CNC machining equipment lubricating oil supply intelligent control method and system based on AI

Abstract

The invention discloses an AI-based CNC machining equipment lubricating oil supply intelligent control method and system, and relates to the technical field of lubricating oil control. The current lubricating oil supply mode is adjusted by judging the pressure grade of each lubricating point; when the adjusting effect is temporarily invalid, the lubricating oil supply weight of each lubricating point is adjusted again, and corresponding state data are obtained for the lubricating points with the adjusted weights to judge the blocking state of the lubricating points; when the lubricating points are in the blocked state, lubricating oil supply of the blocked lubricating points is stopped, and the lubricating oil supply amount is controlled according to the impact degree of sealing rings of the other lubricating points and the oil film covering uniformity and integrity; according to the method, whether the lubricating points are blocked or not can be judged according to actual conditions, the influence on the production progress of CNC equipment is reduced, when the lubricating points are blocked, lubricating oil supply of all the lubricating points can be reasonably controlled according to the actual conditions, lubricating medium waste and secondary faults are reduced, and the service life of the CNC equipment is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of lubricating oil control, and in particular to an AI-based intelligent control method and system for lubricating oil supply for CNC machining equipment. Background Art

[0002] In the context of intelligent manufacturing upgrades, AI-based lubricant supply control technology for CNC machining equipment is gradually replacing traditional static control modes. Traditional methods rely on timing or quantitative rules. While they can guarantee basic lubrication needs, they are prone to excessive losses or insufficient lubrication due to rigid oil supply strategies when faced with dynamic machining scenarios (such as high-speed cutting and variable load conditions). New AI-driven control solutions integrate multi-source sensor data (pressure, temperature, vibration, etc.) and combine deep learning algorithms to build lubrication demand prediction models, enabling precise "on-demand" oil supply control. For example, time-series neural networks are used to dynamically optimize the oil supply cycle and flow ratio to improve energy efficiency. However, existing research on lubricant supply systems primarily focuses on optimizing steady-state operating conditions, and there is still a lack of systematic solutions for real-time fault-tolerant control of sudden mechanical failures (such as valve blockage and pipe blockage).

[0003] In actual production, if the oil pressure at a certain lubrication point is too high, it may be due to a physical failure or blockage factor. For example, the distribution valve is stuck in the closed state due to the intrusion of metal debris, resulting in the oil circuit of the lubrication point being blocked; it may also be due to excessive oil supply, such as unreasonable oil supply weight distribution. If the oil supply is blindly stopped when the pressure at the lubrication point is too high, it may delay the working progress of the CNC equipment. If a lubrication point is blocked and the oil supply to the remaining lubrication points cannot be reasonably controlled, it may cause other lubrication points on the CNC equipment to be over-pressured (exceeding the set amount), which not only causes waste of lubricating medium, but may also lead to secondary failures such as overload failure of the seal ring of the corresponding lubrication point and rupture of the oil film of the precision guide rail, thereby reducing the service life of the CNC equipment. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide an AI-based intelligent control method and system for lubricating oil supply of CNC machining equipment.

[0005] In a first aspect of the present invention, an AI-based intelligent control method for lubricating oil supply of CNC machining equipment is first proposed, the method comprising: Obtaining the pressure of each lubrication point to determine the pressure level of the lubrication point, wherein the pressure level includes a normal level and a warning level; When the pressure level is at the warning level, the current oil supply mode of the lubricating oil is adjusted; and the pressure data of the lubricating point after the adjustment is obtained, and the adjustment effect is determined based on the pressure data of the lubricating point after the adjustment; the adjustment effect includes effective and temporarily ineffective; When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; When a lubrication point is in a blocked state, an alarm is issued to stop the lubricating oil supply to the blocked lubrication point, and the lubricating oil supply to the remaining lubrication points is controlled according to the impact degree of the seal rings and the uniformity and completeness of the oil film coverage.

[0006] Optionally, the pressure of each lubrication point is obtained to determine the pressure level of the lubrication point, where the pressure level includes a normal level and a warning level; The specific steps are: Set a sliding time window and calculate the average pressure within the sliding time window for each lubrication point and standard deviation ; Set the warning threshold. The warning threshold is If the pressure exceeds the warning threshold for three consecutive times in the sliding time window, the pressure level of the lubrication point will be set to the warning level; otherwise, the pressure level of the lubrication point will be set to the normal level.

[0007] Optionally, the step of judging the adjustment effect according to the pressure data of the lubrication point after adjustment is: When the pressure level of the lubrication point is at the warning level, the current working mode of the CNC processing equipment is obtained, and the optimal oil supply mode when the oil supply pressure at the lubrication point reaches the warning level under the same working mode is extracted from the historical processing data of the CNC processing equipment. The lubricating oil supply to the lubrication point is adjusted according to the optimal oil supply mode; Obtain the pressure data of each lubrication point after adjusting the oil supply mode, and calculate the initial pressure drop rate and steady-state pressure value of the pressure data. If the initial pressure drop rate and the steady-state pressure value both meet the preset conditions, it means that the adjustment effect is effective, and lubricating oil continues to be supplied to the lubrication point based on the adjusted oil supply mode; otherwise, it means that the adjustment effect is temporarily invalid.

[0008] Optionally, the steps of re-adjusting the lubricating oil supply weight of each lubrication point are: Obtaining pressure data of each lubrication point after adjusting the oil supply mode, constructing a pressure dynamic evolution diagram sequence, and determining the impact value of each lubrication point based on the pressure dynamic evolution diagram sequence; The influence value at each lubrication point is multiplied by the initial weight, and the reciprocal of the multiplication result is used to adjust the lubricating oil supply weight of each lubrication point.

[0009] Optionally, the steps of obtaining pressure data of each lubrication point after adjustment by adjusting the oil supply mode, constructing a pressure dynamic evolution diagram, and determining the impact value of each lubrication point according to the pressure dynamic evolution diagram are: The pressure of each lubrication point is taken as a node in the dynamic evolution graph, and the edge is constructed based on the time-delayed pressure response: if at time t, the lubrication point The pressure change in Lubrication points caused by time The pressure changes synchronously, then arrive Add directed edges; Obtain the pressure sequences of the two nodes corresponding to the directed edge respectively, calculate the similarity of the pressure sequences of the two nodes using the DTW algorithm, and use the similarity as the weight value of the directed edge; Construct dynamic evolution graphs of multiple time periods to form a dynamic evolution graph sequence; Calculate the total weight of each node in each dynamic evolution graph in the dynamic evolution graph sequence, and calculate the ratio of the total weight of each node to the total weight of all nodes in the dynamic evolution graph, as the influence ratio of the corresponding node in the dynamic evolution graph; Eliminate the maximum influence ratio and the minimum influence ratio of each node in the dynamic evolution graph sequence, and calculate the mean of the remaining influence ratios as the influence value of each lubrication point.

[0010] Optionally, the step of obtaining corresponding status data to determine the blockage status of the lubrication point is: If the pressure of the lubrication point returns to the preset pressure after the lubrication oil supply weight of each lubrication point is readjusted, the lubrication point is in a normal state, and oil is supplied to the lubrication point based on the adjusted lubrication oil supply weight of each lubrication point; If the pressure of the lubrication point does not return to the preset pressure, the lubrication point is in a blocked state and an alarm is issued. The lubricating oil supply weight corresponding to the blocked lubrication point is recorded as 0, and the overall lubricating oil supply is controlled according to the lubricating oil supply weights of the remaining lubrication points and the current impact degree of the sealing ring and the uniformity and completeness of the oil film coverage.

[0011] Optionally, the steps of controlling the lubricating oil supply according to the impact degree of the sealing rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage are as follows: In the preset time period after the lubricating oil supply weight of each lubrication point is readjusted, the preset time period is divided into several unit time windows, the vibration amplitude of the sealing ring in each unit time window is obtained, and the instantaneous impact factor of the sealing ring of each lubrication point in each unit time window is calculated based on the vibration amplitude. The calculation steps of the instantaneous impact factor are as follows: subtract the minimum vibration amplitude acceleration from the maximum vibration amplitude acceleration in the unit time window, and divide the subtraction result by the unit time window length; Calculate the mean instantaneous impact factor of all unit time windows within the preset time period , take the absolute value of the impact factor difference of each adjacent unit time window to obtain the impact change rate, and take the average value to obtain the impact change rate; The proportional weights of the instantaneous impact factor mean and the impact change rate are both set to 0.5, and are multiplied by the corresponding proportional weights to calculate the impact degree of the sealing ring; Calculate the uniformity and completeness of the oil film coverage, and control the overall lubricating oil supply according to the impact degree of the seals of other lubrication points and the uniformity and completeness of the oil film coverage.

[0012] Optionally, the steps for calculating the uniformity and completeness of the oil film coverage are: After re-adjusting the lubricating oil supply weight of each lubrication point, the preset time period is divided into several unit time windows, and the multi-point sampling values of the oil film thickness of the lubrication point in each unit time window are obtained. The mean and standard deviation of the oil film thickness in each unit time window are calculated, and the oil film uniformity factor is calculated. , the calculation formula is: , where and

[0013] Compare the oil film thickness in each unit time window with the preset minimum oil film thickness corresponding to the friction risk, and record the proportion of oil film thickness less than the preset minimum oil film thickness as the oil film rupture rate; The oil film coverage uniformity and integrity are calculated based on the oil film uniformity factor and the oil film rupture rate. The calculation formula is: , where For the uniformity and completeness of oil film coverage, is the number of unit time windows; Indicates the sequence number of the unit time window, and Respectively represent The oil film uniformity factor and oil film rupture rate per unit time window, Indicates the preset weight factor, with a value between 0 and 1.

[0014] In a second aspect of the present invention, an AI-based intelligent control system for lubricating oil supply of CNC machining equipment is proposed, the system comprising: Adjustment module: When the pressure level is at the warning level, the current lubricating oil supply mode is adjusted; the pressure data of the lubrication point after adjustment is obtained, and the adjustment effect is determined based on the pressure data of the lubrication point after adjustment; the adjustment effect includes effective and temporarily ineffective; Blockage judgment module: When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to judge the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; Oil supply control module: When a lubrication point is in a blocked state, an alarm is issued to stop the lubrication oil supply to the blocked lubrication point, and the lubrication oil supply is controlled according to the impact degree of the seal rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage.

[0015] Beneficial effects of the present invention: The present invention proposes an AI-based intelligent control method and system for lubricating oil supply for CNC machining equipment. By determining the pressure level of each lubrication point, the current lubricating oil supply mode is adjusted. The pressure data of the lubrication point after adjustment is obtained, and the adjustment effect is determined based on the pressure data of the lubrication point after adjustment. When the adjustment effect is temporarily ineffective, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage status and the normal status; when the lubrication point is in the blockage state, an alarm is issued, the lubricating oil supply to the blocked lubrication point is stopped, and the lubricating oil supply is controlled according to the impact degree of the sealing ring of the remaining lubrication points and the uniformity and completeness of the oil film coverage; in this way, whether the lubrication point is blocked can be judged according to the actual situation, reducing the impact on the production progress of the CNC equipment, and when the lubrication point is blocked, the lubricating oil supply to all lubrication points can be reasonably controlled according to the actual situation, reducing the waste of lubricating medium and the occurrence of secondary failures, and increasing the service life of the CNC equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 Flowchart of the AI-based intelligent control method for lubricant supply to CNC machining equipment; Figure 2 Framework diagram of the AI-based intelligent control system for lubricant supply to CNC machining equipment. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] The embodiment of the present invention provides an AI-based intelligent control method for lubricating oil supply of CNC machining equipment. Figure 1 , Figure 1 Flowchart of an AI-based intelligent control method for lubricating oil supply to CNC machining equipment provided by an embodiment of the present invention. The method comprises the following steps: Obtain the pressure of each lubrication point to determine the pressure level of the lubrication point. The pressure level includes normal level and warning level; When the pressure level is at the warning level, the current lubricating oil supply mode is adjusted; the pressure data of the lubrication point after adjustment is obtained, and the adjustment effect is judged based on the pressure data of the lubrication point after adjustment; the adjustment effect includes effective and temporarily ineffective; When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; When a lubrication point is in a blocked state, an alarm is issued to stop the lubricating oil supply to the blocked lubrication point, and the lubricating oil supply is controlled according to the impact degree of the sealing rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage.

[0020] Based on the AI-based intelligent control method for lubricating oil supply for CNC machining equipment provided by the embodiment of the present invention, through the above-mentioned method, it is possible to judge whether a lubrication point is clogged according to the actual situation, thereby reducing the impact on the production progress of the CNC equipment. When a lubrication point is clogged, the lubricating oil supply to all lubrication points can be reasonably controlled according to the actual situation, thereby reducing the waste of lubricating medium and the occurrence of secondary failures, and extending the service life of the CNC equipment.

[0021] In one embodiment, the pressure of each lubrication point is obtained to determine the pressure level of the lubrication point, where the pressure level includes a normal level and a warning level; The specific steps are: Set a sliding time window and calculate the average pressure within the sliding time window for each lubrication point and standard deviation ; Set the warning threshold. The warning threshold is If the pressure exceeds the warning threshold for three consecutive times in the sliding time window, the pressure level of the lubrication point will be set to the warning level; otherwise, the pressure level of the lubrication point will be set to the normal level.

[0022] It's important to note that in the lubrication system of CNC machining equipment, a statistical method based on the mean and standard deviation of pressure within a sliding time window is introduced to determine whether lubrication point pressure is abnormal. The warning threshold is set at the mean + 2 times the standard deviation. This approach fully accounts for the normal pressure fluctuation range caused by operating conditions (such as speed and temperature fluctuations) during normal machining, rather than simply relying on a fixed threshold. This method, based on the statistical principle that "99.7% of the data are within ±3σ of the mean, and more than 95% of the data are within ±2σ," can adaptively identify potential abnormal pressure signals that exceed the normal fluctuation range under dynamic conditions. Furthermore, the requirement that "three consecutive sampling points exceed the warning threshold" is set to improve the stability and reliability of the judgment and prevent transient anomalies caused by short-term noise, electromagnetic interference, or occasional sensor errors from being misidentified as true faults. For example, if the average pressure data collected at a lubrication point over the last 10 seconds is 2.5 bar with a standard deviation of 0.2 bar, the warning threshold is 2.5 + 2 × 0.2 = 2.9 bar. If the subsequent three consecutive sampling results are 3.1, 3.2, and 3.3 bar, all exceeding 2.9 bar, it means that the pressure of the lubrication point has continuously deviated from the normal range, has certain trend and risk accumulation characteristics, and should be judged as "warning level".

[0023] In one implementation, this method takes into account both real-time performance and avoids frequent false alarms, and is an important basis for intelligent lubrication control under complex and changing working conditions.

[0024] In one embodiment, when the pressure level is at the warning level, the frequency and duration of lubricating oil supply are adjusted, and pressure data of the lubrication point after adjustment is obtained. The adjustment effect is determined based on the pressure data of the lubrication point after adjustment. The adjustment effect includes the steps of valid and invalid. When the pressure level of the lubrication point is at the warning level, the current working mode of the CNC processing equipment is obtained, and the optimal oil supply mode when the oil supply pressure at the lubrication point reaches the warning level under the same working mode is extracted from the historical processing data of the CNC processing equipment. The lubricating oil supply to the lubrication point is adjusted according to the optimal oil supply mode; Obtain the pressure data of each lubrication point after adjusting the oil supply mode, and calculate the initial pressure drop rate and steady-state pressure value of the pressure data. If the initial pressure drop rate and the steady-state pressure value both meet the preset conditions, it means that the adjustment effect is effective, and lubricating oil continues to be supplied to the lubrication point based on the adjusted oil supply mode; otherwise, it means that the adjustment effect is temporarily invalid.

[0025] It's important to note that in the intelligent control of lubrication systems for CNC machining equipment, to ensure rapid restoration of lubrication system stability in the event of a warning, the oil supply strategy must be dynamically adjusted based on the current equipment operating mode and historical optimization data. The "operating mode" of CNC machining equipment refers to the operating characteristics of the equipment when performing different machining tasks. These can generally be categorized as follows: high-speed cutting mode (such as high-speed milling and drilling), heavy-duty cutting mode (such as high-feed roughing), precision machining mode (such as fine milling and grinding), low-speed idling mode (such as machine startup or standby), and multi-axis linkage mode (such as complex cavity machining). Each mode has different requirements for lubrication rhythm, pressure, and flow. For example, in high-speed cutting mode, due to high spindle speeds and high frictional heat, lubricant not only needs to be delivered promptly but also requires good cooling and flow characteristics. Therefore, the oil supply frequency and oil volume per unit time are generally higher. In precision machining mode, on the other hand, greater emphasis is placed on oil film stability and minimizing oil pressure fluctuations.

[0026] The oil supply mode refers to the strategic configuration of the lubrication system to perform lubrication tasks within a specific operating mode. It mainly includes the following aspects: oil supply frequency (such as once or five times per minute), single oil supply duration (such as 0.5 seconds or 2 seconds per oil supply), oil flow rate per unit time (such as 30ml or 50ml pumping per minute), etc. For example, in heavy-duty cutting mode, if a lubrication point enters the warning state, the best oil supply strategy for similar situations in this mode can be extracted from historical processing records (such as an oil supply frequency of once every 10 seconds, a single oil supply of 1 second, a total flow rate of 60ml / min, and a pressure of 2.8 bar) and applied to make adjustments. The pressure changes at the adjusted lubrication points are then monitored. If the initial pressure drop is rapid (for example, a drop of more than 0.3 bar within 2 seconds) and the steady-state pressure value returns to the normal range, the lubricant has successfully reached the target location, the blockage trend has been alleviated, and the adjustment can be determined to be effective. Conversely, if the pressure drop is not significant, it means that the oil supply strategy has failed to alleviate the blockage or the system still has an anomaly, requiring further response measures. This intelligent control mechanism can significantly improve the adaptability and response efficiency of the lubrication system and reduce the incidence of failures.

[0027] In one embodiment, when the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; Specifically, the steps for readjusting the lubricating oil supply weight of each lubrication point are as follows: Obtaining pressure data of each lubrication point after adjusting the oil supply mode, constructing a pressure dynamic evolution diagram sequence, and determining the impact value of each lubrication point based on the pressure dynamic evolution diagram sequence; Multiply the influence value of each lubrication point by the initial weight, and adjust the lubricating oil supply weight of each lubrication point by the inverse of the multiplication result; In one implementation, when adjustments are temporarily ineffective, rather than directly determining a lubrication point blockage and issuing an alarm, the system first re-adjusts the lubrication supply weights for each lubrication point and then proceeds with lubrication supply. This is primarily due to the fact that abnormal lubrication point pressure does not necessarily equate to actual blockage; temporary response deviations may occur due to temporary load fluctuations, oil supply delays, or disturbances in the surrounding system. Immediately declaring a blockage and issuing an alarm when pressure has barely recovered can easily lead to false alarms, disrupting equipment operation and affecting production continuity. Therefore, the system adopts a more robust judgment approach: analyzing the response trends of each lubrication point to oil supply disturbances by constructing a sequence of dynamic pressure evolution graphs. Based on this, the system re-adjusts the weights and implements more targeted oil supply allocation. This not only further identifies true abnormal points but also mitigates critical conditions where lubrication points may be "on the verge of blockage" but not yet completely inoperative by optimizing the oil supply ratio, thus providing the equipment with an opportunity to "self-repair." This approach improves the system's fault tolerance, reduces unnecessary false alarms, and improves the efficiency of lubrication resource utilization. It also delays or avoids lubrication system maintenance downtime due to misjudgments, ensuring continuous and efficient operation of CNC equipment. This strategy also embodies the AI control system's intelligent "adapt first, then judge" approach, ensuring more accurate judgments and gentler responses.

[0028] In one embodiment, the steps of obtaining pressure data of each lubrication point after adjusting the oil supply mode, constructing a pressure dynamic evolution diagram, and determining the impact value of each lubrication point based on the pressure dynamic evolution diagram are as follows: The pressure of each lubrication point is taken as a node in the dynamic evolution graph, and the edge is constructed based on the time-delayed pressure response: if at time t, the lubrication point The pressure change in Lubrication points caused by time The pressure of the arrive Add directed edges; Obtain the pressure sequences of the two nodes corresponding to the directed edge respectively, calculate the similarity of the pressure sequences of the two nodes using the DTW algorithm, and use the similarity as the weight value of the directed edge; Construct dynamic evolution graphs of multiple time periods to form a dynamic evolution graph sequence; Calculate the total weight of each node in each dynamic evolution graph in the dynamic evolution graph sequence, and calculate the ratio of the total weight of each node to the total weight of all nodes in the dynamic evolution graph, as the influence ratio of the corresponding node in the dynamic evolution graph; Eliminate the maximum influence ratio and the minimum influence ratio of each node in the dynamic evolution graph sequence, and calculate the mean of the remaining influence ratios as the influence value of each lubrication point.

[0029] For example, suppose lubrication point A is in a warning state. Even after adjusting its oil supply pattern, the pressure at point A still hasn't returned to normal. In this case, we can't simply conclude that point A is blocked. This is because the lubrication system is a coupled, complex network with flow interference and pressure transmission relationships between lubrication points. The high pressure at point A might not be solely due to a blockage in its own oil circuit. It could also be due to pressure changes at other lubrication points (such as B, C, and D), which are causing a cascading interference on point A. For example, a sudden or sustained high pressure at an upstream lubrication point could be transmitted through the pipeline to point A, causing a localized pressure buildup there, making it impossible to recover with a single-point adjustment. To further confirm whether point A is truly blocked, a global perspective is needed to redistribute the oil supply weights for all lubrication points. Specifically, based on the time-series pressure data for all lubrication points in the system, a series of dynamic pressure evolution graphs is constructed over multiple time periods. The pressure response relationships between lubrication points at different time points are analyzed. The similarity of pressure changes between lubrication points is calculated using the DTW algorithm to determine the influence value represented by each node in the evolution graph. If point A has a high influence value, it indicates that it carries a significant load on pressure transmission within the system or is easily affected by other parts of the system. Based on this, the original weight is multiplied by the influence value, and the inverse of the product is taken to determine the new lubrication point weight, which is then proportionally distributed to the total system oil supply. For example, if point A's original oil supply weight is 20% and its influence value is 0.4, the adjusted weight becomes 1 / (20% × 0.4), reducing its overall weight relative to other lubrication points, thereby reducing the oil supply load on point A. Other lubrication points, such as B, C, D, and E, will then adjust their oil supply accordingly based on their respective influence values. After the system has run for a period of time, monitor the pressure at point A again. If the pressure at point A drops significantly, the previous high pressure was due to system interference, not blockage. Conversely, if the pressure at point A remains abnormal after the global adjustment, it is generally considered to be blocked. This is because even if the oil supply is reduced and the system interference is alleviated, it will still not be able to absorb oil. This only occurs when there is an obstruction in the lubrication path.

[0030] In one implementation method, the core significance of a global oil supply weight redistribution is to eliminate interference factors through systematic adjustments, so as to more accurately determine whether the pressure abnormality at point A is caused by physical blockage.

[0031] Therefore, specifically, the corresponding status data is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state: If the pressure of the lubrication point returns to the preset pressure after the lubrication oil supply weight of each lubrication point is readjusted, the lubrication point is in a normal state, and oil is supplied to the lubrication point based on the adjusted lubrication oil supply weight of each lubrication point; If the pressure of the lubrication point does not return to the preset pressure, the lubrication point is in a blocked state and an alarm is issued. The lubricating oil supply weight corresponding to the blocked lubrication point is recorded as 0, and the overall lubricating oil supply is controlled according to the lubricating oil supply weights of the remaining lubrication points and the current impact degree of the sealing ring and the uniformity and completeness of the oil film coverage.

[0032] It should be noted that after readjusting the lubricating oil supply weights of each lubrication point and completing the global redistribution, the system will continue to monitor the pressure status of each lubrication point. If the pressure of a lubrication point (such as point A) gradually recovers to the preset safety pressure range (such as dropping from 0.82MPa to 0.68MPa) as the oil supply runs after redistribution, it is judged that the lubrication point has not actually been blocked and is in a normal state, indicating that the previous abnormal pressure may have been caused by non-blocking factors such as system linkage, oil flow disturbance or high-pressure conduction of other lubrication points. At this time, the system can adjust the pressure according to the currently optimized oil supply weights. The system will continue to supply oil to each lubrication point without additional intervention. On the contrary, if the pressure of a lubrication point (such as point A) is still high (for example, still 0.81MPa) after the global oil supply adjustment, it means that it is highly suspected of physical blockage, such as sediment blockage in the oil circuit, valve port adhesion, lubrication channel deformation, etc. At this time, the system will determine that it is in a blocked state and immediately issue an early warning signal to alert the operation and maintenance personnel, and force the corresponding oil supply weight to 0, completely cutting off the oil supply to prevent the system from continuing to supply oil to the blocked point, causing lubricating oil backlog or further pressure increase, which may cause safety hazards.

[0033] At the same time, the system determines whether to adjust the overall oil supply based on the current oil supply weight status of the remaining lubrication points, combined with the impact degree to which the seal ring is subjected (such as mechanical vibration and load impact during operation) and the uniformity and completeness of the oil film coverage (to determine whether lubrication is continuous and sufficient). This ensures that the remaining normal lubrication points maintain good oil supply, avoids lubrication imbalance of the entire system caused by blockage of a certain node, and does not cause overpressure at the remaining lubrication points, thereby effectively improving the stability and adaptability of the lubrication system.

[0034] In one implementation, the benefit of this method is that it not only has the ability to intelligently identify and isolate faulty lubrication points, but also can optimize and reconstruct oil supply resources at the system level, enhance the lubrication system's fault tolerance and emergency response efficiency to blockage faults, significantly improve the safety and lubrication accuracy of system operation, extend equipment service life and reduce maintenance costs.

[0035] In one embodiment, the steps of controlling the lubricating oil supply according to the impact degree of the seal rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage are as follows: In the preset time period after the lubricating oil supply weight of each lubrication point is readjusted, the preset time period is divided into several unit time windows, the vibration amplitude of the sealing ring in each unit time window is obtained, and the instantaneous impact factor of the sealing ring of each lubrication point in each unit time window is calculated based on the vibration amplitude. The calculation steps of the instantaneous impact factor are as follows: subtract the minimum vibration amplitude acceleration from the maximum vibration amplitude acceleration in the unit time window, and divide the subtraction result by the unit time window length; The specific calculation formula is: , where is the unit time window length, For the The maximum vibration amplitude acceleration per unit time window, For the The minimum vibration amplitude acceleration per unit time window; Calculate the mean instantaneous impact factor of all unit time windows within the preset time period , take the absolute value of the impact factor difference of each adjacent unit time window to obtain the impact change rate, and take the average value to obtain the impact change rate; the calculation formula is: , where is the shock change rate, is the number of unit time windows; The proportional weights of the instantaneous impact factor mean and the impact change rate are both set to 0.5, and multiplied by the corresponding proportional weights respectively to calculate the impact degree of the sealing ring; the calculation formula is: Where, Impact strength of the sealing ring.

[0036] Calculate the uniformity and completeness of the oil film coverage, and control the overall lubricating oil supply according to the impact degree of the seals of other lubrication points and the uniformity and completeness of the oil film coverage.

[0037] It's important to note that seal ring impact resistance measures the severity of mechanical impact to a lubrication point during oil supply, thus providing a basis for dynamically adjusting lubricant supply. Specifically, seal ring impact resistance reflects the severity and instability of the vibration amplitude changes at the lubrication point over a given period of time. A higher value indicates a harsher environment and more unstable operating conditions, making it more difficult for lubricant to adhere or more susceptible to being thrown off. Therefore, it's not advisable to continue supplying large amounts of lubricant to avoid wasting resources and potentially causing thermal damage or seal failure due to oil film rupture.

[0038] The advantages of this calculation method in one implementation are: First, it integrates information on both impact intensity and volatility, providing a more comprehensive picture of the seal's operating conditions. Second, the required data (i.e., maximum and minimum acceleration) is derived from common triaxial vibration sensors, offering real-time performance, simplified data acquisition, and suitability for online monitoring systems. By quantifying this impact intensity, oil supply can be proactively reduced when a lubrication point detects adverse operating conditions, thereby avoiding lubricant waste or exacerbating equipment damage and achieving a more precise, energy-efficient, and long-term lubrication management strategy. In a system with multiple lubrication points (e.g., A, B, and C), if oil supply to point A is interrupted due to a blockage, the system dynamically adjusts the oil supply weighting based on the impact intensity and oil film coverage integrity coefficient of points B and C, increasing or decreasing their respective weights to achieve oil redistribution and maintain optimal lubrication system performance.

[0039] In one embodiment, the steps for calculating the uniformity and completeness of the oil film coverage are: After re-adjusting the lubricating oil supply weight of each lubrication point, the preset time period is divided into several unit time windows, and the multi-point sampling values of the oil film thickness of the lubrication point in each unit time window are obtained. The mean and standard deviation of the oil film thickness in each unit time window are calculated, and the oil film uniformity factor is calculated. , the calculation formula is: , where and are the standard deviation and mean, respectively; Compare the oil film thickness in each unit time window with the preset minimum oil film thickness corresponding to the friction risk, and record the proportion of oil film thickness less than the preset minimum oil film thickness as the oil film rupture rate; The oil film coverage uniformity and integrity are calculated based on the oil film uniformity factor and the oil film rupture rate. The calculation formula is: , where For the uniformity and completeness of oil film coverage, is the number of unit time windows; Indicates the sequence number of the unit time window, and Respectively represent The oil film uniformity factor and oil film rupture rate per unit time window, Indicates the preset weight factor, with a value between 0 and 1.

[0040] It's important to note that the film coverage uniformity and integrity calculation measures the uniformity and integrity of the lubricant film distribution at a lubrication point during operation. Specifically, it comprehensively assesses whether the oil film thickness distribution on the lubrication point surface is uniform (i.e., whether the differences between different measurement points are minimal) and whether there is widespread oil film breakdown (i.e., the proportion of areas with thickness below a preset safety threshold). A higher value indicates a more uniform oil film distribution and improved film integrity at the lubrication point, indicating that additional lubrication oil replenishment is no longer necessary. Conversely, increased oil supply should be considered to repair localized weak areas.

[0041] In one implementation, this calculation method offers several advantages: First, it integrates static distribution (uniformity) and dynamic safety (rupture rate) metrics to provide a more comprehensive and stable criterion for determining oil film status. Second, by using relative and proportional indicators such as standard deviation and rupture rate, it offers adaptability and versatility to diverse lubrication conditions. Third, sampled data can be acquired and updated online via a sensor array, making it suitable for deployment in embedded or intelligent lubrication systems. Overall, this method not only helps promptly identify oil film degradation trends and reduce the risk of friction damage, but also provides a basis for subsequent adjustments to lubricant levels.

[0042] In one embodiment, the steps of controlling the lubricating oil supply according to the impact degree of the seal rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage are as follows: Compare the seal impact degree of the remaining lubrication points with the critical impact degree, and compare the oil film coverage uniformity and integrity with the preset minimum coverage uniformity and integrity. If the seal impact degree is less than the critical impact degree but the oil film coverage uniformity and integrity is less than the preset minimum coverage uniformity and integrity, increase the lubricating oil supply weight of the corresponding lubrication point and appropriately increase the lubricating oil supply; If the impact degree of the sealing ring is less than the critical impact degree and the oil film coverage uniformity and integrity is not less than the preset minimum coverage uniformity and integrity, the oil supply weight of the corresponding lubrication point will not be changed; the current lubricating oil supply amount will not be changed; If the impact degree of the sealing ring is not less than the critical impact degree, and the oil film coverage uniformity and integrity is not less than the preset minimum coverage uniformity and integrity, the oil supply weight of the corresponding lubrication point is reduced, and the lubricating oil supply is appropriately reduced; If the impact degree of the sealing ring is not less than the critical impact degree, but the oil film coverage uniformity and integrity is less than the preset minimum coverage uniformity and integrity, the lubricating oil supply to the corresponding lubrication point will be stopped and an alarm will be issued.

[0043] It should be noted that, assuming that lubrication point A in the current system has stopped supplying oil due to a blockage, only points B and C are considered. If the impact resistance of the seal ring at point B is below the critical value, it indicates stable operation without strong vibration, but its oil film coverage uniformity and integrity are below the minimum standard. This means that although the current operation is stable, the lubrication film is poorly distributed and there is a risk of localized wear. In this case, the system should moderately increase the oil supply weight of point B, for example, from the original 0.3 to 0.35, and increase the total supply accordingly to compensate.

[0044] For example, if the impact degree and oil film coverage of point C are both at normal levels, that is, the impact degree is low and the oil film integrity is good, the system does not need to adjust the oil supply weight of point C, and continues to maintain the status quo without changing the overall oil supply.

[0045] If it is found that the impact degree of a lubrication point (such as point D) is higher than the critical value, it means that its operating state has become unstable, but the oil film coverage still meets the standard, indicating that the lubrication is saturated but cannot alleviate the impact. At this time, the oil supply to this point should be appropriately reduced to avoid over-lubrication causing oil temperature increase or energy waste.

[0046] The most serious situation is: a certain point (such as point E) has both large sealing ring impact fluctuations and incomplete oil film, which means that this point may face the dual risks of severe wear and lubrication failure. The system should immediately suspend oil supply and trigger the alarm mechanism to prompt the operation and maintenance personnel to conduct key inspections and emergency maintenance.

[0047] One way to achieve this is through the above-mentioned refined regulation, which can not only achieve on-demand distribution of lubricating oil among multiple nodes, improve lubrication efficiency and mechanical life, but also serve as an auxiliary mechanism for early fault warning, providing technical support for the intelligent operation and maintenance of industrial equipment.

[0048] Based on the same inventive concept, the present invention also provides an AI-based intelligent control system for lubricating oil supply of CNC processing equipment. Figure 2 , Figure 2 This is a framework diagram of an AI-based intelligent control system for lubricating oil supply for CNC machining equipment provided in an embodiment of the present invention. The system includes: Pressure level judgment module: obtains the pressure of each lubrication point to judge the pressure level of the lubrication point. The pressure level includes normal level and warning level; Adjustment module: When the pressure level is at the warning level, the current lubricating oil supply mode is adjusted; the pressure data of the lubrication point after adjustment is obtained, and the adjustment effect is judged based on the pressure data of the lubrication point after adjustment; the adjustment effect includes effective and temporarily ineffective; Blockage judgment module: When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the adjustment is obtained to judge the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; Oil supply control module: When a lubrication point is in a blocked state, an alarm is issued to stop the lubrication oil supply to the blocked lubrication point, and the lubrication oil supply is controlled according to the impact degree of the seal rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage.

[0049] Based on the AI-based intelligent control method for lubricating oil supply for CNC machining equipment provided by the embodiment of the present invention, through the above-mentioned method, it is possible to judge whether a lubrication point is clogged according to the actual situation, thereby reducing the impact on the production progress of the CNC equipment. When a lubrication point is clogged, the lubricating oil supply to all lubrication points can be reasonably controlled according to the actual situation, thereby reducing the waste of lubricating medium and the occurrence of secondary failures, and extending the service life of the CNC equipment.

[0050] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An AI-based intelligent control method for lubricating oil supply of CNC machining equipment, characterized in that: The following steps are involved: Obtaining the pressure of each lubrication point to determine the pressure level of the lubrication point, wherein the pressure level includes a normal level and a warning level; When the pressure level is at the warning level, the current oil supply mode of the lubricating oil is adjusted; and the pressure data of the lubricating point after the adjustment is obtained, and the adjustment effect is determined based on the pressure data of the lubricating point after the adjustment; the adjustment effect includes effective and temporarily ineffective; When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to determine the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; When a lubrication point is in a blocked state, an alarm is issued to stop the lubricating oil supply to the blocked lubrication point, and the lubricating oil supply to the remaining lubrication points is controlled according to the impact degree of the seal rings and the uniformity and completeness of the oil film coverage.

2. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 1 is characterized in that: Obtaining the pressure of each lubrication point to determine the pressure level of the lubrication point, wherein the pressure level includes a normal level and a warning level; The specific steps are: Set a sliding time window and calculate the average pressure within the sliding time window for each lubrication point and standard deviation ; Set the warning threshold. The warning threshold is ; If the pressure exceeds the warning threshold for three consecutive times in the sliding time window, the pressure level of the lubrication point will be set to the warning level; otherwise, the pressure level of the lubrication point will be set to the normal level.

3. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 1 is characterized in that: The steps to judge the adjustment effect based on the pressure data of the lubrication point after adjustment are as follows: When the pressure level of the lubrication point is at the warning level, the current working mode of the CNC processing equipment is obtained, and the optimal oil supply mode when the oil supply pressure at the lubrication point reaches the warning level under the same working mode is extracted from the historical processing data of the CNC processing equipment. The lubricating oil supply to the lubrication point is adjusted according to the optimal oil supply mode; Obtain the pressure data of each lubrication point after adjusting the oil supply mode, and calculate the initial pressure drop rate and steady-state pressure value of the pressure data. If the initial pressure drop rate and steady-state pressure value both meet the preset conditions, it means that the adjustment effect is effective, and lubricating oil continues to be supplied to the lubrication point based on the adjusted oil supply mode; Otherwise, it means that the adjustment effect is temporarily invalid.

4. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 1 is characterized in that: The steps to readjust the lubricating oil supply weight of each lubrication point are: Obtaining pressure data of each lubrication point after adjusting the oil supply mode, constructing a pressure dynamic evolution diagram sequence, and determining the impact value of each lubrication point based on the pressure dynamic evolution diagram sequence; The influence value at each lubrication point is multiplied by the initial weight, and the reciprocal of the multiplication result is used to adjust the lubricating oil supply weight of each lubrication point.

5. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 4 is characterized in that: The steps of obtaining the pressure data of each lubrication point after adjusting the oil supply mode, constructing a pressure dynamic evolution diagram, and determining the impact value of each lubrication point based on the pressure dynamic evolution diagram are as follows: The pressure of each lubrication point is taken as a node in the dynamic evolution graph, and the edge is constructed based on the time-delayed pressure response: if at time t, the lubrication point The pressure change in Lubrication points caused by time The pressure changes synchronously, then arrive Add directed edges; Obtain the pressure sequences of the two nodes corresponding to the directed edge respectively, calculate the similarity of the pressure sequences of the two nodes using the DTW algorithm, and use the similarity as the weight value of the directed edge; Construct dynamic evolution graphs of multiple time periods to form a dynamic evolution graph sequence; Calculate the total weight of each node in each dynamic evolution graph in the dynamic evolution graph sequence, and calculate the ratio of the total weight of each node to the total weight of all nodes in the dynamic evolution graph, as the influence ratio of the corresponding node in the dynamic evolution graph; Eliminate the maximum influence ratio and the minimum influence ratio of each node in the dynamic evolution graph sequence, and calculate the mean of the remaining influence ratios as the influence value of each lubrication point.

6. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 1 is characterized in that: The steps to obtain the corresponding status data to determine the blockage status of the lubrication point are: If the pressure of the lubrication point returns to the preset pressure after the lubrication oil supply weight of each lubrication point is readjusted, the lubrication point is in a normal state, and oil is supplied to the lubrication point based on the adjusted lubrication oil supply weight of each lubrication point; If the pressure of the lubrication point does not return to the preset pressure, the lubrication point is in a blocked state and an alarm is issued. The lubricating oil supply weight corresponding to the blocked lubrication point is recorded as 0, and the overall lubricating oil supply is controlled according to the lubricating oil supply weights of the remaining lubrication points and the current impact degree of the sealing ring and the uniformity and completeness of the oil film coverage.

7. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 1 is characterized in that: The steps to control the lubricating oil supply according to the impact degree of the seal rings of other lubrication points and the uniformity and completeness of the oil film coverage are as follows: In the preset time period after the lubricating oil supply weight of each lubrication point is readjusted, the preset time period is divided into several unit time windows, the vibration amplitude of the sealing ring in each unit time window is obtained, and the instantaneous impact factor of the sealing ring of each lubrication point in each unit time window is calculated based on the vibration amplitude. The calculation steps of the instantaneous impact factor are as follows: subtract the minimum vibration amplitude acceleration from the maximum vibration amplitude acceleration in the unit time window, and divide the subtraction result by the unit time window length; Calculate the mean instantaneous impact factor of all unit time windows within the preset time period , take the absolute value of the impact factor difference of each adjacent unit time window to obtain the impact change rate, and take the average value to obtain the impact change rate; The proportional weights of the instantaneous impact factor mean and the impact change rate are both set to 0.5, and are multiplied by the corresponding proportional weights to calculate the impact degree of the sealing ring; Calculate the uniformity and completeness of the oil film coverage, and control the overall lubricating oil supply according to the impact degree of the seals of other lubrication points and the uniformity and completeness of the oil film coverage.

8. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 7 is characterized in that: The steps to calculate the uniformity and completeness of oil film coverage are: After re-adjusting the lubricating oil supply weight of each lubrication point, the preset time period is divided into several unit time windows, and the multi-point sampling values of the oil film thickness of the lubrication point in each unit time window are obtained. The mean and standard deviation of the oil film thickness in each unit time window are calculated, and the oil film uniformity factor is calculated. , the calculation formula is: , where and are the standard deviation and mean, respectively; Compare the oil film thickness in each unit time window with the preset minimum oil film thickness corresponding to the friction risk, and record the proportion of oil film thickness less than the preset minimum oil film thickness as the oil film rupture rate; The oil film coverage uniformity and integrity are calculated based on the oil film uniformity factor and the oil film rupture rate. The calculation formula is: , where For the uniformity and completeness of oil film coverage, is the number of unit time windows; Indicates the sequence number of the unit time window, and Respectively represent The oil film uniformity factor and oil film rupture rate per unit time window, Indicates the preset weight factor, with a value between 0 and 1.

9. The AI-based intelligent control method for lubricating oil supply of CNC machining equipment according to claim 7 is characterized in that: The steps to control the lubricating oil supply according to the impact degree of the seal rings of other lubrication points and the uniformity and completeness of the oil film coverage are as follows: Compare the seal impact degree of the remaining lubrication points with the critical impact degree, and compare the oil film coverage uniformity and integrity with the preset minimum coverage uniformity and integrity. If the seal impact degree is less than the critical impact degree but the oil film coverage uniformity and integrity is less than the preset minimum coverage uniformity and integrity, increase the lubricating oil supply weight of the corresponding lubrication point; If the impact degree of the sealing ring is less than the critical impact degree and the oil film coverage uniformity and integrity is not less than the preset minimum coverage uniformity and integrity, the oil supply weight of the corresponding lubrication point will not be changed; If the impact degree of the sealing ring is not less than the critical impact degree, and the oil film coverage uniformity and integrity is not less than the preset minimum coverage uniformity and integrity, the oil supply weight of the corresponding lubrication point is reduced; If the impact degree of the sealing ring is not less than the critical impact degree, but the oil film coverage uniformity and integrity is less than the preset minimum coverage uniformity and integrity, the lubricating oil supply to the corresponding lubrication point will be stopped and an alarm will be issued.

10. An AI-based intelligent control system for lubricating oil supply of CNC processing equipment, used to implement the AI-based intelligent control method for lubricating oil supply of CNC processing equipment according to any one of claims 1 to 9, characterized in that: The system comprises: Pressure level judgment module: obtains the pressure of each lubrication point to judge the pressure level of the lubrication point, and the pressure level includes normal level and warning level; Adjustment module: When the pressure level is at the warning level, the current lubricating oil supply mode is adjusted; the pressure data of the lubrication point after adjustment is obtained, and the adjustment effect is determined based on the pressure data of the lubrication point after adjustment; the adjustment effect includes effective and temporarily ineffective; Blockage judgment module: When the adjustment effect is temporarily invalid, the lubricating oil supply weight of each lubrication point is readjusted, and the corresponding status data of the lubrication point after the weight adjustment is obtained to judge the blockage status of the lubrication point; the blockage status includes the actual blockage state and the normal state; Oil supply control module: When a lubrication point is in a blocked state, an alarm is issued to stop the lubrication oil supply to the blocked lubrication point, and the lubrication oil supply is controlled according to the impact degree of the seal rings of the remaining lubrication points and the uniformity and completeness of the oil film coverage.

Images