Punching machine equipment data analysis system and method based on multi-source data fusion

Abstract

The application discloses a punch equipment data analysis system and method based on multi-source data fusion, and relates to the technical field of punch equipment maintenance.The method comprises the following steps: collecting punch equipment data; constructing a lubrication demand index model; automatically adjusting model parameters and triggering an abnormal processing mechanism when an oil product index anomaly is detected; constructing an oil quality lubrication attenuation coupling curve; when the lubricating oil quality detection data does not reach the oil quality threshold, the oil product replacement process is preferentially triggered; when the lubricating oil quality detection data reaches the oil quality threshold, the lubricating detection frequency and the oil supplement amount are dynamically adjusted; based on the lubrication demand index model and the oil quality lubrication attenuation coupling curve, instant, advance, timing or emergency lubricating oil spraying instructions are generated, and spraying data is recorded.The application sets an oil quality pre-detection and abnormal response mechanism, preferentially excludes unqualified oil product risks, ensures that equipment is timely lubricated under high-risk working conditions, and reduces wear and failure probability.

Description

Technical Field

[0001] This invention relates to the field of punch press equipment maintenance technology, specifically to a punch press equipment data analysis system and method based on multi-source data fusion. Background Technology

[0002] As a key piece of equipment in the machining field, the operational stability of punch presses is closely related to lubrication effectiveness. Traditional lubrication management for punch presses primarily employs timed, quantitative, or simple threshold alarm methods, which have significant technical limitations: Firstly, fixed-cycle lubrication strategies cannot adapt to changes in actual operating conditions, easily leading to over-lubrication causing resource waste or insufficient lubrication causing equipment wear; secondly, existing technologies often rely solely on single operating parameters (such as temperature or vibration) for judgment, lacking integrated analysis of multi-dimensional data such as lubricant quality, equipment load, and environmental factors, resulting in low accuracy in fault prediction. Especially in high-intensity, high-precision modern stamping production, traditional methods struggle to assess oil film changes in real time, making preventative maintenance impossible.

[0003] While existing technologies have developed lubrication control systems based on sensor data, they generally suffer from problems such as limited data processing dimensions, delayed model updates, and inadequate anomaly response mechanisms, resulting in insufficient adaptability and reliability under complex operating conditions. Furthermore, existing systems lack sufficient modeling accuracy for lubricant performance degradation, failing to accurately predict oil film failure times and severely impacting the timeliness of lubrication decisions. Summary of the Invention

[0004] The purpose of this invention is to provide a data analysis system and method for punching equipment based on multi-source data fusion, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a data analysis method for punching machine equipment based on multi-source data fusion, the method comprising:

[0006] S100: Collect data from the punch press equipment, including real-time operating parameters of the punch press, historical lubrication data, and lubricating oil quality test data.

[0007] S200: Construct a lubrication demand index model, using the real-time operating parameters of the punch press as initial state parameters and the lubricating oil quality test data as correction factors input into the model; when abnormal oil indicators are detected, the model parameters are automatically adjusted and an anomaly handling mechanism is triggered.

[0008] S300: Construct a lubrication attenuation coupling curve based on oil quality, using lubricating oil quality test data as the curve starting point parameter, and setting differentiated attenuation coefficients according to oil quality grade;

[0009] S400 When the lubricating oil quality test data does not reach the oil quality threshold, the oil replacement process is triggered first. When the lubricating oil quality test data reaches the oil quality threshold, the lubrication test frequency and oil replenishment amount are dynamically adjusted based on the lubrication demand index calculated by the lubrication demand index model and the lubrication attenuation coupling curve of oil quality.

[0010] The S500 generates instant, advance, timed, or emergency lubricating oil spraying commands based on the lubrication demand index model and the lubrication attenuation coupling curve of oil quality, and records the spraying data.

[0011] According to the above scheme, the real-time operating parameters of the punch press include the number of strokes, load intensity, stamping frequency, temperature of key parts of the equipment, ambient temperature and humidity, and material hardness; historical lubrication data includes historical oil replenishment amount, historical lubrication cycle, and historical oil film failure records; and lubricating oil quality test data is the test data before the lubricating oil is sprayed, including viscosity, moisture content, impurity particle size, and effective concentration of additives.

[0012] According to the above scheme, step S200 includes:

[0013] S210. Using the number of strokes, load intensity, stamping frequency, temperature of key parts of the equipment, ambient temperature and humidity, and material hardness in the real-time operating parameters of the punch press as basic parameters, establish an initial lubrication requirement assessment value.

[0014] S220: Using viscosity, moisture content, impurity particle size and effective additive concentration in lubricating oil quality test data as correction factors, the initial lubrication demand assessment value is corrected in real time through multi-parameter weighted fusion calculation.

[0015] The lubrication demand index is calculated using the following formula:

[0016] LI=∑(ω i ×P i )×∏(1+α j ×Q j );

[0017] Where LI represents the lubrication demand index; ω i P represents the weight coefficient of the i-th real-time running parameter; i α represents the normalized value of the i-th real-time running parameter; j Q represents the correction factor for the j-th lubricating oil quality parameter; j This represents the normalized deviation value of the j-th lubricating oil quality parameter;

[0018] S230. When the viscosity of the lubricating oil exceeds the preset operating parameter range of the equipment, the moisture content exceeds the specified safety threshold of the equipment, the particle size of the impurities reaches the preset pollution level, or the effective concentration of the additive is lower than the standard threshold, the lubrication demand index model automatically increases the weight of the corresponding abnormal indicators in the calculation of the lubrication demand index and triggers the abnormal handling mechanism. The abnormal handling mechanism includes starting the secondary verification process of lubricating oil quality and generating abnormal warning information.

[0019] The relevant thresholds for real-time operating parameters of the punch press, including the safe upper limit of temperature of key parts of the equipment, the rated range of load intensity and the design limit of punching frequency, are determined based on the hardware parameters of the equipment's motor power and mechanical structure strength to ensure that the operating parameters do not exceed the physical tolerance range of the equipment.

[0020] The basic quality thresholds for lubricating oil, including the standard viscosity range, upper limit of water content, limit of impurity particle size grade, and minimum effective concentration threshold of additives, are set according to the lubricating oil model and specifications specified in the equipment manual and the oil performance indicators recommended by the manufacturer, to ensure that the physical and chemical properties of the oil are suitable for the lubrication needs of the equipment.

[0021] The lubrication demand index model automatically increases the weighting of corresponding abnormal indicators in the lubrication demand index calculation, as shown in the following formula:

[0022] ω' j =ω j ×(1+β×|Q j |);

[0023] Where, ω' j denoted as the adjusted weighting coefficient; β represents the abnormal influence factor; |Qj| represents the absolute value of the deviation of the j-th parameter;

[0024] According to the above scheme, step S300 includes:

[0025] S310. Using the viscosity, moisture content, impurity particle size, and effective additive concentration in the lubricating oil quality test data as initial input parameters, determine the initial state parameters of the lubrication attenuation coupling curve of the oil quality.

[0026] S320. Based on the various index values ​​of lubricating oil quality test data, the oil is divided into different oil quality grades. The oil quality grade is based on the viscosity compliance range, the safe range of water content, the degree of impurity particle size contamination, and the degree of deviation from the effective concentration threshold of additives.

[0027] S330: Different lubrication attenuation coefficients are configured for oils of different quality grades, and the attenuation coefficients are negatively correlated with the oil quality grade.

[0028] S340. Based on the initial state parameters and the attenuation coefficient, construct the oil quality lubrication attenuation coupling curve. The oil quality lubrication attenuation coupling curve is used to represent the law of change of lubrication efficiency of oil over time during the operation of the punch press.

[0029] The oil film thickness decay model is shown in the following formula:

[0030] E(t) = E0 × e (-λ×t) ×∑(γ k ×S k );

[0031] Where E(t) represents the lubrication efficiency at time t; E0 represents the initial lubrication efficiency; λ represents the attenuation coefficient; t represents the running time; γ k S is denoted as the influence coefficient of the k-th real-time operating parameter on lubrication efficiency; k It is represented as the normalized value of the k-th real-time running parameter;

[0032] The formula for predicting oil film failure time is as follows:

[0033] T f =-ln(E t / E0) / λ;

[0034] Among them, T f E represents the predicted oil film failure time. t E0 represents the critical lubrication efficiency threshold; E0 represents the initial lubrication efficiency; and λ represents the attenuation coefficient.

[0035] According to the above plan, the oil quality grades include the first grade range, the second grade range, and the third grade range;

[0036] The first level range is when all quality testing parameters are within the standard operating range; the second level range is when any quality testing parameter is close to the critical threshold; and the third level range is when any quality testing parameter exceeds the allowable range.

[0037] The formula for classifying oil quality grades is as follows:

[0038] G=round(∑(δ j ×R j )×3);

[0039] Where G represents the quality level; δ j R represents the weight of the j-th quality parameter; j This indicates that the j-th quality parameter is qualified; round() means rounding to the nearest integer.

[0040] According to the above scheme, step S400 includes:

[0041] S410. Compare the viscosity, moisture content, impurity particle size, and effective additive concentration in the lubricating oil quality test data with the preset oil quality thresholds respectively.

[0042] S420. When any detection parameter fails to reach the corresponding oil quality threshold, the oil replacement process is triggered, an oil replacement command is generated, and the lubrication and oil replenishment process is paused.

[0043] S430. When all detection parameters reach the oil quality threshold, obtain the lubrication demand index output in real time by the lubrication demand index model, and read the lubrication performance change trend reflected by the oil quality lubrication attenuation coupling curve.

[0044] S440. Based on the numerical range of the lubrication demand index and the predicted value of the remaining effective time of the oil film, dynamically adjust the sampling frequency of lubrication detection and the amount of lubricating oil supplied in a single oil replenishment operation.

[0045] The formula for the amount of lubricating oil supplied in a single oil replenishment operation is as follows:

[0046] V = V0 × (LI / LI0) × (E0 / E(t));

[0047] Where V represents the actual oil replenishment amount; V0 represents the baseline oil replenishment amount; LI represents the current lubrication demand index; LI0 represents the baseline lubrication demand index; E0 represents the initial lubrication efficiency; and E(t) represents the current lubrication efficiency.

[0048] The sampling frequency for lubrication testing is calculated using the following formula:

[0049] F=F0×(1+k×(1-E(t) / E0))×LI;

[0050] Where F represents the adjusted detection frequency; F0 represents the reference detection frequency; k represents the adjustment coefficient; and E(t) / E0 represents the relative lubrication efficiency.

[0051] The baseline parameter thresholds include the baseline oil replenishment amount, the baseline detection frequency, and the baseline lubrication demand index. These are determined by the statistical average of equipment operating stability data over historical lubrication cycles and are dynamically adjusted as the equipment's service life increases.

[0052] S450. When the lubrication demand index is in the high demand range, increase the detection frequency and oil replenishment amount proportionally; when the remaining effective time of the oil film is close to the warning threshold, activate the preventive oil replenishment mechanism.

[0053] According to the above scheme, the oil quality thresholds include the standard range of viscosity, the upper limit of water content, the limit of impurity particle size grade, and the minimum threshold of effective additive concentration.

[0054] The high demand range is defined as when the lubrication demand index exceeds the preset upper limit of the normal operating range;

[0055] The warning threshold is a preset oil film failure safety margin based on the equipment type;

[0056] The relevant thresholds for the lubrication demand index, including the upper limit of the normal working range and the upper limit of the preset threshold, are based on quantitative research on the lubrication demand of punch presses in the industry and are calibrated with the critical value of equipment wear risk under typical working conditions to ensure that the index thresholds can accurately reflect the urgency of lubrication demand.

[0057] The relevant thresholds for oil film failure, including the critical lubrication efficiency threshold, the oil film failure early warning time threshold, and the remaining effective safety margin of the oil film, are determined by referring to the oil film load-bearing capacity theory and fatigue failure model in lubrication engineering, combined with experimental data on the friction characteristics of the metal contact surface of the punch press.

[0058] According to the above scheme, step S500 includes:

[0059] When the lubrication demand index reaches the upper limit of the preset threshold, an immediate spraying command is generated; when the lubrication demand index is within the preset threshold range, the oil film failure time is predicted by combining the oil quality lubrication attenuation coupling curve, and if the predicted oil film failure time is less than the oil film failure warning time threshold, an early spraying command is generated; when the punch press completes the preset number of punchings or reaches the preset running time, if the lubrication demand index has not reached the threshold but the remaining effective time of the oil film is less than the safety margin, a timed spraying command is generated; when the emergency lubrication assessment mechanism is triggered, an emergency spraying command is directly generated, and the equipment operating parameters and oil quality data at the time of spraying are recorded simultaneously.

[0060] According to the above scheme, when the system receives multiple spraying instructions at the same time, only the instruction with the highest priority is executed, and the remaining instructions are automatically put into the execution queue. The instruction priorities from high to low are emergency spraying instructions, immediate spraying instructions, advance spraying instructions, and timed spraying instructions.

[0061] The equipment operating parameters during spraying include load intensity, pressing frequency, and temperature of key components;

[0062] Oil quality data includes the results of viscosity, moisture content, and particle size analysis.

[0063] The punch press equipment data analysis system based on multi-source data fusion includes: a data acquisition module, a lubrication demand module, an energy efficiency prediction module, a lubrication strategy module, and a spraying execution module.

[0064] The data acquisition module includes a real-time data acquisition module, a historical data storage module, and an oil quality testing module. The real-time data acquisition module is used to collect data on stroke count, load intensity, stamping frequency, temperature of key equipment components, ambient temperature and humidity, and material hardness. The historical data storage module is used to store historical oil replenishment amounts, historical lubrication cycles, and historical oil film failure records. The oil quality testing module is used to collect data on viscosity, moisture content, impurity particle size, and effective additive concentration before lubricating oil is sprayed.

[0065] The lubrication demand module includes an initial assessment module, a parameter correction module, and an anomaly response module. The initial assessment module establishes an initial lubrication demand assessment value based on real-time operating parameters. The parameter correction module performs weighted fusion correction on the initial assessment value using oil quality data. The anomaly response module adjusts the model parameters and triggers a secondary verification and early warning mechanism when anomalies in oil indicators are detected.

[0066] The energy efficiency prediction module includes a decay curve construction module and a failure trend prediction module. The decay curve construction module starts with oil quality data, configures differentiated decay coefficients according to quality grade, and generates a lubrication efficiency decay coupling curve. The failure trend prediction module outputs the remaining effective time of the oil film and the failure warning time through the decay curve.

[0067] The lubrication strategy module includes an oil quality judgment module, a parameter adjustment module, and a prevention mechanism module. The oil quality judgment module compares oil quality data with thresholds to trigger a replacement process or qualified oil processing logic. The parameter adjustment module dynamically adjusts the lubrication detection frequency and single oil replenishment amount based on the lubrication demand index and decay curve. The prevention mechanism module activates enhanced detection and oil replenishment mechanisms when the oil film is close to the warning threshold in the high demand range.

[0068] The spraying execution module includes an instruction production module and a priority scheduling module. The instruction production module generates emergency spraying instructions, immediate spraying instructions, advance spraying instructions, or timed spraying instructions based on the lubrication demand index, oil film failure prediction, and emergency mechanism. The priority scheduling module executes only the instruction with the highest priority when multiple instructions conflict, and the remaining instructions automatically enter the waiting queue.

[0069] Compared with the prior art, the beneficial effects of the present invention are:

[0070] 1. This invention integrates real-time operating parameters of the punch press, historical data, and oil quality data to construct a dynamic lubrication demand model and performance decay curve, thereby achieving quantitative prediction of lubrication demand and accurate prediction of oil film failure, avoiding the blindness of traditional fixed-cycle lubrication.

[0071] 2. This invention sets up a pre-detection and anomaly response mechanism for oil quality to prioritize the elimination of risks from substandard oil; and through multi-command priority scheduling and emergency spraying mechanisms, it ensures timely lubrication of equipment under high-risk operating conditions, reducing wear and the probability of failure.

[0072] 3. This invention combines historical data with real-time feedback to achieve strategy self-optimization, thereby improving the continuous operation capability and service life of punching equipment. Attached Figure Description

[0073] Figure 1 This is a flowchart illustrating the steps of the punching machine data analysis method based on multi-source data fusion according to the present invention.

[0074] Figure 2 This is a schematic diagram of the structure of the punching machine data analysis system based on multi-source data fusion according to the present invention. Detailed Implementation

[0075] 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.

[0076] Example: Figures 1-2 As shown, the present invention provides a technical solution, a data analysis method for punching equipment based on multi-source data fusion, the method comprising the following steps:

[0077] S100: Collect data from the punch press equipment, including real-time operating parameters of the punch press, historical lubrication data, and lubricating oil quality test data.

[0078] Specifically, the real-time operating parameters of the punch press include the number of strokes, load intensity, stamping frequency, temperature of key parts of the equipment, ambient temperature and humidity, and material hardness; historical lubrication data includes historical oil replenishment amount, historical lubrication cycle, and historical oil film failure records; and lubricating oil quality test data is the test data before the lubricating oil is sprayed, including viscosity, moisture content, impurity particle size, and effective concentration of additives.

[0079] For example: collecting three types of data from a punch press running continuously for 8 hours:

[0080] Real-time operating parameters:

[0081] Stroke count: 120 strokes / minute, P1=0.8, normalized value, full scale 150 strokes / minute; Load strength: 80kN, P2=0.7, rated load 100kN; Stamping frequency: 30 strokes / minute, P3=0.6, design limit 50 strokes / minute; Critical component temperature: 55℃, P4=0.5, safety upper limit 70℃; Ambient temperature and humidity: 25℃, 60%, P5=0.4; Material hardness: HB180, P6=0.3, processing material is ordinary carbon steel;

[0082] Historical lubrication data:

[0083] Historical oil replenishment volume: average 200mL per replenishment, 8 hours of lubrication; historical oil film failure record: no failures due to insufficient lubrication in the past 3 months;

[0084] Lubricating oil quality test data:

[0085] Pre-spray testing: Hydraulic oil, model X001; Viscosity: 48 mm 2 / s, Q1=0.1, standard range 40-50mm 2 / s; Moisture content: 0.03%, Q2=0.2, upper limit 0.05%; Impurity particle size: ISO18 / 15, Q3=0.1, grade limit 20 / 17; Additive effective concentration: 92%, Q4=0.05, minimum threshold 90%.

[0086] This is just an example and is not intended to be limiting.

[0087] S200: Construct a lubrication demand index model, using the real-time operating parameters of the punch press as initial state parameters and the lubricating oil quality test data as correction factors input into the model; when abnormal oil indicators are detected, the model parameters are automatically adjusted and an anomaly handling mechanism is triggered.

[0088] Specifically, step S200 includes:

[0089] S210. Using the number of strokes, load intensity, stamping frequency, temperature of key parts of the equipment, ambient temperature and humidity, and material hardness in the real-time operating parameters of the punch press as basic parameters, establish an initial lubrication requirement assessment value.

[0090] For example: Real-time parameter weighting coefficient ω i The parameters are as follows: number of strokes 0.2, load strength 0.25, stamping frequency 0.2, temperature 0.15, temperature and humidity 0.1, and material hardness 0.1.

[0091] The initial assessment value is 0.2×0.8+0.25×0.7+0.2×0.6+0.15×0.5+0.1×0.4+0.1×0.3=0.615;

[0092] S220: Using viscosity, moisture content, impurity particle size and effective additive concentration in lubricating oil quality test data as correction factors, the initial lubrication demand assessment value is corrected in real time through multi-parameter weighted fusion calculation.

[0093] The lubrication demand index is calculated using the following formula:

[0094] LI=∑(ω i ×P i )×∏(1+α j ×Q j );

[0095] Where LI represents the lubrication demand index; ω i P represents the weight coefficient of the i-th real-time running parameter; i α represents the normalized value of the i-th real-time running parameter; j Q represents the correction factor for the j-th lubricating oil quality parameter; j This represents the normalized deviation value of the j-th lubricating oil quality parameter;

[0096] For example: quality parameter correction factor α j All are 0.1;

[0097] The lubrication demand index LI = 0.615 × (1 + 0.1 × 0.1) × (1 + 0.1 × 0.2) × (1 + 0.1 × 0.1) × (1 + 0.1 × 0.05) = 0.615 × 1.01 × 1.02 × 1.01 × 1.005 ≈ 0.65;

[0098] S230. When the viscosity of the lubricating oil exceeds the preset operating parameter range of the equipment, the moisture content exceeds the specified safety threshold of the equipment, the particle size of the impurities reaches the preset pollution level, or the effective concentration of the additive is lower than the standard threshold, the lubrication demand index model automatically increases the weight of the corresponding abnormal indicators in the calculation of the lubrication demand index and triggers the abnormal handling mechanism. The abnormal handling mechanism includes starting the secondary verification process of lubricating oil quality and generating abnormal warning information.

[0099] Furthermore, the threshold values ​​related to the real-time operating parameters of the punch press, including the safe upper limit of temperature of key parts of the equipment, the rated range of load intensity and the design limit of punching frequency, are determined based on the hardware parameters of the equipment's motor power and mechanical structure strength to ensure that the operating parameters do not exceed the physical tolerance range of the equipment.

[0100] The basic quality thresholds for lubricating oil, including the standard viscosity range, upper limit of water content, limit of impurity particle size grade, and minimum effective concentration threshold of additives, are set according to the lubricating oil model and specifications specified in the equipment manual and the oil performance indicators recommended by the manufacturer, to ensure that the physical and chemical properties of the oil are suitable for the lubrication needs of the equipment.

[0101] The lubrication demand index model automatically increases the weighting of corresponding abnormal indicators in the lubrication demand index calculation, as shown in the following formula:

[0102] ω' j =ω j ×(1+β×|Q j |);

[0103] Where, ω' j denoted as the adjusted weighting coefficient; β represents the abnormal influence factor; |Qj| represents the absolute value of the deviation of the j-th parameter.

[0104] S300: Construct a lubrication attenuation coupling curve based on oil quality, using lubricating oil quality test data as the curve starting point parameter, and setting differentiated attenuation coefficients according to oil quality grade;

[0105] Specifically, step S300 includes:

[0106] S310. Using the viscosity, moisture content, impurity particle size, and effective additive concentration in the lubricating oil quality test data as initial input parameters, determine the initial state parameters of the lubrication attenuation coupling curve of the oil quality.

[0107] S320. Based on the various index values ​​of lubricating oil quality test data, the oil is divided into different oil quality grades. The oil quality grade is based on the viscosity compliance range, the safe range of water content, the degree of impurity particle size contamination, and the degree of deviation from the effective concentration threshold of additives.

[0108] S330: Different lubrication attenuation coefficients are configured for oils of different quality grades, and the attenuation coefficients are negatively correlated with the oil quality grade.

[0109] S340. Based on the initial state parameters and the attenuation coefficient, construct the oil quality lubrication attenuation coupling curve. The oil quality lubrication attenuation coupling curve is used to represent the law of change of lubrication efficiency of oil over time during the operation of the punch press.

[0110] Furthermore, the oil film thickness decay model is shown in the following formula:

[0111] E(t) = E0 × e (-λ×t) ×∑(γ k ×S k );

[0112] Where E(t) represents the lubrication efficiency at time t; E0 represents the initial lubrication efficiency; λ represents the attenuation coefficient; t represents the running time; γ k S is denoted as the influence coefficient of the k-th real-time operating parameter on lubrication efficiency; k It is represented as the normalized value of the k-th real-time running parameter;

[0113] Furthermore, the oil film failure time is predicted using the following formula:

[0114] T f =-ln(E t / E0) / λ;

[0115] Among them, T f E represents the predicted oil film failure time. t E0 represents the critical lubrication efficiency threshold; E0 represents the initial lubrication efficiency; and λ represents the attenuation coefficient.

[0116] Furthermore, the oil quality grades include a first grade range, a second grade range, and a third grade range; the first grade range is when all quality test parameters are within the standard operating range, the second grade range is when any quality test parameter is close to the critical threshold, and the third grade range is when any quality test parameter exceeds the allowable range.

[0117] Furthermore, the formula for classifying oil quality grades is as follows:

[0118] G=round(∑(δ j ×R j )×3);

[0119] Where G represents the quality level; δ j R represents the weight of the j-th quality parameter; j This indicates that the j-th quality parameter is qualified; round() means rounding to the nearest integer.

[0120] For example: initial lubrication efficiency E0=95, quality parameter qualification rate R j Viscosity 0.95, Moisture 0.9, Impurities 0.95, Additives 0.98, Weight δ j Both are 0.25;

[0121] Oil quality grade G = roundd(∑(δ) j ×R j )×3)=round((0.25×0.95+0.25×0.9+0.25×0.95+0.25×0.98)×3)=round(0.945×3)=3, corrected to the first level;

[0122] The degradation coefficient of Grade 1 oil is λ = 0.05 / h (the highest quality grade, with the slowest degradation).

[0123] Real-time parameter influence coefficient γ k All are 0.17, normalized value S k Same as P i ;

[0124] Lubrication efficiency formula E(t) = 95 × e (-0.05t) ×0.17×3.3≈95×e (-0.05t) ×0.561

[0125] After running for 4 hours, E(4) = 95 × e (-0.2) ×0.561≈95×0.819×0.561≈44.2.

[0126] S400 When the lubricating oil quality test data does not reach the oil quality threshold, the oil replacement process is triggered first. When the lubricating oil quality test data reaches the oil quality threshold, the lubrication test frequency and oil replenishment amount are dynamically adjusted based on the lubrication demand index calculated by the lubrication demand index model and the lubrication attenuation coupling curve of oil quality.

[0127] Specifically, step S400 includes:

[0128] S410. Compare the viscosity, moisture content, impurity particle size, and effective additive concentration in the lubricating oil quality test data with the preset oil quality thresholds respectively.

[0129] S420. When any detection parameter fails to reach the corresponding oil quality threshold, the oil replacement process is triggered, an oil replacement command is generated, and the lubrication and oil replenishment process is paused.

[0130] S430. When all detection parameters reach the oil quality threshold, obtain the lubrication demand index output in real time by the lubrication demand index model, and read the lubrication performance change trend reflected by the oil quality lubrication attenuation coupling curve.

[0131] For example: Currently, LI = 0.65, E(t) = 44.2;

[0132] The remaining effective time of the oil film, Tf, is approximately 1.15 / 0.05 = 23 hours.

[0133] S440. Based on the numerical range of the lubrication demand index and the predicted value of the remaining effective time of the oil film, dynamically adjust the sampling frequency of lubrication detection and the amount of lubricating oil supplied in a single oil replenishment operation.

[0134] The formula for the amount of lubricating oil supplied in a single oil replenishment operation is as follows:

[0135] V = V0 × (LI / LI0) × (E0 / E(t));

[0136] Where V represents the actual oil replenishment amount; V0 represents the baseline oil replenishment amount; LI represents the current lubrication demand index; LI0 represents the baseline lubrication demand index; E0 represents the initial lubrication efficiency; and E(t) represents the current lubrication efficiency.

[0137] The sampling frequency for lubrication testing is calculated using the following formula:

[0138] F=F0×(1+k×(1-E(t) / E0))×LI;

[0139] Where F represents the adjusted detection frequency; F0 represents the reference detection frequency; k represents the adjustment coefficient; and E(t) / E0 represents the relative lubrication efficiency.

[0140] The baseline parameter thresholds include the baseline oil replenishment amount, the baseline detection frequency, and the baseline lubrication demand index. These are determined by the statistical average of equipment operating stability data over historical lubrication cycles and are dynamically adjusted as the equipment's service life increases.

[0141] For example: Refueling amount: Baseline refueling amount V0=200mL, baseline index LI0=0.5, E0=95;

[0142] V=200×(0.65 / 0.5)×(95 / 44.2)=200×1.3×2.15≈559mL;

[0143] Detection frequency: Base frequency F0 = 1 time / 2 hours, adjustment coefficient k = 0.5;

[0144] F = 1 × (1 + 0.5 × (1 - 44.2 / 95)) × 0.65 ≈ 1 × (1 + 0.5 × 0.535) × 0.65 ≈ 1 × 1.267 × 0.65 ≈ 0.82 times / hour, which means approximately one test every 73 minutes;

[0145] S450. When the lubrication demand index is in the high demand range, increase the detection frequency and oil replenishment amount proportionally; when the remaining effective time of the oil film is close to the warning threshold, activate the preventive oil replenishment mechanism.

[0146] For example: If the current LI=0.65 < high demand range (LI>0.7), and the remaining oil film time is 23 hours > the warning threshold of 8 hours, the enhancement mechanism will not be activated for the time being.

[0147] Furthermore, the oil quality thresholds include the standard viscosity range, upper limit of moisture content, limit of impurity particle size grade, and minimum threshold of effective additive concentration; the high demand range is defined as the lubrication demand index exceeding the preset upper limit of the normal operating range; the warning threshold is the preset oil film failure safety margin based on the equipment type; the lubrication demand index-related thresholds, including the upper limit of the normal operating range and the upper limit of the preset threshold, are based on quantitative research on the lubrication demand of punch presses in the industry, combined with calibration of the critical value of equipment wear risk under typical working conditions, to ensure that the index thresholds can accurately reflect the urgency of lubrication demand; the oil film failure-related thresholds, including the critical lubrication efficiency threshold, the oil film failure warning time threshold, and the remaining effective safety margin of the oil film, are determined with reference to the oil film load-bearing capacity theory and fatigue failure model in lubrication engineering, combined with experimental data on the friction characteristics of the metal contact surface of the punch press.

[0148] The S500 generates instant, advance, timed, or emergency lubricating oil spraying commands based on the lubrication demand index model and the lubrication attenuation coupling curve of oil quality, and records the spraying data.

[0149] Specifically, step S500 includes:

[0150] When the lubrication demand index reaches the upper limit of the preset threshold, an immediate spraying command is generated; when the lubrication demand index is within the preset threshold range, the oil film failure time is predicted by combining the oil quality lubrication attenuation coupling curve, and if the predicted oil film failure time is less than the oil film failure warning time threshold, an early spraying command is generated; when the punch press completes the preset number of punchings or reaches the preset running time, if the lubrication demand index has not reached the threshold but the remaining effective time of the oil film is less than the safety margin, a timed spraying command is generated; when the emergency lubrication assessment mechanism is triggered, an emergency spraying command is directly generated, and the equipment operating parameters and oil quality data at the time of spraying are recorded simultaneously.

[0151] Furthermore, when the system receives multiple spraying commands simultaneously, only the command with the highest priority is executed, and the remaining commands automatically enter the execution queue. The command priorities from high to low are emergency spraying command, immediate spraying command, advance spraying command, and timed spraying command. The equipment operating parameters during spraying include load intensity, stamping frequency, and temperature of key parts. Oil quality data includes viscosity, moisture content, and impurity particle size detection results.

[0152] For example: the lubrication demand index LI=0.65 < the preset threshold upper limit of 0.9, and the immediate spraying condition is not met; the predicted oil film failure time of 23 hours is greater than the warning threshold of 8 hours, and the advance spraying condition is not met; the punch press has been running for 8 hours, the preset running time, the lubrication demand index has not reached the threshold, but the remaining effective time of the oil film of 23 hours is greater than the safety margin of 5 hours, so no timed instruction is generated for the time being; the emergency assessment mechanism has not been triggered, and there is no emergency instruction.

[0153] Final result: No spraying command needs to be generated at present. Follow the adjusted frequency, spraying once every 73 minutes, continuously monitor, and perform oil replenishment once the parameters meet the standards.

[0154] This invention provides another technical solution: a data analysis method for punch press equipment based on multi-source data fusion, in which the lubrication demand index of the punch press exceeds the upper limit of the threshold during continuous high-load operation.

[0155] Real-time operating parameters: Stroke count 140 times / minute, P1=0.93; Load strength 95kN, P2=0.95; Stamping frequency: 45 times / minute, P3=0.9; Critical component temperature 68℃, P4=0.97, other parameters close to full scale; Ambient temperature and humidity: 30℃, 65%, P5=0.60; Material hardness: HB200, P6=0.50, for processing high-strength steel;

[0156] Lubricating oil quality: viscosity 52mm 2 / s, Q1=0.4, exceeding the standard upper limit by 40mm 2 / s, moisture 0.045%, Q2=0.9, close to the upper limit.

[0157] Lubrication demand index (0-1.5):

[0158] The initial assessment value is 0.2×0.93+0.25×0.95+0.2×0.90+0.15×0.97+0.1×0.60+0.1×0.50=0.186+0.238+0.18+0.146+0.06+0.05=0.86;

[0159] After correction, LI = 0.86 × 1.04 × 1.09 × 1.03 × 1.02 ≈ 0.86 × 1.19 ≈ 1.02, which exceeds the preset threshold limit of 0.9;

[0160] Because LI=1.02≥0.9, an immediate spraying command is triggered. Simultaneously, the following data is recorded: load intensity 95kN, stamping frequency 45 times / minute, critical component temperature 68℃, and oil viscosity 52mm. 2 / s, moisture content 0.045%.

[0161] This invention provides another technical solution, which accelerates the decline in lubrication efficiency and predicts that the oil film failure time is lower than the warning threshold;

[0162] Initial lubrication efficiency E0=92, quality rating, oil grade 2, decay coefficient λ=0.08 / h;

[0163] Real-time parameter influence coefficient γ k All are 0.17, normalized value S k: 0.85, 0.75, 0.70, 0.60, 0.55 and 0.45;

[0164] The impact factor is 0.17 × (0.85 + 0.75 + 0.70 + 0.60 + 0.55 + 0.45) = 0.17 × 3.9 = 0.663;

[0165] After running for 7 hours, E(7) = 92 × e (-0.08×7) ×0.663=92×e (-0.56) ×0.663≈92×0.571×0.663≈92×0.379≈34.9;

[0166] Critical lubrication efficiency E t =30,T f =-ln(E t / E0) / λ=-ln(30 / 92) / 0.08≈(-ln0.326) / 0.08≈1.12 / 0.08=14 hours;

[0167] The oil film failure warning time threshold is 0.3 × T0, where T0 is the baseline oil film life of 40 hours and the threshold is 12 hours. The current predicted failure time is 14 hours < 12 hours. Due to continuous decay, the time will decrease after t=7.5 hours. f =11 hours, triggering the advance spraying command, with a planned replenishment of oil within 1 hour, and the replenishment amount calculated according to the formula as 620mL.

[0168] This invention provides another technical solution: lubricating oil is found to be contaminated with impurities, and the parameters exceed the safety threshold.

[0169] Lubricating oil quality test data: Viscosity: 58 mm 2 / s, exceeding the standard range by 40-50mm 2 / s; Moisture content: 0.07%, exceeding the upper limit by 0.05%; Impurity particle size: ISO22 / 19, exceeding the grade limit 20 / 17; Effective additive concentration: 85%, below the minimum threshold of 90%;

[0170] Four parameters were detected to be abnormal, with an abnormality impact factor β=2.0. The adjusted weight ω' j =ω j ×(1+β×|Q j |)(Q j (These are absolute deviations; viscosity Q1 = 0.8, moisture Q2 = 0.4, impurities Q3 = 1.0, additives Q4 = 0.5).

[0171] The revised lubrication demand index LI=1.3, triggering the emergency lubrication assessment mechanism;

[0172] The emergency spraying command is generated directly with the highest priority, interrupting the currently pending timed command and recording the following information simultaneously: load intensity of 85kN, stamping frequency of 30 times / minute, oil impurity particle size of 22 / 19, and moisture content of 0.07% during pollution.

[0173] This invention provides another technical solution: a punch press equipment data analysis system based on multi-source data fusion. The system includes: a data acquisition module, a lubrication demand module, an energy efficiency prediction module, a lubrication strategy module, and a spraying execution module.

[0174] The data acquisition module includes a real-time data acquisition module, a historical data storage module, and an oil quality testing module. The real-time data acquisition module is used to collect data on stroke count, load intensity, stamping frequency, temperature of key equipment components, ambient temperature and humidity, and material hardness. The historical data storage module is used to store historical oil replenishment amounts, historical lubrication cycles, and historical oil film failure records. The oil quality testing module is used to collect data on viscosity, moisture content, impurity particle size, and effective additive concentration before lubricating oil is sprayed.

[0175] The lubrication demand module includes an initial assessment module, a parameter correction module, and an anomaly response module. The initial assessment module establishes an initial lubrication demand assessment value based on real-time operating parameters. The parameter correction module performs weighted fusion correction on the initial assessment value using oil quality data. The anomaly response module adjusts the model parameters and triggers a secondary verification and early warning mechanism when anomalies in oil indicators are detected.

[0176] The energy efficiency prediction module includes a decay curve construction module and a failure trend prediction module. The decay curve construction module starts with oil quality data, configures differentiated decay coefficients according to quality grade, and generates a lubrication efficiency decay coupling curve. The failure trend prediction module outputs the remaining effective time of the oil film and the failure warning time through the decay curve.

[0177] The lubrication strategy module includes an oil quality judgment module, a parameter adjustment module, and a prevention mechanism module. The oil quality judgment module compares oil quality data with thresholds to trigger a replacement process or qualified oil processing logic. The parameter adjustment module dynamically adjusts the lubrication detection frequency and single oil replenishment amount based on the lubrication demand index and decay curve. The prevention mechanism module activates enhanced detection and oil replenishment mechanisms when the oil film is close to the warning threshold in the high demand range.

[0178] The spraying execution module includes an instruction production module and a priority scheduling module. The instruction production module generates emergency spraying instructions, immediate spraying instructions, advance spraying instructions, or timed spraying instructions based on the lubrication demand index, oil film failure prediction, and emergency mechanism. The priority scheduling module executes only the instruction with the highest priority when multiple instructions conflict, and the remaining instructions automatically enter the waiting queue.

[0179] 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. Punch press equipment data analysis method based on multi-source data fusion, characterized in that: The method comprises: S100, collecting punch equipment data, the punch equipment data comprising punch real-time running parameters, historical lubrication data and lubricating oil quality detection data; S200, constructing a lubrication demand index model, taking the punch real-time running parameters as initial state parameters and taking the lubricating oil quality detection data as a correction factor input model; when an oil product index anomaly is detected, automatically adjusting model parameters and triggering an abnormal processing mechanism; S300, constructing an oil product quality lubrication attenuation coupling curve, taking the lubricating oil quality detection data as a curve starting point parameter and setting a differential attenuation coefficient according to oil product quality grades; S310, taking viscosity, moisture content, impurity particle size and additive effective concentration in the lubricating oil quality detection data as initial input parameters to determine initial state parameters of the oil product quality lubrication attenuation coupling curve; S320, based on index values of the lubricating oil quality detection data, dividing oil products into different oil product quality grades, the oil product quality grade division being based on a viscosity standard interval, a moisture content safety range, an impurity particle size pollution grade and a deviation degree of an additive effective concentration threshold value; S330, configuring corresponding lubrication attenuation coefficients for oil products of different quality grades, the attenuation coefficients being in a negative correlation relationship with the oil product quality grades; S340, based on the initial state parameters and the attenuation coefficients, constructing an oil product quality lubrication attenuation coupling curve, the oil product quality lubrication attenuation coupling curve being used to represent a law of lubrication efficiency of oil products changing with time during punch operation; S400, when the lubricating oil quality detection data does not reach an oil product quality threshold value, preferentially triggering an oil product replacement process, when the lubricating oil quality detection data reaches the oil product quality threshold value, based on a lubrication demand index calculated by the lubrication demand index model and the oil product quality lubrication attenuation coupling curve, dynamically adjusting lubrication detection frequency and oil replenishment amount; S500, based on the lubrication demand index model and the oil product quality lubrication attenuation coupling curve, generating immediate, advance, timing or emergency lubricating oil spraying instructions and recording spraying data.

2. The punch equipment data analysis method based on multi-source data fusion according to claim 1, characterized in that: the punch real-time running parameters comprise stroke frequency, load intensity, punching frequency, equipment key position temperature, environmental temperature and humidity and material hardness; the historical lubrication data comprise historical oil replenishment amount, historical lubrication period and historical oil film failure record; and the lubricating oil quality detection data are detection data before lubricating oil spraying, comprising viscosity, moisture content, impurity particle size and additive effective concentration.

3. The punch press equipment data analysis method based on multi-source data fusion according to claim 2, characterized in that: Step S200 comprises: S210, taking stroke frequency, load intensity, punching frequency, equipment key position temperature, environmental temperature and humidity and material hardness in the punch real-time running parameters as basic parameters to establish an initial lubrication demand evaluation value; S220, taking viscosity, moisture content, impurity particle size and additive effective concentration in the lubricating oil quality detection data as correction factors to perform real-time correction on the initial lubrication demand evaluation value through multi-parameter weighted fusion calculation; S230. When the viscosity of the lubricating oil exceeds the preset operating parameter range of the equipment, the moisture content exceeds the specified safety threshold of the equipment, the particle size of the impurities reaches the preset pollution level, or the effective concentration of the additive is lower than the standard threshold, the lubrication demand index model automatically increases the weight of the corresponding abnormal indicators in the calculation of the lubrication demand index and triggers the abnormal handling mechanism. The abnormal handling mechanism includes starting the secondary verification process of lubricating oil quality and generating abnormal early warning information.

4. The data analysis method for punching equipment based on multi-source data fusion according to claim 1, characterized in that: The oil quality grades include a first grade range, a second grade range, and a third grade range; The first level range is when all quality testing parameters are within the standard operating range, the second level range is when any quality testing parameter is close to the critical threshold, and the third level range is when any quality testing parameter exceeds the allowable range.

5. The punch press equipment data analysis method based on multi-source data fusion according to claim 1, characterized in that: Step S400 includes: S410. The viscosity, moisture content, impurity particle size and effective concentration of additives in the lubricating oil quality test data are compared with preset oil quality thresholds respectively. S420. When any detection parameter fails to reach the corresponding oil quality threshold, the oil replacement process is triggered, an oil replacement command is generated, and the lubrication and oil replenishment process is paused. S430. When all detection parameters reach the oil quality threshold, obtain the lubrication demand index output in real time by the lubrication demand index model, and read the lubrication efficiency change trend reflected by the oil quality lubrication attenuation coupling curve. S440. Based on the numerical range of the lubrication demand index and the predicted value of the remaining effective time of the oil film, dynamically adjust the sampling frequency of lubrication detection and the amount of lubricating oil supplied in a single oil replenishment operation. S450. When the lubrication demand index is in the high demand range, increase the detection frequency and oil replenishment amount proportionally; when the remaining effective time of the oil film is close to the warning threshold, activate the preventive oil replenishment mechanism.

6. The data analysis method for punching equipment based on multi-source data fusion according to claim 5, characterized in that: The oil quality thresholds include the standard viscosity range, the upper limit of moisture content, the limit of impurity particle size grade, and the minimum threshold of effective additive concentration. The high demand range is defined as the lubrication demand index being greater than the preset upper limit of the normal working range; The warning threshold is a preset safe time margin for oil film failure based on the equipment type.

7. The punch press equipment data analysis method based on multi-source data fusion according to claim 1, characterized in that: Step S500 includes: When the lubrication demand index reaches the upper limit of the preset threshold, an immediate spraying command is generated; when the lubrication demand index is within the preset threshold range, the oil film failure time is predicted by combining the oil quality lubrication attenuation coupling curve, and if the predicted oil film failure time is less than the oil film failure warning time threshold, an early spraying command is generated; when the punch press completes the preset number of punchings or reaches the preset running time, if the lubrication demand index has not reached the threshold but the remaining effective time of the oil film is less than the safety margin, a timed spraying command is generated; when the emergency lubrication assessment mechanism is triggered, an emergency spraying command is directly generated, and the equipment operating parameters and oil quality data at the time of spraying are recorded simultaneously.

8. The punch equipment data analysis method based on multi-source data fusion according to claim 7, characterized in that: When the system simultaneously receives multiple spraying instructions, only the instruction with the highest current priority is executed, and the remaining instructions automatically enter the execution queue, with the instruction priority from high to low being emergency spraying instruction, immediate spraying instruction, advance spraying instruction, and timing spraying instruction; The equipment operating parameters at the time of spraying include load intensity, stamping frequency, and key position temperature; The oil quality data includes viscosity, moisture content, and impurity particle size detection results.

9. The punch press equipment data analysis system based on multi-source data fusion, applied to the punch press equipment data analysis method based on multi-source data fusion in any one of claims 1-8, characterized in that: The system includes a data acquisition module, a lubrication demand module, an energy efficiency prediction module, a lubrication strategy module, and a spraying execution module; The data acquisition module includes a real-time data acquisition module, a historical data storage module, and an oil quality detection module; the real-time data acquisition module is used to acquire stroke frequency, load intensity, stamping frequency, equipment key position temperature, environmental temperature and humidity, and material hardness; the historical data storage module is used to store historical oil replenishment amount, historical lubrication period, and historical oil film failure record; the oil quality detection module is used to acquire viscosity, moisture content, impurity particle size, and additive effective concentration before lubricating oil spraying; The lubrication demand module includes an initial evaluation module, a parameter correction module, and an abnormal response module; the initial evaluation module establishes an initial lubrication demand evaluation value based on real-time operating parameters; the parameter correction module corrects the initial evaluation value by weighted fusion of oil quality data; the abnormal response module adjusts model parameters when detecting oil product index abnormalities, triggering secondary verification and early warning mechanism; The energy efficiency prediction module includes a decay curve construction module and a failure trend prediction module; the decay curve construction module takes oil quality data as the starting point, configures differential decay coefficients according to quality grades, and generates a lubrication efficiency decay coupling curve; the failure trend prediction module outputs oil film remaining effective time and failure warning time through the decay curve; The lubrication strategy module includes an oil product qualification judgment module, a parameter adjustment module, and a prevention mechanism module; the oil product qualification judgment module compares oil quality data with threshold values, triggering replacement process or qualified oil product processing logic; the parameter adjustment module dynamically adjusts lubrication detection frequency and single oil replenishment amount in combination with lubrication demand index and decay curve; the prevention mechanism module starts intensive detection and oil replenishment mechanism when in high demand interval or when oil film is close to warning threshold; The spraying execution module includes an instruction production module and a priority scheduling module; the instruction production module generates emergency spraying instruction, immediate spraying instruction, advance spraying instruction, or timing spraying instruction based on lubrication demand index, oil film failure prediction, and emergency mechanism; the priority scheduling module executes only the instruction with the highest current priority when multiple instructions conflict, and the remaining instructions automatically enter the execution queue.

Images