, and a piston. Their operational status is affected by various factors such as air pressure, temperature, material characteristics, and operating frequency. For example, in coal silo applications, high dust levels can cause sensor clogging, and high temperatures can affect the stability of electronic components. Frequent firing operations can lead to cannon body wear and seal aging, resulting in air leakage. Traditional single-sensor monitoring cannot comprehensively characterize the equipment's status, distinguish between "normal performance degradation" and "precursor malfunctions," and is unable to adapt to the monitoring needs of different operating conditions.

[0003] Traditional air cannon monitoring technology suffers from data silos. There is no data linkage between the air cannon and the upstream air compressor system or the downstream level gauge / blockage detection system. This may result in blind spraying when the air pressure is insufficient or ineffective spraying when the material flow is normal, wasting energy and aggravating equipment wear. Summary of the Invention

[0004] The purpose of this invention is to provide an integrated air cannon equipment status monitoring system and method to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An integrated air cannon equipment status monitoring system and method, the method comprising the following steps:

[0007] S100: Use sensors to collect operating condition data of the integrated air cannon, and set a standard data parsing module in the communication protocol to fuse the collected operating condition data.

[0008] Furthermore, the specific steps for fusing the collected data from different operating conditions are as follows:

[0009] S101. Add a standard data parsing module for the operating condition sensor to the existing communication protocol of the integrated air cannon. The operating condition sensor is used to collect operating condition data, including air source pressure data, material status data, and environmental parameters. The air source pressure data includes the real-time pressure P of the air storage tank. tank Gas supply pipeline pressure P line Instantaneous flow rate Q flow Material status data includes the silo level height H. level Blockage signal B block Downstream equipment load current Iload The blockage signal is represented by a Boolean value; environmental parameters include ambient temperature T. ambient Ambient humidity S humidity A standard data parsing module is added to the communication protocol to achieve effective fusion of data from different types of operating conditions, solve the problem of inconsistent data formats from multiple sensors, and improve data utilization efficiency.

[0010] S102. Set the working condition data acquisition time step Δt. All working condition sensors acquire data at the same time and at the same time interval to ensure data time alignment. For each different working condition data, calculate the average value of the data change after the corresponding historical data change based on the historical data change amount as the change threshold. Each working condition data has a preset corresponding change threshold. When the change amount of different working condition data acquired by the working condition sensor at the time step Δt is greater than the corresponding change threshold, the data is judged to be abnormal and discarded.

[0011] By setting time steps to align data and using thresholds based on historical changes to filter out abnormal data, we ensure that the data input into the model is accurate and noise-free, providing high-quality data support for subsequent decision-making.

[0012] After screening, the abnormal operating condition data were merged and packaged into a single operating condition data package called Data. packet ={timestamp, P sys M status E env}, where P sys M represents the gas source pressure data. status E represents material status data. env Indicates environmental parameters.

[0013] S200. Construct an injection demand assessment model. Calculate the blockage signal component, load change component, time factor component, and historical pattern component using data from different operating conditions as inputs to the injection demand assessment model. Weighted fusion is then used to obtain the demand index.

[0014] Furthermore, the specific steps for obtaining the demand index through weighted fusion are as follows:

[0015] S201. Calculate the blockage signal component using the blockage signal, using the following formula:

[0016] ;

[0017] In the formula, D block w represents the blocking signal component. b This represents the congestion weighting coefficient;

[0018] S202. Extract the reference load current I of the downstream equipment of the integrated air cannon during normal operation. base and the maximum allowable current Imax The formula for calculating the load variation component is:

[0019] ;

[0020] In the formula, D load Indicates the load variation component; I load (t) represents the downstream device load current collected at the current time t;

[0021] S203, Extract the last spray time t of the air cannon. last The formula for calculating the time factor component is:

[0022] ;

[0023] In the formula, D time The time factor component is represented by exp, the natural exponential function is represented by t, the current time is represented by τ, and the time constant is represented by τ.

[0024] S204. Extract historical operating condition events and calculate the Euclidean distance between the operating condition data in each historical operating condition event and the current operating condition data.

[0025] The formula for calculating historical pattern components is:

[0026] ;

[0027] In the formula, D history R represents the historical pattern component. success (i) represents the clearing success rate of the i-th historical working condition event, where the clearing success rate is the number of successful clearings in the i-th historical working condition event divided by the total number of blockages; w similar (i) represents the similarity weight of the i-th historical working condition event, and the similarity weight is inversely proportional to the Euclidean distance;

[0028] S205. Assign weights of α, β, γ, and δ to the four components of congestion signal, load change, time factor, and historical pattern, respectively, and standardize them, α+β+γ+δ=1; use the weights to sum the standardized values ​​of the four components to obtain the demand index; comprehensively characterize the injection demand through the four components of congestion signal, load change, time factor, and historical pattern, which can more realistically reflect the urgency of clearing congestion compared to judging by a single indicator.

[0029] The system presets injection thresholds and a reserve threshold. When the demand index is greater than the injection threshold, it determines that immediate injection is needed to clear the blockage; when the demand index is greater than the reserve threshold but less than the injection threshold, it determines that planned injection is needed; and when the demand index is less than the reserve threshold, it determines that injection should not be performed. By setting injection and reserve thresholds, the demand is divided into three levels: "immediate injection," "planned injection," and "no injection," avoiding blind injection or missing key blockage clearing opportunities, and balancing the blockage clearing effect with energy consumption.

[0030] S300: For the calculation of the theoretical optimal number of air cannons, the theoretical optimal number of air cannons is optimized by judging the real-time pressure of the air tank, and two principles are set to constrain the theoretical optimal number of air cannons; a time-constrained dynamic calculation interval is constructed.

[0031] Furthermore, the specific steps for constructing the time-constrained dynamic calculation interval are as follows:

[0032] S301. In the integrated air cannon, all air cannons are randomly divided into four groups as basic groups. The total number of air cannon devices in the integrated air cannon is extracted. Based on expert experience, the ideal number of air cannon devices in each group is set. The ideal group number is obtained by dividing the total number of air cannon devices by the ideal number of air cannon devices in each group and rounding up. The ideal group number is then compared with the basic group number and the minimum value is extracted as the theoretically optimal group number G. ideal When determining whether to perform current or planned injection for clearing blockages, concurrent injection is executed in each group based on the theoretically optimal number of groups. The theoretically optimal number of groups is calculated by combining the total number of air cannons and expert experience, and is dynamically adjusted according to the real-time pressure of the air tank to avoid action failure caused by concurrent injection when the pressure is insufficient, or waste of resources when the pressure is sufficient.

[0033] S302, Preset minimum concurrent pressure threshold, when the real-time pressure P of the gas storage tank... tank If the pressure is below the minimum concurrent pressure threshold, the gas tank pressure is considered insufficient; otherwise, the pressure is considered sufficient.

[0034] When the pressure in the gas storage tank is insufficient, a concurrent group number adjustment strategy is executed, specifically: G actual =max(2,G) ideal -1), G actual This indicates the latest number of concurrent groups; select 2 and G. ideal -1 is used for boundary protection to avoid low system efficiency due to excessively low concurrency.

[0035] When the gas storage tank pressure is sufficient, the theoretical optimal number of groups is not adjusted, and the theoretical optimal number of groups is used as the latest concurrent number of groups;

[0036] S303. After determining the latest number of concurrent groups, two principles are followed when determining the members of each group's air cannon equipment: the proximity principle and the efficiency optimization principle.

[0037] The proximity principle means that during the screening process, the air cannon device with the closest Euclidean distance is selected as a member of the same group.

[0038] The principle of optimal efficiency means selecting air cannon equipment as members of the same group based on the success rate of collaboration from high to low. The success rate of collaboration means the proportion of successful collaborative clearing times among the same air cannon equipment in the total number of clearing times in history. The principle of proximity reduces the spatial redundancy of collaborative clearing of air cannons. The principle of optimal efficiency selects members based on the historical success rate of collaboration, improves the coordination of equipment within the group, and increases the success rate of clearing.

[0039] S304, User-defined basic injection interval T base Extract the standard pressure P of the gas storage tank. nominal The injection interval time is dynamically calculated using the following formula:

[0040] ;

[0041] In the formula, T interval D represents the injection interval time. demand The demand index is used to represent the pressure dimension coefficient. The ratio of the standard argon gas in the storage tank to the real-time pressure of the storage tank is used as the pressure dimension coefficient. The lower the real-time pressure of the storage tank, the larger the ratio, and the longer the calculated injection interval time, thus reducing the execution frequency and avoiding excessive consumption of the real-time pressure of the storage tank. The reciprocal of the demand index is used as the demand dimension coefficient. The larger the demand index, the smaller the demand dimension coefficient, and the shorter the injection interval time, requiring high-frequency execution to meet urgent needs. An interval time calculation model is constructed by using the storage tank pressure ratio and the reciprocal of the demand index. When the pressure is low, the interval is extended to avoid excessive consumption, and when the demand is high, the interval is shortened to meet urgent needs. At the same time, time constraints are set to prevent the interval from being too long or too short.

[0042] Set a time constraint 0 ≤ T interval ≤99:59:59, where 99:59:59 represents 99 hours, 59 minutes, and 59 seconds.

[0043] S400: Construct an air source availability assessment model to calculate the maximum number of concurrent actions, and dynamically schedule the air cannons based on the maximum number of concurrent actions;

[0044] Furthermore, the specific steps for dynamically scheduling the air cannon based on the maximum number of concurrent actions are as follows:

[0045] S401, The gas source availability assessment model is specifically as follows:

[0046] ;

[0047] In the formula, N con This represents the maximum number of concurrent actions, where floor represents rounding down, and P minV represents the minimum pressure required for the air cannon to operate effectively. tank E represents the volume of the gas storage tank. per k represents the amount of gas consumed in a single injection. safe This indicates the safety factor; the maximum number of concurrent actions is calculated using the gas source availability assessment model to avoid insufficient gas supply due to excessive concurrency and to ensure effective pressure supply for each injection action.

[0048] S402. Initialize and set up the execution queue and waiting queue;

[0049] When immediate spraying to clear blockages is required, professionals assign a priority coefficient to each group of concurrent air cannons based on its location. The priority of each group of concurrent air cannons is calculated by multiplying the priority coefficient by the demand index. Let N be the number of concurrent air cannons in the current execution queue. current When N current <N con When the highest priority air cannon concurrent group is entered into the execution queue for immediate spraying, it is entered into the waiting queue. In the waiting queue, the air cannon concurrent groups are sorted from highest to lowest priority. Priority is set for groups that require immediate spraying, and execution permissions are dynamically allocated in combination with the execution queue and the waiting queue. High-urgency and high-priority clearing tasks are handled first to improve emergency response efficiency.

[0050] S403. Calculate the pressure recovery time in the gas storage tank using the following formula:

[0051] ;

[0052] In the formula, T recover P represents the pressure recovery time. target P represents the target pressure of the gas storage tank. current Q represents the current pressure of the gas storage tank after injection. com Q represents the air output of the air compressor. con Indicates the current gas consumption;

[0053] In the waiting queue, a maximum waiting time is set for each group of air cannons to fire concurrently. The queue is iterated and compared; if the pressure recovery time is less than the maximum waiting time for the corresponding group, the group will fire after the pressure recovery time has elapsed. If the pressure recovery time is greater than the maximum waiting time, all concurrent firing of air cannons within the corresponding group is canceled, and single firing is performed instead. Based on the comparison between the pressure recovery time and the maximum waiting time, the system flexibly selects between "waiting before firing concurrently" or "canceling concurrent firing and switching to single firing," ensuring the need for clearing congestion while avoiding exacerbating congestion due to excessive waiting time.

[0054] S500 records the number of actions of each air cannon, calculates the spray effect score, energy efficiency index, and equipment health index, and optimizes and adjusts the weights of the spray demand assessment model.

[0055] Furthermore, the specific steps for optimizing and adjusting the weights of the injection demand assessment model are as follows:

[0056] S501. After dynamically adjusting the air cannon's spray action and completing the spraying and unblocking process, extract the load current deviation of the downstream equipment before spraying and the load current of the downstream equipment after spraying, and calculate the spraying effect score using the following formula:

[0057] ;

[0058] In the formula, S eff Indicates the spray effect score, △I before Indicates the deviation of the load current of the downstream equipment before injection, ΔI after This indicates the deviation of the downstream equipment load current after injection; the deviation of the downstream equipment load current before injection is equal to the downstream load current before injection minus the reference load current; the deviation of the downstream equipment load current after injection is equal to the downstream load current after injection minus the reference load current.

[0059] S502. Calculate the energy efficiency index and equipment health index using the following formula:

[0060] , ;

[0061] In the formula, E eff H represents the energy efficiency index. health E represents the equipment health indicator. con N represents the energy consumed by the jet. fault N represents the number of equipment failures. total Indicates the total number of times the equipment was operated;

[0062] S503. During each injection unblocking operation, the injection effect score, energy efficiency index, and equipment health index are recorded as historical data. For the four weights Ω = (α, β, γ, δ) in the demand assessment model, the gradient descent algorithm is used to calculate the gradient ΔΩ of the average injection effect score with respect to each weight. The corresponding weight is then updated using this gradient. The formula is as follows:

[0063] ;

[0064] In the formula, Ω new Ω represents the updated weights. old This represents the old weights before the update. This represents the learning rate. Constraints are set such that the sum of the four new weights after the update is one, and each new weight is no less than 0.05. Using the gradient descent algorithm, the weights of the injection demand assessment model are dynamically updated based on historical evaluation data. This allows the model to adapt to changes in different operating conditions, continuously improving the accuracy of demand judgment and avoiding decision-making biases caused by model rigidity. The constraint that the sum of the weights is 1 and each weight is no less than 0.05 ensures that all four components play a role, preventing any one component from being excessively ignored and guaranteeing the model's balance.

[0065] The S600 features a bilingual (Chinese and English) operating interface, creating an intelligent integrated air cannon control panel.

[0066] Furthermore, the specific steps for constructing an intelligent integrated air cannon control panel are as follows:

[0067] S601. Set up an operation interface that can be switched between Chinese and English, and set up a working condition monitoring dashboard, intelligent decision suggestion panel, parameter adjustment interface and effect feedback view in the operation interface.

[0068] The operating condition monitoring panel displays real-time gas source pressure data, material status data, and environmental parameters; it also displays the demand index using trend charts.

[0069] The intelligent decision suggestion panel displays the calculated theoretical optimal number of groups and the injection interval time, waiting for the user to confirm the settings with one click; it also displays the demand index judgment results and the air cannon dynamic adjustment results.

[0070] The parameter adjustment interface has a manual adjustment function and displays the updated weights calculated by the system to recommend to the user.

[0071] The effect feedback view displays the total number of spray actions for each air cannon, and displays the spray effect score, energy efficiency index, and equipment health index through bar charts.

[0072] An integrated air cannon equipment status monitoring system includes a data acquisition module, a demand assessment module, an intelligent grouping module, a spray dynamic scheduling module, an assessment and update module, and an intelligent interface module.

[0073] The data acquisition module is used to collect the operating condition data of the integrated air cannon using sensors, and a standard data parsing module is set in the communication protocol to fuse the collected operating condition data.

[0074] The demand assessment module is used to construct an injection demand assessment model. It uses data from different operating conditions to calculate the blockage signal component, load change component, time factor component, and historical pattern component as inputs to the injection demand assessment model, and then weights and fuses them to obtain the demand index.

[0075] The intelligent grouping module is used to calculate the theoretically optimal number of groups for integrated air cannons and to construct a time-constrained dynamic calculation interval.

[0076] The injection dynamic scheduling module is used to construct an air source availability assessment model, calculate the maximum number of concurrent actions, and dynamically schedule the air cannons based on the maximum number of concurrent actions.

[0077] The evaluation and update module is used to record the number of actions of each air cannon, calculate the spray effect score, energy efficiency index, equipment health index, and optimize the weight of the spray demand evaluation model.

[0078] The intelligent interface module is used to set up Chinese and English operating interfaces to build an intelligent integrated air cannon operating panel.

[0079] The requirements assessment module includes a component calculation unit, a weighted fusion unit, and a judgment unit;

[0080] The component calculation unit is used to calculate the blockage signal component, load change component, time factor component and historical pattern component using data from different operating conditions.

[0081] The weighted fusion unit is used to perform weighted fusion after assigning weights to different components and standardizing them.

[0082] The judgment unit is used to preset the injection threshold and the preparation threshold, and uses the demand index to determine whether to inject immediately to clear blockages or to inject according to a plan.

[0083] The evaluation and update module includes a spray evaluation unit and a weight update unit;

[0084] The spray evaluation unit is used to record the number of actions of each air cannon and calculate the spray effect score, energy efficiency index, and equipment health index.

[0085] The weight update unit is used to calculate the gradient of the average spray effect score with respect to each weight using the gradient descent algorithm, and then update the corresponding weight using the gradient.

[0086] Compared with the prior art, the beneficial effects of the present invention are:

[0087] 1. This invention achieves autonomous adjustment of multiple factors through the following design:

[0088] Demand assessment level: Integrates four major factors: blockage, load, time, and history, to avoid the limitations of a single factor; Execution strategy level: Takes into account constraints such as gas source pressure, concurrency capability, and grouping efficiency, and dynamically adjusts the number of groups, interval time, and injection method; Continuous optimization level: Combines blockage clearing effect, energy consumption, and equipment health to update weights, enabling the system to autonomously adapt to changes in operating conditions without frequent manual intervention.

[0089] 2. The overall solution of this invention avoids ineffective injection through precise demand judgment, improves the success rate of blockage clearing through intelligent grouping and dynamic scheduling, and continuously improves decision-making accuracy through weighting, ultimately achieving the goal of "more efficient blockage clearing, more reasonable energy consumption, and more durable equipment." Compared with traditional manual operation or simple automatic control, it can reduce unnecessary gas consumption, reduce the risk of equipment failure, and reduce the cost of manual intervention.

[0090] 3. This invention connects scattered working condition data, decision data, and feedback data through data fusion, multi-component modeling, and linked index updates, constructing a closed-loop data link of "collection-analysis-decision-execution-evaluation-optimization" to avoid the one-sidedness of decision-making caused by information silos. Attached Figure Description

[0091] Figure 1 This is a schematic diagram illustrating the steps of an integrated air cannon equipment status monitoring method according to the present invention. Detailed Implementation

[0092] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0093] Example: Figure 1 As shown, the present invention provides a technical solution.

[0094] This application was used in a large thermal power plant's raw coal bunker unblocking scenario. The bunker is equipped with 16 integrated air cannons (numbered A1-A16), distributed across four sections (four cannons per section), to address the problem of poor material flow caused by adhesion and bridging on the inner walls of the bunker. Operating conditions included: large fluctuations in raw coal moisture content (ambient humidity 30%-80%), and dynamic changes in material flow rate with the power generation load (downstream feeder load current fluctuation range 5A-15A). A balance needed to be struck between unblocking efficiency, air source conservation, and equipment lifespan.

[0095] Initial parameter settings:

[0096] Data acquisition: time step Δt = 10s, threshold values ​​for data changes under various operating conditions (e.g., pressure change threshold of gas storage tank ±0.05MPa, load current change threshold ±2A).

[0097] Injection demand assessment model: Initial weights Ω = (α = 0.35, β = 0.3, γ = 0.2, δ = 0.15), injection threshold = 0.8, preparatory threshold = 0.4; clogging weight coefficient w b=0.9, time constant τ=3600s (1 hour).

[0098] Grouping parameters: Ideal number of air cannons per group = 4, minimum concurrent pressure threshold = 0.8 MPa; basic injection time interval 30 min, time constraint 0 ≤ T interval ≤99:59:59.

[0099] Air source and control parameters: Minimum effective pressure P for air cannon operation min =0.6MPa, gas consumption per injection E per =0.1m³, safety factor k safe =0.8; maximum waiting time per group = 10 min; learning rate η = 0.05.

[0100] An integrated air cannon equipment status monitoring system and method, the method comprising the following steps:

[0101] S100: Use sensors to collect operating condition data of the integrated air cannon, and set a standard data parsing module in the communication protocol to fuse the collected operating condition data.

[0102] The specific steps for fusing the collected data from different operating conditions are as follows:

[0103] S101. Add a standard data parsing module for the operating condition sensor to the existing communication protocol of the integrated air cannon. The operating condition sensor is used to collect operating condition data, including air source pressure data, material status data, and environmental parameters. The air source pressure data includes the real-time pressure P of the air storage tank. tank Gas supply pipeline pressure P line Instantaneous flow rate Q flow Material status data includes the silo level height H. level Blockage signal B block Downstream equipment load current I load The blockage signal is represented by a Boolean value; environmental parameters include ambient temperature T. ambient Ambient humidity S humidity A standard data parsing module is added to the communication protocol to achieve effective fusion of data from different types of operating conditions, solve the problem of inconsistent data formats from multiple sensors, and improve data utilization efficiency.

[0104] S102. Set the working condition data acquisition time step Δt. All working condition sensors acquire data at the same time and at the same time interval to ensure data time alignment. For each different working condition data, calculate the average value of the data change after the corresponding historical data change based on the historical data change amount as the change threshold. Each working condition data has a preset corresponding change threshold. When the change amount of different working condition data acquired by the working condition sensor at the time step Δt is greater than the corresponding change threshold, the data is judged to be abnormal and discarded.

[0105] By setting time steps to align data and using thresholds based on historical changes to filter out abnormal data, we ensure that the data input into the model is accurate and noise-free, providing high-quality data support for subsequent decision-making.

[0106] After screening, the abnormal operating condition data were merged and packaged into a single operating condition data package called Data. packet ={timestamp, P sys M status E env}, where P sys M represents the gas source pressure data. status E represents material status data. env Indicates environmental parameters.

[0107] The sensor synchronously collects data at time steps Δt = 10s:

[0108] Gas source pressure data: P tank =1.1MPa, P line =1.05MPa, Q flow =0.3 m³ / min;

[0109] Material status data: H level =6.2m (warehouse height 8m), B block =1 (partial blockage), I load (t) = 13A;

[0110] Environmental parameters: T ambient =25℃, S humidity =65%.

[0111] Data filtering: The change in each data point is less than the corresponding threshold (e.g., P). tank The change was 0.02 MPa < 0.05 MPa, with no abnormal data.

[0112] Fusion Packaging: D atapacket ={Timestamp: 2024-05-2010:00:00, P sys ={1.1,1.05,0.3},M status ={6.2,1,13},E env ={25,65}}.

[0113] S200. Construct an injection demand assessment model. Calculate the blockage signal component, load change component, time factor component, and historical pattern component using data from different operating conditions as inputs to the injection demand assessment model. Weighted fusion is then used to obtain the demand index.

[0114] The specific steps for obtaining the demand index through weighted fusion are as follows:

[0115] S201. Calculate the blockage signal component using the blockage signal, using the following formula:

[0116] ;

[0117] In the formula, D block w represents the blocking signal component. b This represents the congestion weighting coefficient;

[0118] S202. Extract the reference load current I of the downstream equipment of the integrated air cannon during normal operation. base and the maximum allowable current I max The formula for calculating the load variation component is:

[0119] ;

[0120] In the formula, D load Indicates the load variation component; I load (t) represents the downstream device load current collected at the current time t;

[0121] S203, Extract the last spray time t of the air cannon. last The formula for calculating the time factor component is:

[0122] ;

[0123] In the formula, D time The time factor component is represented by exp, the natural exponential function is represented by t, the current time is represented by τ, and the time constant is represented by τ.

[0124] S204. Extract historical operating condition events and calculate the Euclidean distance between the operating condition data in each historical operating condition event and the current operating condition data.

[0125] The formula for calculating historical pattern components is:

[0126] ;

[0127] In the formula, D history R represents the historical pattern component. success(i) represents the clearing success rate of the i-th historical working condition event, where the clearing success rate is the number of successful clearings in the i-th historical working condition event divided by the total number of blockages; w similar (i) represents the similarity weight of the i-th historical working condition event, and the similarity weight is inversely proportional to the Euclidean distance;

[0128] S205. Assign weights of α, β, γ, and δ to the four components of congestion signal, load change, time factor, and historical pattern, respectively, and standardize them, α+β+γ+δ=1; use the weights to sum the standardized values ​​of the four components to obtain the demand index; comprehensively characterize the injection demand through the four components of congestion signal, load change, time factor, and historical pattern, which can more realistically reflect the urgency of clearing congestion compared to judging by a single indicator.

[0129] The system presets injection thresholds and a reserve threshold. When the demand index is greater than the injection threshold, it determines that immediate injection is needed to clear the blockage; when the demand index is greater than the reserve threshold but less than the injection threshold, it determines that planned injection is needed; and when the demand index is less than the reserve threshold, it determines that injection should not be performed. By setting injection and reserve thresholds, the demand is divided into three levels: "immediate injection," "planned injection," and "no injection," avoiding blind injection or missing key blockage clearing opportunities, and balancing the blockage clearing effect with energy consumption.

[0130] Component calculation:

[0131] Blockage signal component: D block =w b ×B block =0.9 × 1 = 0.9;

[0132] Load variation component: Known I base =8A (normal reference current), I max =16A, substituting into the formula, we get D. load =(13-8) / (16-8)=0.625;

[0133] Time factor component: last injection time t last =2024-05-2009:20:00, tt last =2400s, D time =1-exp(-2400 / 3600)=1-0.513=0.487;

[0134] Historical pattern component: Extract 5 similar historical conditions, and the success rate R of clearing the congestion is the 3 conditions with the smallest Euclidean distance. success The similarity weights are 0.9, 0.85, and 0.8 respectively, with a similarity weight w. similar It is inversely proportional to the distance, and D is calculated. history =0.86.

[0135] Demand index calculation: After standardization and weighted summation, demand index = 0.35×0.9 + 0.3×0.625 + 0.2×0.487 + 0.15×0.86 ≈ 0.78.

[0136] Demand assessment: 0.4 < 0.78 < 0.8, therefore it is determined to be "planned jet cleaning".

[0137] S300: For the calculation of the theoretical optimal number of air cannons, the theoretical optimal number of air cannons is optimized by judging the real-time pressure of the air tank, and two principles are set to constrain the theoretical optimal number of air cannons; a time-constrained dynamic calculation interval is constructed.

[0138] The specific steps for constructing a time-constrained dynamic calculation interval are as follows:

[0139] S301. In the integrated air cannon, all air cannons are randomly divided into four groups as basic groups. The total number of air cannon devices in the integrated air cannon is extracted. Based on expert experience, the ideal number of air cannon devices in each group is set. The ideal group number is obtained by dividing the total number of air cannon devices by the ideal number of air cannon devices in each group and rounding up. The ideal group number is then compared with the basic group number and the minimum value is extracted as the theoretically optimal group number G. ideal When determining whether to perform current or planned injection for clearing blockages, concurrent injection is executed in each group based on the theoretically optimal number of groups. The theoretically optimal number of groups is calculated by combining the total number of air cannons and expert experience, and is dynamically adjusted according to the real-time pressure of the air tank to avoid action failure caused by concurrent injection when the pressure is insufficient, or waste of resources when the pressure is sufficient.

[0140] S302, Preset minimum concurrent pressure threshold, when the real-time pressure P of the gas storage tank... tank If the pressure is below the minimum concurrent pressure threshold, the gas tank pressure is considered insufficient; otherwise, the pressure is considered sufficient.

[0141] When the pressure in the gas storage tank is insufficient, a concurrent group number adjustment strategy is executed, specifically: G actual =max(2,G) ideal -1), G actual This indicates the latest number of concurrent groups; select 2 and G. ideal -1 is used for boundary protection to avoid low system efficiency due to excessively low concurrency.

[0142] When the gas storage tank pressure is sufficient, the theoretical optimal number of groups is not adjusted, and the theoretical optimal number of groups is used as the latest concurrent number of groups;

[0143] S303. After determining the latest number of concurrent groups, two principles are followed when determining the members of each group's air cannon equipment: the proximity principle and the efficiency optimization principle.

[0144] The proximity principle means that during the screening process, the air cannon device with the closest Euclidean distance is selected as a member of the same group.

[0145] The principle of optimal efficiency means selecting air cannon equipment as members of the same group based on the success rate of collaboration from high to low. The success rate of collaboration means the proportion of successful collaborative clearing times among the same air cannon equipment in the total number of clearing times in history. The principle of proximity reduces the spatial redundancy of collaborative clearing of air cannons. The principle of optimal efficiency selects members based on the historical success rate of collaboration, improves the coordination of equipment within the group, and increases the success rate of clearing.

[0146] S304, User-defined basic injection interval T base Extract the standard pressure P of the gas storage tank. nominal The injection interval time is dynamically calculated using the following formula:

[0147] ;

[0148] In the formula, T interval D represents the injection interval time. demand The demand index is used to represent the pressure dimension coefficient. The ratio of the standard argon gas in the storage tank to the real-time pressure of the storage tank is used as the pressure dimension coefficient. The lower the real-time pressure of the storage tank, the larger the ratio, and the longer the calculated injection interval time, thus reducing the execution frequency and avoiding excessive consumption of the real-time pressure of the storage tank. The reciprocal of the demand index is used as the demand dimension coefficient. The larger the demand index, the smaller the demand dimension coefficient, and the shorter the injection interval time, requiring high-frequency execution to meet urgent needs. An interval time calculation model is constructed by using the storage tank pressure ratio and the reciprocal of the demand index. When the pressure is low, the interval is extended to avoid excessive consumption, and when the demand is high, the interval is shortened to meet urgent needs. At the same time, time constraints are set to prevent the interval from being too long or too short.

[0149] Set a time constraint 0 ≤ T interval ≤99:59:59, where 99:59:59 represents 99 hours, 59 minutes, and 59 seconds.

[0150] Theoretical optimal number of groups: Total number of air cannons: 16, ideal number of groups = 16 ÷ 4 = 4, basic number of groups = 4, G ideal =min(4,4)=4.

[0151] Group number optimization: Real-time pressure P of the gas storage tank tank =1.1MPa > 0.8MPa (concurrent minimum pressure threshold), pressure is sufficient, G actual =4.

[0152] Group execution: Following the principles of proximity (closest Euclidean distance between air cannons in the same compartment) and optimal efficiency (top 4 combinations with the highest historical success rate in coordinated operations), the final grouping is as follows:

[0153] Group 1: A1, A2, A3, A4 (Warehouse Segment 1, 92% success rate in coordination);

[0154] Group 2: A5, A6, A7, A8 (Warehouse Segment 2, 89% success rate in coordination);

[0155] Group 3: A9, A10, A11, A12 (Warehouse segment 3, 90% success rate in coordination);

[0156] Group 4: A13, A14, A15, A16 (Warehouse segment 4, with a collaboration success rate of 88%).

[0157] The dynamic interval time = 30min × (1.2 / 1.1) × (1 / 0.78) ≈ 30 × 1.09 × 1.28 ≈ 41min, that is, the spray interval is set to 41min.

[0158] S400: Construct an air source availability assessment model to calculate the maximum number of concurrent actions, and dynamically schedule the air cannons based on the maximum number of concurrent actions;

[0159] The specific steps for dynamically scheduling the air cannon based on the maximum number of concurrent actions are as follows:

[0160] S401, The gas source availability assessment model is specifically as follows:

[0161] ;

[0162] In the formula, N con This represents the maximum number of concurrent actions, where floor represents rounding down, and P min V represents the minimum pressure required for the air cannon to operate effectively. tank E represents the volume of the gas storage tank. per k represents the amount of gas consumed in a single injection. safe This indicates the safety factor; the maximum number of concurrent actions is calculated using the gas source availability assessment model to avoid insufficient gas supply due to excessive concurrency and to ensure effective pressure supply for each injection action.

[0163] S402. Initialize and set up the execution queue and waiting queue;

[0164] When immediate spraying to clear blockages is required, professionals assign a priority coefficient to each group of concurrent air cannons based on its location. The priority of each group of concurrent air cannons is calculated by multiplying the priority coefficient by the demand index. Let N be the number of concurrent air cannons in the current execution queue. current When N current <N conWhen the highest priority air cannon concurrent group is selected, it is entered into the execution queue; otherwise, it is entered into the waiting queue. In the waiting queue, the air cannon concurrent groups are sorted from highest to lowest priority. Priority is set for groups with immediate spraying needs. Execution permissions are dynamically allocated in combination with the execution queue and the waiting queue to prioritize high-urgency and high-priority clearing tasks and improve emergency response efficiency.

[0165] S403. Calculate the pressure recovery time in the gas storage tank using the following formula:

[0166] ;

[0167] In the formula, T recover P represents the pressure recovery time. target P represents the target pressure of the gas storage tank. current Q represents the current pressure of the gas storage tank after injection. com Q represents the air output of the air compressor. con Indicates the current gas consumption;

[0168] In the waiting queue, a maximum waiting time is set for each group of air cannons to fire concurrently. The queue is iterated and compared; if the pressure recovery time is less than the maximum waiting time for the corresponding group, the group will fire after the pressure recovery time has elapsed. If the pressure recovery time is greater than the maximum waiting time, all concurrent firing of air cannons within the corresponding group is canceled, and single firing is performed instead. Based on the comparison between the pressure recovery time and the maximum waiting time, the system flexibly selects between "waiting before firing concurrently" or "canceling concurrent firing and switching to single firing," ensuring the need for clearing congestion while avoiding exacerbating congestion due to excessive waiting time.

[0169] Maximum concurrent actions: Substituted into the gas source availability assessment model, N con =floor[(P tank ×V tank -P min ×V tank )×k safe / E per =floor[(1.1×10-0.6×10)×0.8 / 0.1]=floor[5×0.8 / 0.1]=40, actual maximum concurrent action count = 40.

[0170] Queue initialization: Execute queue N current =0, waiting queue is empty.

[0171] Priority Calculation: Professionals set priority coefficients based on the importance of the warehouse segment (Group 1 = 1.2, Group 2 = 1.0, Group 3 = 1.1, Group 4 = 0.9). Priority = Priority Coefficient × Demand Index.

[0172] Group 1: 1.2 × 0.78 = 0.936; Group 2: 1.0 × 0.78 = 0.78; Group 3: 1.1 × 0.78 = 0.858; Group 4: 0.9 × 0.78 = 0.702.

[0173] Scheduled execution: N current =0 < 4. Since the total number of air groups is 16, which is much smaller than the maximum concurrent action, all groups can be sorted by priority and entered into the execution queue for immediate spraying to clear blockages.

[0174] Pressure recovery time calculation: P after injection current =1.1-(4×0.1) / 10=1.06MPa, target pressure Pt arget =1.1MP, Q con =0.3 m³ / min, T recover =(1.1-1.06)×10 / (0.8-0.3)=0.4 / 0.5=0.8min<10min (maximum waiting time), no need to adjust the spraying method, and there is no grouping in the waiting queue;

[0175] S500 records the number of actions of each air cannon, calculates the spray effect score, energy efficiency index, and equipment health index, and optimizes and adjusts the weights of the spray demand assessment model.

[0176] The specific steps for optimizing and adjusting the weights of the injection demand assessment model are as follows:

[0177] S501. After dynamically adjusting the air cannon's spray action and completing the spraying and unblocking process, extract the load current deviation of the downstream equipment before spraying and the load current of the downstream equipment after spraying, and calculate the spraying effect score using the following formula:

[0178] ;

[0179] In the formula, S eff Indicates the spray effect score, △I before Indicates the deviation of the load current of the downstream equipment before injection, ΔI after This indicates the deviation of the downstream equipment load current after injection; the deviation of the downstream equipment load current before injection is equal to the downstream load current before injection minus the reference load current; the deviation of the downstream equipment load current after injection is equal to the downstream load current after injection minus the reference load current.

[0180] S502. Calculate the energy efficiency index and equipment health index using the following formula:

[0181] , ;

[0182] In the formula, E effH represents the energy efficiency index. health E represents the equipment health indicator. con N represents the energy consumed by the jet. fault N represents the number of equipment failures. total Indicates the total number of times the equipment was operated;

[0183] S503. During each injection unblocking operation, the injection effect score, energy efficiency index, and equipment health index are recorded as historical data. For the four weights Ω = (α, β, γ, δ) in the demand assessment model, the gradient descent algorithm is used to calculate the gradient ΔΩ of the average injection effect score with respect to each weight. The corresponding weight is then updated using this gradient. The formula is as follows:

[0184] ;

[0185] In the formula, Ω new Ω represents the updated weights. old This represents the old weights before the update. This represents the learning rate. Constraints are set such that the sum of the four new weights after the update is one, and each new weight is no less than 0.05. Using the gradient descent algorithm, the weights of the injection demand assessment model are dynamically updated based on historical evaluation data. This allows the model to adapt to changes in different operating conditions, continuously improving the accuracy of demand judgment and avoiding decision-making biases caused by model rigidity. The constraint that the sum of the weights is 1 and each weight is no less than 0.05 ensures that all four components play a role, preventing any one component from being excessively ignored and guaranteeing the model's balance.

[0186] The S600 features a bilingual (Chinese and English) operating interface, creating an intelligent integrated air cannon control panel.

[0187] The specific steps for constructing an intelligent integrated air cannon control panel are as follows:

[0188] S601. Set up an operation interface that can be switched between Chinese and English, and set up a working condition monitoring dashboard, intelligent decision suggestion panel, parameter adjustment interface and effect feedback view in the operation interface.

[0189] The operating condition monitoring panel displays real-time gas source pressure data, material status data, and environmental parameters; it also displays the demand index using trend charts.

[0190] The intelligent decision suggestion panel displays the calculated theoretical optimal number of groups and the injection interval time, waiting for the user to confirm the settings with one click; it also displays the demand index judgment results and the air cannon dynamic adjustment results.

[0191] The parameter adjustment interface has a manual adjustment function and displays the updated weights calculated by the system to recommend to the user.

[0192] The effect feedback view displays the total number of spray actions for each air cannon, and displays the spray effect score, energy efficiency index, and equipment health index through bar charts.

[0193] An integrated air cannon equipment status monitoring system includes a data acquisition module, a demand assessment module, an intelligent grouping module, a spray dynamic scheduling module, an assessment and update module, and an intelligent interface module.

[0194] The data acquisition module is used to collect the operating condition data of the integrated air cannon using sensors, and a standard data parsing module is set in the communication protocol to fuse the collected operating condition data.

[0195] The demand assessment module is used to construct an injection demand assessment model. It uses data from different operating conditions to calculate the blockage signal component, load change component, time factor component, and historical pattern component as inputs to the injection demand assessment model, and then weights and fuses them to obtain the demand index.

[0196] The intelligent grouping module is used to calculate the theoretically optimal number of groups for integrated air cannons and to construct a time-constrained dynamic calculation interval.

[0197] The injection dynamic scheduling module is used to construct an air source availability assessment model, calculate the maximum number of concurrent actions, and dynamically schedule the air cannons based on the maximum number of concurrent actions.

[0198] The evaluation and update module is used to record the number of actions of each air cannon, calculate the spray effect score, energy efficiency index, equipment health index, and optimize the weight of the spray demand evaluation model.

[0199] The intelligent interface module is used to set up Chinese and English operating interfaces to build an intelligent integrated air cannon operating panel.

[0200] The requirements assessment module includes a component calculation unit, a weighted fusion unit, and a judgment unit;

[0201] The component calculation unit is used to calculate the blockage signal component, load change component, time factor component and historical pattern component using data from different operating conditions.

[0202] The weighted fusion unit is used to perform weighted fusion after assigning weights to different components and standardizing them.

[0203] The judgment unit is used to preset the injection threshold and the preparation threshold, and uses the demand index to determine whether to inject immediately to clear blockages or to inject according to a plan.

[0204] The evaluation and update module includes a spray evaluation unit and a weight update unit;

[0205] The spray evaluation unit is used to record the number of actions of each air cannon and calculate the spray effect score, energy efficiency index, and equipment health index.

[0206] The weight update unit is used to calculate the gradient of the average spray effect score with respect to each weight using the gradient descent algorithm, and then update the corresponding weight using the gradient.

[0207] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method of equipment condition monitoring for an integrated air cannon, characterized by: The method includes the following steps: S100: Use sensors to collect operating data of the integrated air cannon; S200. Construct an injection demand assessment model. Calculate the blockage signal component, load change component, time factor component, and historical pattern component using data from different operating conditions as inputs to the injection demand assessment model. Weighted fusion is used to obtain the demand index. The demand index is then used to determine whether injection is needed to clear the blockage. The specific steps for determining whether jet cleaning is needed using the demand index are as follows: The four components—congestion signal component, load change component, time factor component, and historical pattern component—are assigned weights of α, β, γ, and δ, respectively, and then standardized, with α+β+γ+δ=1. The demand index is obtained by weighted summation of the standardized values ​​of the four components using the weights. The system presets injection thresholds and a reserve threshold. When the demand index is greater than the injection threshold, it determines that immediate injection is needed to clear the blockage. When the demand index is greater than the reserve threshold but less than the injection threshold, it determines that planned injection is needed to clear the blockage. When the demand index is less than the reserve threshold, it determines that injection is not needed. S300: For the calculation of the theoretical optimal number of air cannons, the theoretical optimal number of air cannons is optimized by judging the real-time pressure of the air tank, and two principles are set to constrain the theoretical optimal number of air cannons; a time-constrained dynamic calculation interval is constructed. The specific steps for calculating the theoretically optimal number of sets are as follows: In an integrated air cannon system, all air cannons are randomly divided into four groups as the basic groups. The total number of air cannon devices in the integrated air cannon system is extracted. Based on expert experience, the ideal number of air cannon devices in each group is set. The ideal group number is obtained by dividing the total number of air cannon devices by the ideal number of air cannon devices in each group and rounding up. The ideal group number is then compared with the basic group number across the four groups, and the minimum value is taken as the theoretically optimal number of groups G. ideal When determining between current injection for clearing blockages and planned injection for clearing blockages, concurrent injection is performed in each group based on the theoretically optimal number of groups. The time constraint is: 0 ≤ T interval ≤ 99:59:59, where 99:59:59 means 99 hours 59 minutes 59 seconds. S400: Construct an air source availability assessment model to calculate the maximum number of concurrent actions, initialize and set the execution queue and waiting queue, and dynamically schedule the air cannons according to the maximum number of concurrent actions; S500 calculates the spray effect score, energy efficiency index, and equipment health index, and optimizes and adjusts the weights of the spray demand assessment model. The S600 features a bilingual (Chinese and English) operating interface, creating an intelligent integrated air cannon control panel.

2. The method of claim 1, wherein: The optimal number of theoretically optimal groups in S300 is specifically determined by judging the real-time pressure of the gas storage tank: A preset concurrent minimum pressure threshold is set, when the real-time pressure P tank If the pressure is less than the concurrent minimum pressure threshold, it is determined that the gas tank is insufficient; otherwise, it is determined that the pressure is sufficient. When the pressure in the gas storage tank is insufficient, a concurrent group number adjustment strategy is executed, specifically: G actual =max(2,G) ideal -1), Gactual represents the latest number of concurrent groups, choose 2 and G ideal -1 is used for boundary protection at the maximum value; When the gas storage tank pressure is sufficient, the theoretical optimal number of groups is not adjusted, and the theoretical optimal number of groups is used as the latest concurrent number of groups.

3. The method of claim 1, wherein: The two principles in S300 include the proximity principle and the efficiency optimization principle; The proximity principle means that during the screening process, the air cannon device with the closest Euclidean distance is selected as a member of the same group. The principle of optimal efficiency means selecting air cannon devices as members of the same group based on the success rate of collaboration from high to low. The success rate of collaboration means the proportion of successful collaborative clearing times among the same air cannon devices in the total number of clearing times in the history of clearing blockages.

4. The method for monitoring the equipment status of an integrated air cannon according to claim 1, characterized in that: The specific steps for dynamically scheduling the air cannon based on the maximum number of concurrent actions in S400 are as follows: Initialize and set up the execution queue and waiting queue; When immediate spraying to clear blockages is required, professionals assign a priority coefficient to each group of concurrent air cannons based on its location. The priority of each group of concurrent air cannons is calculated by multiplying the priority coefficient by the demand index. Let N be the number of concurrent air cannons in the current execution queue. current When N current <N con When the priority is high, the highest priority concurrent air cannon group is entered into the execution queue and spraying is executed immediately; otherwise, it is entered into the waiting queue, where concurrent air cannon groups are sorted in descending order of priority. N con represents the maximum number of concurrent actions; Calculate the pressure recovery time in the gas storage tank; Set the maximum waiting time for each group of air cannons to fire concurrently in the waiting queue. Iterate through and compare the groups. If the pressure recovery time is less than the maximum waiting time of the corresponding group, execute the spraying of the corresponding group after the pressure recovery time. If the pressure recovery time is greater than the maximum waiting time of the corresponding group, cancel the concurrent execution of all air cannons in the corresponding group and execute them as single shots instead.

5. The method of claim 1, wherein: The specific steps for optimizing and adjusting the weights of the injection demand assessment model in S500 are as follows: Each time the spraying and unblocking process is performed, the spraying effect score, energy efficiency index, and equipment health index are recorded as historical data. For the four weights Ω = (α, β, γ, δ) in the demand assessment model, the gradient descent algorithm is used to calculate the gradient ΔΩ of the average spraying effect score with respect to the weight for each weight. The gradient is used to update the corresponding weights, and the constraint condition is set that the sum of the four new weights after the update is one, and each new weight after the update is not less than 0.

05.

6. An integrated air cannon system for monitoring the condition of a device using the method of any one of claims 1 to 5, wherein: The equipment status monitoring system includes a data acquisition module, a demand assessment module, an intelligent grouping module, a dynamic spraying scheduling module, an assessment and update module, and an intelligent interface module. The data acquisition module is used to collect the operating condition data of the integrated air cannon using sensors, and a standard data parsing module is set in the communication protocol to fuse the collected operating condition data. The demand assessment module is used to construct an injection demand assessment model. It uses data from different operating conditions to calculate the blockage signal component, load change component, time factor component, and historical pattern component as inputs to the injection demand assessment model, and then weights and fuses them to obtain the demand index. The intelligent grouping module is used to calculate the theoretically optimal number of groups for integrated air cannons and to construct a time-constrained dynamic calculation interval. The injection dynamic scheduling module is used to construct an air source availability assessment model, calculate the maximum number of concurrent actions, and dynamically schedule the air cannons based on the maximum number of concurrent actions. The evaluation and update module is used to record the number of actions of each air cannon, calculate the spray effect score, energy efficiency index, equipment health index, and optimize the weight of the spray demand evaluation model. The intelligent interface module is used to set up Chinese and English operating interfaces to build an intelligent integrated air cannon operating panel.

7. An integrated air cannon plant condition monitoring system according to claim 6, characterized in that: The demand assessment module includes a component calculation unit, a weighted fusion unit, and a judgment unit; The component calculation unit is used to calculate the blockage signal component, load change component, time factor component and historical pattern component using data from different operating conditions. The weighted fusion unit is used to perform weighted fusion after assigning weights to different components and standardizing them. The judgment unit is used to preset the injection threshold and the preparation threshold, and uses the demand index to determine whether to inject immediately to clear blockages or to inject according to a plan.

8. An integrated air cannon plant condition monitoring system according to claim 6, characterized in that: The evaluation and update module includes a spray evaluation unit and a weight update unit; The spray evaluation unit is used to record the number of actions of each air cannon and calculate the spray effect score, energy efficiency index, and equipment health index. The weight update unit is used to calculate the gradient of the average spray effect score with respect to each weight using the gradient descent algorithm, and then update the corresponding weight using the gradient.

Images