High-low pressure switching control method and system for thin oil station of mill

Abstract

The invention discloses a mill thin oil station high-low pressure switching control method and system, and relates to the technical field of industrial automation control. The method comprises the steps that a state curve is determined by monitoring the mechanical state of a mill; triggering a reflecting layer port to execute dual-order judgment to trigger a target switching condition; a switching strategy is determined through a switching controller according to the target switching condition, and the mill numerical control center drives to execute switching between the high-pressure system and the low-pressure system; and pre-stamping, valve body synchronization, flow transfer, steady-state adjustment and post-steady-state compensation are taken as a decision chain. The technical problems that the lubrication effect and the operation stability of the mill are affected and the industrial real-time accurate control requirement in the grinding scene cannot be met due to the fact that the dynamic state data of the mill during operation are difficult to accurately capture and the switching condition is difficult to judge by a traditional control means are solved, and the purposes of improving the high-low pressure switching opportunity and mode matching degree and improving the control accuracy are achieved. And the lubricating effect and the operation stability of the mill are optimized, and the technical effect of industrial real-time precise control requirements in a grinding scene is met.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, and in particular to a high-low pressure switching control method and system for a mill thin oil station. Background Technology

[0002] During the grinding process, the stable operation of the mill is crucial to production efficiency and equipment lifespan. Real-time industrial control is key to ensuring precise regulation of mill operating conditions. The high-low pressure switching control of the mill's thin oil station is crucial to the adaptability of the lubrication system to the mill's operating state. Current technologies often rely on conventional control logic and simple monitoring methods to manage high-low pressure switching. While these methods are effective under relatively stable conditions, as industrial production demands higher precision and stability from mills, their limitations become apparent when applied to the complex and variable actual operating conditions of mills. Due to the dynamic changes in bearing clearance friction and oil circuit characteristics during mill operation, traditional control methods struggle to accurately capture multi-dimensional state data and intelligently determine switching conditions. This results in poor matching between the timing and mode of high-low pressure switching, affecting the mill's lubrication effect and operational stability, and failing to meet the demands of real-time, precise industrial control. Summary of the Invention

[0003] This application provides a high-low pressure switching control method and system for a mill thin oil station, which is used to solve the technical problem that traditional control methods are difficult to accurately capture dynamic state data of the mill during operation and determine switching conditions, thus affecting the mill's lubrication effect and operational stability, and failing to meet the industrial real-time and precise control requirements in grinding scenarios.

[0004] The first aspect of this application provides a high-low pressure switching control method for a mill thin oil station. The method includes: determining a state curve by monitoring the mill's mechanical condition, wherein the monitored elements include at least bearing clearance friction and oil circuit characteristics; triggering a reflection layer port based on the state curve, performing a two-stage judgment based on state transition probability and curve slope, and triggering a target switching condition, wherein the target switching condition includes a switching mode and a switching state, the switching mode being a quasi-switching mode or a safe conservative mode, and the switching state being a high-pressure-low-pressure or low-pressure-high-pressure switching; performing directional triggering and switching drive decision-making of the switching controller based on the target switching condition, determining a switching strategy, and driving the switching strategy to perform switching control management between the high-pressure system and the low-pressure system according to the mill's numerical control center; wherein the decision chain is pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation.

[0005] A second aspect of this application provides a high-low pressure switching control system for a mill thin oil station. The system includes: a state curve acquisition module, used to determine a state curve by monitoring the mechanical state of the mill, wherein the monitored elements include at least bearing clearance friction and oil circuit characteristics; a target switching condition triggering module, used to trigger the reflective layer port according to the state curve, perform a two-stage judgment based on state transition probability and curve slope, and trigger a target switching condition, wherein the target switching condition includes a switching mode and a switching state, the switching mode being a quasi-switching mode or a safe conservative mode, and the switching state being a high-pressure-low-pressure or low-pressure-high-pressure switching; a switching strategy acquisition module, used to execute directional triggering and switching drive decisions of the switching controller according to the target switching condition, determine a switching strategy, and drive the switching strategy to perform switching control management between the high-pressure system and the low-pressure system according to the mill CNC center; and a decision chain acquisition module, used to establish a decision chain of pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application achieves precise switching between high and low pressure systems in the mill's thin oil station by monitoring the mill's operating status from multiple dimensions, constructing a state curve, and determining the state based on historical data and real-time slope via the reflective layer port. Driven by a dual-branch, fifth-order decision logic, the switching controller ensures stable operation of the mill in grinding scenarios, meets the requirements of industrial real-time control for the adaptability and accuracy of the lubrication system, and improves the timing and mode matching of high and low pressure switching, optimizes the mill's lubrication effect and operational stability, and meets the technical requirements of industrial real-time precise control in grinding scenarios. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 This is a flowchart illustrating the high-low pressure switching control method for the mill thin oil station provided in this application embodiment.

[0009] Figure 2 This is a schematic diagram of the high-low pressure switching control system for the mill thin oil station provided in the embodiments of this application.

[0010] Figure labeling: Module 1 for state curve acquisition, Module 2 for target switching condition triggering, Module 3 for switching strategy acquisition, and Module 4 for decision chain acquisition. Detailed Implementation

[0011] This application provides a high-low pressure switching control method and system for a mill thin oil station, which is used to solve the technical problem that traditional control methods are difficult to accurately capture dynamic state data of the mill during operation and determine switching conditions, thus affecting the mill's lubrication effect and operational stability, and failing to meet the industrial real-time and precise control requirements in grinding scenarios.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, a high-low pressure switching control method for a mill thin oil station includes: Step A100: Determine the state curve by performing mechanical condition monitoring of the mill, wherein the monitoring elements include at least bearing clearance friction and oil circuit characteristics.

[0015] In this embodiment, the oil circuit features include oil film thickness and oil viscosity.

[0016] Specifically, based on the monitoring elements including bearing clearance friction and oil circuit characteristics, multi-state axial direction is determined to construct a state coordinate system. Multi-position monitoring is triggered when the mill starts. The monitoring data is integrated in the state coordinate system and the state curve is updated synchronously in time sequence. The specific steps are explained in detail in A110-A30.

[0017] Step A200: Based on the state curve, trigger the reflective layer port, perform a two-stage determination based on the state transition probability and the curve slope, and trigger the target switching condition. The target switching condition includes a switching mode and a switching state. The switching mode is a quasi-switching mode or a safe and conservative mode, and the switching state is a high-voltage-low-voltage or low-voltage-high-voltage switching.

[0018] In this embodiment, the reflective layer port is constructed by retrieving historical switching records, mining a Markov state chain based on the state transition probability driven by high and low voltage, and then performing judgment condition transformations based on this state chain. This port is embedded in the mill CNC center. The state transition probability is based on high and low voltage drive and reflects the probability value of the mill's operating state transitioning from one node to another.

[0019] Optionally, during mill operation, the status curve is continuously updated with monitoring data, forming a continuous time-series status record. When the target switching condition is triggered, the sampling interval is first set according to the mill's task conditions. Then, the front-end sensor array is set according to this sampling interval to drive the automated acquisition of monitoring elements, coordinate arrangement, and updating of the status curve. The specific steps are explained in detail in A210-A220.

[0020] Next, the reflection layer port is triggered to locate the upper node of the updated state curve and the lower node that serves as the real-time update node. Then, the state transition probability is determined according to the second judgment condition. If the preset probability is not met, the handover decision is terminated; if it is met, the target handover state and mode are determined and added to the target handover conditions. The specific steps are explained in detail in A230-A270.

[0021] By using a two-stage determination based on state transition probability and curve slope, the target switching conditions, which include switching modes and switching states, are accurately triggered, providing an accurate decision-making basis for high and low pressure switching of the mill thin oil station.

[0022] Step A300: Based on the target switching conditions, execute the directional triggering and switching drive decision of the switching controller, determine the switching strategy, and drive the switching strategy to perform switching control management between the high-pressure system and the low-pressure system according to the mill CNC center.

[0023] In this embodiment, the mill CNC center is a center that embeds a reflective layer port and a switching controller, and is used to drive the switching strategy to execute the switching control management between the high-voltage system and the low-voltage system.

[0024] In one embodiment of this application, firstly, the target response branch is activated according to the target switching mode, and then the high-low voltage switching logic decision is made based on the branch guided by the target switching state, thereby determining the switching strategy. The specific steps are described in detail in A340-A350.

[0025] After determining the switching strategy, the mill CNC center first receives the specific execution parameters included in the strategy. These parameters cover five stages: pre-pressurization, valve synchronization, flow transfer, steady-state regulation, and post-steady-state compensation. Examples include the target pre-pressurization pressure value, valve action time difference, flow transfer rate range, steady-state pressure fluctuation threshold, and post-steady-state compensation duration. Since both the reflector port and the switching controller are embedded in the CNC center, this tight integration allows the CNC center to quickly parse the logical instructions in the strategy, avoiding data transmission delays that could affect switching timeliness.

[0026] Subsequently, the CNC center sends drive signals to the switching controller based on the analysis results. The switching controller coordinates the actions of each component through dual-channel interaction with the high-pressure and low-pressure systems. Taking low-pressure-high-pressure switching as an example, in the pre-pressurization stage, the CNC center drives the switching controller to activate the high-pressure oil pump, which raises the output pressure to the preset 15MPa within 100ms. At the same time, the high-pressure filter filters oil impurities to prevent particles from entering the high-pressure pipeline. In the valve synchronization stage, the high-pressure control valve is controlled to open from the closed state to 80% within 30ms, while the low-pressure system valve is simultaneously closed from the open state to 20%, ensuring no instantaneous flow interruption during oil circuit switching. In the flow transfer stage, according to the rate requirement of 5L / s in the strategy, the ratio of the high-pressure oil pump speed to the low-pressure oil pump speed is adjusted to ensure a smooth transfer of oil flow from the low-pressure system to the high-pressure system. At this time, the flow sensor in the high-pressure pipeline feeds back data to the CNC center in real time to dynamically correct rate deviations.

[0027] Upon entering the steady-state adjustment phase, the CNC center monitors changes in oil film thickness and bearing clearance friction via a front-end sensor array. If the oil film thickness stabilizes at 0.08mm ± 0.005mm, the bearing clearance friction remains below 0.02mm, and the high-pressure system pressure fluctuation is less than 0.5MPa, then steady-state is considered achieved. If a deviation occurs, the drive switching controller fine-tunes the opening of the high-pressure control valve within ±5% until the parameters meet the standards. In the subsequent steady-state compensation phase, the CNC center maintains the current high-pressure system operating parameters for 3 seconds to ensure the oil film structure stabilizes before terminating the switching drive, completing the switching management from low pressure to high pressure.

[0028] For high-pressure to low-pressure switching, the process is similar but in the opposite direction: during the pre-pressurization stage, the output pressure of the high-pressure oil pump is reduced to 5MPa; during the valve body synchronization stage, the high-pressure control valve is closed and the low-pressure valve is opened; during the flow transfer stage, the flow of the high-pressure system is reduced and the flow of the low-pressure system is increased; finally, the pressure is stabilized at 3MPa±0.3MPa through steady-state regulation; and the subsequent steady-state compensation also lasts for 3 seconds to consolidate the low-pressure lubrication state.

[0029] By analyzing and driving the switching strategy through the mill CNC center, and combining the interaction between the switching controller and the high and low pressure systems, precise switching management between the high-pressure system and the low-pressure system, which includes components such as high-pressure oil pumps, filters, pipelines, and valves, is achieved according to the preset strategy, ensuring that the mill can obtain appropriate lubrication pressure under different operating conditions.

[0030] Step A400: The decision chain is pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation.

[0031] Specifically, pre-pressurization is an operation that applies a certain pressure to the target system (such as a high-pressure system switching from low pressure to high pressure) before the high-low pressure switch begins. Its purpose is to eliminate air gaps within the system, pre-fill the oil circuits with fluid and establish initial pressure, avoiding shocks caused by sudden pressure changes during the switch. For example, when switching to a high-pressure system, pre-pressurization will pre-raise the pressure in the high-pressure pipeline to 60%-70% of the target pressure. This provides a stable pressure base for subsequent valve operation and flow transfer, reducing pressure fluctuations during the switchover process.

[0032] Valve synchronization refers to the process where, after pre-pressurization, the control valves of the high and low pressure systems operate simultaneously according to a preset timing sequence. For example, when switching from low pressure to high pressure, the valves in the low-pressure system gradually close, while the valves in the high-pressure system simultaneously and gradually open, maintaining synchronicity in their actions. Its function is to prevent instantaneous interruption or overload of the oil circuit caused by unidirectional valve operation, achieving a smooth transition of the oil circuit and preventing oil shock or flow interruption caused by asynchronous valve operation, thus protecting mill components from damage.

[0033] Flow transfer is the process of gradually transferring oil flow from the current system to the target system based on the synchronous operation of the valve body. By adjusting the oil pump speed or valve opening, the flow is transferred from the original system to the target system at a preset rate (such as increasing or decreasing the flow rate by a certain value per second). Its function is to achieve dynamic distribution of lubrication flow to meet the flow requirements of the target system. The effect is to ensure the continuity of key parameters such as oil film thickness and oil viscosity during the switching process, and to avoid lubrication failure caused by sudden changes in flow.

[0034] Steady-state regulation is a step in fine-tuning the pressure, flow rate, and other parameters of the target system after flow transfer has been completed. By monitoring factors such as bearing clearance friction and oil film thickness in real time, the output power of the oil pump or the valve opening is dynamically corrected to stabilize the system parameters within the target range. Its function is to bring the system to the ideal operating state after the switch, ensuring that the mill obtains stable and suitable lubrication conditions under the target pressure mode, thus reducing mechanical wear.

[0035] Post-steady-state compensation is a step that maintains the target parameter state and makes minor compensation adjustments after steady-state adjustment. It typically lasts for a period of time, such as 3-5 seconds, during which system parameters are continuously monitored. If minor fluctuations occur, such as pressure fluctuations within ±0.2 MPa, immediate fine-tuning is performed. Its purpose is to consolidate the stable state after switching and prevent parameter rebound caused by system inertia or external disturbances. The effect is to further improve stability after switching, ensure the mill operates under reliable lubrication conditions for a long time, and extend the service life of the equipment.

[0036] Furthermore, step A100 in the method provided in this application embodiment includes: A110: The oil circuit characteristics include at least oil film thickness and oil viscosity.

[0037] A120: Based on the monitoring elements, determine the multi-state axes and construct a state coordinate system.

[0038] A130: Upon startup of the mill, multi-position monitoring based on monitoring elements is triggered, and the monitoring is integrated in the state coordinate system to synchronize the state curve, wherein the state curve is updated in time sequence with the monitoring data.

[0039] Specifically, during mill operation, the key elements that need to be monitored are first identified. These include oil circuit characteristics such as oil film thickness and oil viscosity, which, combined with bearing clearance friction, constitute a complete monitoring element system. Based on these three types of monitoring elements, corresponding multi-state axes are determined. For example, bearing clearance friction is the first axis x, oil film thickness is the second axis y, and oil viscosity is the third axis z. This constructs a three-dimensional state coordinate system, with each axis corresponding to the numerical range of a monitoring element, thereby achieving a multi-dimensional quantitative characterization of the mill's operating state.

[0040] The moment the mill starts, the system automatically activates a preset multi-position monitoring mechanism. At this time, sensors deployed at key parts of the mill bearings begin to collect bearing clearance friction data in real time. Simultaneously, dedicated sensors installed in the oil circuit system detect oil film thickness and oil viscosity. The data acquisition processes for these three monitoring elements are completely parallel, ensuring the synchronization of information acquisition. For example, 10 seconds after startup, the sensors collect data showing a bearing clearance friction of 0.018 mm, an oil film thickness of 0.07 mm, and an oil viscosity of 36 cSt. This real-time data is instantly transmitted to the mill's data processing unit via an internal communication link. Based on the aforementioned state coordinate system, where bearing clearance friction is the first axis, oil film thickness is the second axis, and oil viscosity is the third axis, this unit maps these three sets of data to the three dimensions of the coordinate system, forming a specific coordinate point (0.018, 0.07, 36).

[0041] As the mill continues to run, the sensors repeat the above acquisition process at fixed intervals. New detection data are constantly generated and transmitted to the processing unit, which in turn forms new coordinate points in the state coordinate system. These coordinate points are connected in chronological order to form a state curve that is updated in real time with the monitoring data, fully reflecting the dynamic changes in the operating state of the mill after startup.

[0042] By constructing a multi-dimensional condition monitoring and dynamic curve update system, accurate and real-time characterization of the mill's operating status was achieved, providing a reliable condition basis for subsequent high and low pressure switching control.

[0043] Furthermore, step A300 in the method provided in this application embodiment includes: A310: The state curve is stored in the mill data center.

[0044] A320: Establish a reflective layer port and switching controller and embed them in the mill CNC center.

[0045] A330: Through communication network access, establish the interaction between the state curve repository and the reflective layer port, and establish dual-path interaction between the switching controller and the high-voltage system and the low-voltage system. The high-voltage system consists of a high-pressure oil pump, a high-pressure filter, a high-pressure pipeline, and a high-pressure control valve.

[0046] In this embodiment, the mill data center is the location in the mill used to store state curves.

[0047] Optionally, during mill operation, state curves reflecting its mechanical state are continuously generated. These curves include time-series changes in monitoring elements such as bearing clearance friction, oil film thickness, and oil viscosity. For ease of subsequent analysis and retrieval, these state curves are stored in real-time in the mill's data center. For example, after each grinding cycle, the corresponding state curves are saved according to timestamps. The data center can hold nearly six months of historical curve data, providing ample historical reference for the reflector layer port.

[0048] Meanwhile, the completed reflective layer port and switching controller will be deployed in the mill's CNC center using an embedded method. This deployment method can shorten the data transmission path, reduce signal delay, and ensure rapid response of control commands. For example, by embedding the decision logic of the reflective layer port and the decision program of the switching controller into a dedicated chip in the CNC center, it can work stably in the complex operating environment of the mill, unaffected by external interference.

[0049] Subsequently, a multi-directional interaction mechanism is constructed using communication network access. On one hand, the state curve repository establishes real-time interaction with the reflection layer port. When determining switching conditions, the reflection layer port can retrieve the required historical and real-time state curve data from the repository at any time. For example, when it is necessary to analyze state transition patterns, it can quickly obtain curve segments of nearly one hour. On the other hand, the switching controller establishes dual-path interaction with the high-pressure system and the low-pressure system. The high-pressure system consists of a high-pressure oil pump, a high-pressure filter, a high-pressure pipeline, and a high-pressure control valve. Dual-path interaction ensures that the switching controller can not only send switching commands to these components, such as starting the high-pressure oil pump and opening the high-pressure control valve, but also receive their real-time operating status feedback, such as the current pressure of the high-pressure pipeline and the cleanliness of the filter, to achieve closed-loop monitoring of the switching process.

[0050] By storing state curves, embedding core components, and establishing multi-directional communication, the high-low pressure switching control of the mill thin oil station is provided with efficient data support, rapid response capability, and precise interaction mechanism, ensuring the stability and reliability of subsequent switching decisions and execution.

[0051] Furthermore, step A320 in the method provided in this application embodiment includes: A321: Retrieve historical switching records and mine Markov state chains, using the state transition probability driven by high and low voltage as the mining benchmark.

[0052] A322: Based on the Markov state chain, perform judgment condition transformation to construct the reflection layer port, wherein the judgment condition includes mode dimension and high / low voltage switching dimension.

[0053] In this embodiment, the Markov state chain is obtained by retrieving historical switching records and using the state transition probability driven by high and low voltage as the mining benchmark.

[0054] Specifically, when constructing the reflective layer port, the first step is to retrieve historical switching records from the mill's past operations. These records contain state data such as bearing clearance friction, oil film thickness, and oil viscosity during high-low pressure system switching under different operating conditions, along with the corresponding switching results. Based on this historical data, and using the state transition probability under high-low pressure drive as a benchmark, a Markov state chain is extracted. For example, in the historical records, when the bearing clearance friction increases from 0.02mm to 0.03mm and the oil film thickness decreases from 0.06mm to 0.05mm, the transition from low-pressure to high-pressure state occurs 400 times, accounting for 40% of the total number of transitions. This 40% is the probability of that state transition, and this series of states and their corresponding transition probabilities constitute the Markov state chain.

[0055] Next, the first and second slope threshold groups are determined according to the Markov state chain. The first judgment condition of the mode dimension is determined by mapping the threshold and the switching mode. The second judgment condition of the high and low voltage switching dimension is determined by combining the preset probability of state transition based on the monitoring element transition. Then, the reflection layer port is constructed. The specific steps are explained in detail in A312-1-A312-4.

[0056] By retrieving historical switching records and mining the Markov state chain driven by high and low voltage, a basis for the state transition rules of the reflection layer port construction is provided, supporting the subsequent transformation of judgment conditions.

[0057] Furthermore, step A312 in the method provided in this application embodiment includes: A312-1: Based on the Markov state chain, determine the first slope threshold group and the second slope threshold group, wherein the first slope threshold is a high slope threshold, and the thresholds in the group correspond one-to-one with the monitoring elements.

[0058] A312-2: By mapping the first slope threshold to the quasi-switching mode, and the second slope threshold to the safe and conservative mode, the first judgment condition of the mode dimension is determined.

[0059] A312-3: Based on the Markov state chain, determine the second judgment condition for the high-low voltage switching dimension, wherein the condition is based on the preset probability of state transition based on the transition of monitoring elements.

[0060] A312-4: Construct the reflection layer port according to the first judgment condition and the second judgment condition.

[0061] In this embodiment, the first judgment condition is determined by mapping a first slope threshold group (high slope threshold) to a quasi-switching mode and a second slope threshold group (low slope threshold) to a safe and conservative mode, and is used to determine the switching mode to be adopted based on the slope of the state curve. The second judgment condition is a switching mode determined based on a Markov state chain and using the preset probability of state transition of the monitored element as the basis, and is used to determine whether high-low voltage switching is feasible.

[0062] Specifically, firstly, based on the Markov state chain obtained from the mining, the first slope threshold group and the second slope threshold group are determined. The Markov state chain contains the state transition laws of the mill in different operating stages, such as the state transition from low pressure to high pressure during startup to quickly establish an oil film, and the state transition from high pressure to low pressure during normal operation to maintain stable lubrication.

[0063] Based on these patterns, corresponding thresholds are set for monitoring elements such as bearing clearance friction, oil film thickness, and oil viscosity. The first slope threshold group consists of high slope thresholds; for example, the first threshold for bearing clearance friction is set to 0.00015 mm / ms, the first threshold for oil film thickness is set to -0.00025 mm / ms (negative values ​​indicate decreasing thickness), and the first threshold for oil viscosity is set to -0.05 cSt / ms (negative values ​​indicate decreasing viscosity). These thresholds correspond to scenarios with rapidly changing operating conditions. The second slope threshold group consists of low slope thresholds; for example, the second threshold for bearing clearance friction is set to 0.00008 mm / ms, the second threshold for oil film thickness is set to -0.00015 mm / ms, and the second threshold for oil viscosity is set to -0.02 cSt / ms, corresponding to scenarios with relatively stable operating conditions.

[0064] Next, the first judgment condition for the mode dimension is determined. When the slope of the monitored element exceeds the first slope threshold group, it indicates that the operating condition is unstable and the variables of each element are changing rapidly. For example, during the rapid establishment of the oil film in the start-up phase, the oil film thickness decreases at a large slope. At this time, the quasi-switching mode is triggered to respond quickly. When the slope is within the range of the second slope threshold group, it indicates that the operating condition is stable and the element changes slowly. For example, during normal operation, the oil film thickness remains stable and the slope is small. At this time, the safe and conservative mode is triggered to perform fine switching of multi-level analysis.

[0065] Furthermore, the pressure requirements for lubricating oil differ during mill startup and operation. For example, during startup, the mill components are in a state of static friction, requiring higher oil pressure to quickly establish an oil film and reduce wear during startup. During normal operation, the friction state is relatively stable, and a lower and more stable oil pressure is sufficient to meet lubrication needs. High-low pressure switching control can automatically adjust the lubricating oil pressure according to the mill's operating state, ensuring suitable lubrication conditions at different stages. For instance, in dynamic conditions, when the required oil film thickness changes, a brief switch to high pressure can be made to replenish the oil film, followed by a switch to low pressure for steady-state drive. The slope at this point is a transient variable. For example, if the slope remains consistently high, it indicates unstable operating conditions, meaning the variables are changing rapidly. Therefore, a high degree of immediacy in adjustment is required to avoid delays that cannot keep up with the changing conditions. A quasi-switching mode is adopted, where switching is allowed immediately upon mode triggering for rapid decision-making and response. Otherwise, a conservative mode is used, which involves deploying refined switching parameters based on multi-level analysis of the variable elements.

[0066] Simultaneously, the second judgment condition for determining the high-low pressure switching dimension is based on the Markov state chain, namely, the preset probability of state transitions based on the monitoring element transitions. For example, the preset probability of transitioning from a low-pressure state to a high-pressure state (such as the need for high pressure to establish an oil film during startup) is set to 0.7, and the preset probability of transitioning from a high-pressure state to a low-pressure state (such as switching to low pressure to maintain lubrication during normal operation) is set to 0.65. These preset probabilities of state transitions are derived from the statistical analysis of the actual occurrence probability of similar state transitions in historical data.

[0067] Finally, the first and second judgment conditions are integrated to construct the reflection layer port. After integration, when the reflection layer port is working, it will make judgments based on these two conditions simultaneously: on the one hand, it determines whether it is a quasi-switching mode or a safe and conservative mode by matching the slope with the threshold; on the other hand, it determines whether the handover can be executed by whether the state transition probability meets the standard. The two together constitute the core judgment logic of the reflection layer port, ensuring that it can accurately trigger subsequent handover decisions.

[0068] By determining the slope threshold group and the preset probability of state transition based on the Markov state chain, a reflective layer port containing judgment conditions of mode and high and low pressure switching dimensions is constructed, which realizes the accurate judgment of high and low pressure switching under different working conditions of the mill and adapts to the differences in lubrication requirements during the start-up and operation stages.

[0069] Furthermore, step A320 in the method provided in this application embodiment includes: A323: Using pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation as the fifth-order decision logic, and using the historical switching records, the first supervised training is performed to determine the slow response branch.

[0070] A324: For the slow response branch, perform lightweight black-box transfer training to determine the fast response branch.

[0071] A325: Parallelize the slow response branch and the fast response branch, establish connections between the branch output port and the high-voltage system and the low-voltage system, and determine the switching controller.

[0072] In this embodiment, the slow response branch is determined using a fifth-order decision logic of pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation, and is trained with historical switching records after the first supervised training. The fast response branch is determined after lightweight black-box transfer training of the slow response branch, and runs in parallel with the slow response branch to quickly respond to high-low pressure switching requirements.

[0073] Specifically, when constructing the switching controller, a five-order decision logic framework—pre-pressurization, valve synchronization, flow transfer, steady-state regulation, and post-steady-state compensation—is initially established. A large amount of historical switching records are collected, containing key data for each stage of the transition from low pressure to high pressure or vice versa under different mill operating conditions. This data includes, for example, the pressure value during pre-pressurization, the time difference of valve synchronization, the flow transfer rate, the pressure fluctuation range after steady-state regulation, and the duration of post-steady-state compensation. This historical data is then used as training samples for the first supervised training of the five-order decision logic.

[0074] During training, the system learns the optimal parameter combinations for each decision-making stage under different operating conditions. For example, under heavy-load start-up conditions, the pre-pressurization pressure needs to reach a certain threshold to ensure smooth valve body synchronization. Through repeated iterative adjustments, the slow-response branch is finally determined. This branch can handle complex and stable operating conditions with a detailed decision-making process. For instance, in the flow transfer stage, it accurately calculates the relationship between the flow rate change rate and the oil film thickness to ensure a smooth switching process. Its average decision response time is approximately 80ms.

[0075] For the identified slow-response branch, lightweight black-box transfer learning is used to obtain the fast-response branch. This training does not focus on the multi-stage recursive analysis process within the slow-response branch, but only extracts its input data (such as the slope characteristics of the state curve and the transfer probability value) and corresponding output results (such as switching commands and execution parameters) under various typical operating conditions. Through transfer learning, the fast-response branch directly establishes a mapping relationship between input and output, ignoring complex intermediate calculations. For example, when the input is a certain slope threshold and transfer probability, the fast-response branch can directly output the corresponding switching action without repeating the detailed analysis of the slow-response branch, thus shortening its average decision response time to 20ms. This is suitable for handling sudden operating conditions requiring rapid response, such as a sudden drop in oil film thickness.

[0076] The slow-response and fast-response branches are deployed in parallel, operating independently yet in coordination. The slow-response branch primarily handles stable operating conditions requiring precise adjustments, such as high-low pressure switching during normal mill operation, ensuring accurate control of each component through fifth-order logic. The fast-response branch addresses scenarios with sudden changes in operating conditions requiring rapid action, such as a sudden increase in bearing clearance friction, quickly triggering switching commands. Simultaneously, output ports are established for both branches, connecting to the high-pressure and low-pressure systems (composed of high-pressure oil pumps, high-pressure filters, etc.), ensuring that the switching commands output by each branch directly affect the corresponding system components, ultimately forming a complete switching controller.

[0077] By constructing a switching controller with dual response branches, the accuracy and speed of high-low pressure switching are balanced, enabling the switching control to adapt to the fine adjustment requirements of complex and stable operating conditions, as well as the rapid response requirements of sudden operating conditions, thereby improving the overall adaptability of high-low pressure switching in the mill thin oil station.

[0078] Furthermore, step A200 in the method provided in this application embodiment includes: A210: Set the sampling interval according to the mill's operating conditions.

[0079] A220: Based on the sampling interval, set up the front-end sensor array, drive the automated collection and coordinate arrangement of monitoring elements, and update the state curve.

[0080] In this embodiment, the front-end sensor array is configured according to a set sampling interval and is used to drive the automated acquisition and coordinate arrangement of monitoring elements to update the state curve.

[0081] In one embodiment, the real-time requirements for monitoring the operating status of the mill vary under different task conditions, so it is necessary to set the sampling interval according to the specific operating conditions. For example, during the mill start-up phase, the operating conditions change rapidly, and in order to capture the state fluctuations during the oil film establishment process in a timely manner, the sampling interval can be set to 30ms; while during the stable operation phase, the operating conditions are relatively smooth, and the sampling interval can be extended to 50ms to balance monitoring accuracy and system load.

[0082] Next, a corresponding front-end sensor array is configured according to the set sampling interval. This array includes a displacement sensor for detecting bearing clearance friction, an optical sensor for monitoring oil film thickness, and a viscosity sensor for measuring oil viscosity. These sensors are activated synchronously at preset intervals to collect data. The collected data is automatically transmitted to the processing unit and arranged along the axes of the state coordinate system, i.e., the bearing clearance friction value corresponds to the x-axis, the oil film thickness value corresponds to the y-axis, and the oil viscosity value corresponds to the z-axis, forming coordinate points. As sampling continues, new coordinate points are continuously generated and connected in chronological order, and the original state curve is updated accordingly, fully reflecting the current operating status trend of the mill.

[0083] By setting the sampling interval according to the operating conditions and driving the sensor array to automatically collect and arrange data, the dynamic update of the state curve is realized, providing real-time and accurate state basis for triggering subsequent target switching conditions.

[0084] Furthermore, step A200 in the method provided in this application embodiment includes: A230: Trigger the reflection layer port to locate the upper and lower nodes of the updated state curve, wherein the lower node is a real-time update node.

[0085] A240: Based on the second judgment condition, perform a state transition probability determination. If the transition from the upper node to the lower node does not meet the preset probability, terminate the switching decision.

[0086] A250: If the preset probability is met, determine the target switching state, wherein the target switching state is high voltage-low voltage or low voltage-high voltage.

[0087] A260: Based on the upper and lower nodes, the target switching mode is determined by slope calculation and threshold matching.

[0088] A270: Add the target switching state and target switching mode to the target switching conditions.

[0089] Optionally, after triggering the reflective layer port, the updated state curve is first located to distinguish between upper and lower nodes. The upper node refers to the monitoring data point at a historical moment in the state curve, which includes information such as bearing clearance friction, oil film thickness, and oil viscosity at that moment. For example, the node data at time t1 during the stable operation phase of the mill is 0.02mm bearing clearance friction, 0.06mm oil film thickness, and 35cSt oil viscosity. The lower node is the data point at the current moment that is updated in real time. For example, the node data at time t2 (50ms interval from t1) is 0.03mm bearing clearance friction, 0.05mm oil film thickness, and 33cSt oil viscosity. This node is continuously refreshed with the latest monitoring data.

[0090] Next, when determining the state transition probability from the upper-level node to the lower-level node based on the second judgment condition, it is first clarified that the preset probability threshold in the second judgment condition is determined based on the Markov state chain. The Markov state chain originates from the mining of a large number of historical switching records. By statistically analyzing the probability distribution of successful and effective switching in similar state transitions (i.e., changes in monitoring elements from the upper-level node to the lower-level node) in history, a value that can both ensure switching reliability and adapt to changes in operating conditions is selected as the threshold. This threshold represents the minimum standard for the system to determine that the state transition is feasible.

[0091] The calculation of the actual migration probability involves combining the monitoring features of the current parent node (historical state) and the child node (real-time state) with the migration patterns of similar states recorded in the Markov state chain. Specifically, it involves finding the frequency in historical records of situations where the current parent node's state is similar to the current child node's state, and then migrating to a situation similar to the child node's state. This frequency serves as the current actual migration probability, reflecting the degree to which the current state change matches historical patterns. If the actual migration probability reaches or exceeds a preset threshold, it indicates that the current state migration conforms to a valid historical pattern, and the switching decision can continue. If it does not reach the threshold, it indicates that the current state change deviates from a valid pattern, and to avoid erroneous operations, the switching decision process is terminated directly.

[0092] Once the state transition probability meets the conditions, the target switching state is determined based on the changing trends of the monitoring data from the upper and lower nodes. If, compared to the upper node, the lower node experiences increased bearing clearance friction and decreased oil film thickness (as shown in the change from t1 to t2 above), it indicates that the mill currently requires higher pressure lubricating oil to maintain oil film stability. In this case, the target switching state is determined to be low-pressure to high-pressure switching. Conversely, if the bearing clearance friction decreases and the oil film thickness increases, the target switching state is high-pressure to low-pressure switching.

[0093] Subsequently, the slope of the curve between the upper-level node and the lower-level node is calculated and matched with a preset threshold group to determine the target switching mode. The slope calculation is based on the time interval. For example, if the time interval from t1 to t2 is 50ms, the slope of bearing clearance friction is (0.03-0.02) / 50=0.0002mm / ms, the slope of oil film thickness is (0.05-0.06) / 50=-0.0002mm / ms, and the slope of oil viscosity is (33-35) / 50=-0.04cSt / ms. These slopes are compared with the first slope threshold group (high slope threshold, such as bearing clearance friction 0.00015 mm / ms, oil film thickness -0.00025 mm / ms, oil viscosity -0.05 cSt / ms) and the second slope threshold group (low slope threshold, such as bearing clearance friction 0.00008 mm / ms, oil film thickness -0.00015 mm / ms, oil viscosity -0.02 cSt / ms). Since the above slopes all exceed the first slope threshold group, the target switching mode is determined to be a quasi-switching mode; if the slope does not exceed the first slope threshold group but meets the second slope threshold group, the target switching mode is a safe and conservative mode.

[0094] Finally, the determined target switching state (such as low-voltage to high-voltage switching) is integrated with the target switching mode (such as quasi-switching mode) to form a complete target switching condition.

[0095] By locating status nodes, determining migration probabilities, and identifying switching states and modes, precise target switching conditions are constructed, providing a clear execution basis for the high and low pressure switching control of the mill thin oil station.

[0096] Furthermore, step A300 in the method provided in this application embodiment includes: A340: Based on the target switching mode, execute branch-oriented triggering to activate the target response branch.

[0097] A350: Guided by the target switching state, the system makes logical decisions on high-low voltage switching based on the target response branch to determine the switching strategy.

[0098] In one embodiment, when determining the switching strategy by executing the directional triggering and switching drive decision of the switching controller, branch directional triggering is first executed according to the determined target switching mode. The target switching mode is divided into a quasi-switching mode and a safe and conservative mode, corresponding to the fast response branch and the slow response branch in the switching controller, respectively. For example, when the target switching mode is a quasi-switching mode, it indicates that the mill operating conditions change rapidly, such as a rapid increase in bearing clearance friction and a sharp decrease in oil film thickness during the startup phase. At this time, a rapid system response is required, so directional triggering is performed and the fast response branch is activated. If the target switching mode is a safe and conservative mode, it indicates that the operating conditions are relatively stable, such as the gradual change of monitored elements during normal mill operation. In this case, the slow response branch is activated to ensure the precision of the decision.

[0099] Subsequently, guided by the target switching state, logical decisions for high-low pressure switching are made based on the activated target response branch. The target switching state includes high-pressure-low-pressure switching and low-pressure-high-pressure switching, which determines the direction of switching. For example, if the target switching state is low-pressure-high-pressure switching and the fast response branch is activated, the branch will quickly formulate execution steps based on the fifth-order decision logic of pre-pressurization-valve synchronization-flow transfer-steady-state adjustment-post-steady-state compensation; if the slow response branch is activated, it is also based on the fifth-order logic, but each step is more refined, as shown in Table 1.

[0100] After the above steps, the final switching strategy will include specific execution parameters, such as pressure values, valve opening, flow rate, and duration at each stage. These parameters are matched with the operating characteristics of the high-pressure system (high-pressure oil pump, high-pressure filter, etc.) and the low-pressure system to ensure a smooth and efficient switching process.

[0101] By activating the corresponding response branch according to the target switching mode and making logical decisions guided by the target switching state, a switching strategy adapted to different working conditions was determined, realizing precise control and efficient execution of high and low pressure switching of the mill thin oil station.

[0102] Table 1: Parameters of Fifth-Order Decision Logic Decision-making stage Key parameters Quasi-switching mode parameters Safety Conservative Mode Parameters Pre-stamping Target pressure 12MPa (achieved within 20ms) 10MPa (achieved within 50ms) Valve body synchronization Valve opening change rate 90% / 30ms 30%→60%→90% (Graded adjustment) Traffic transfer Transfer rate 8L / s 5L / s steady-state regulation Pressure fluctuation range ±0.2MPa ±0.1MPa Post-steady-state compensation Compensation duration 50ms 100ms In summary, the high-low pressure switching control method for the mill thin oil station provided in this application has the following technical effects: This application monitors the mechanical state of the mill to determine the state curve, obtains the target switching conditions through two-stage judgment at the reflective layer port, determines the switching strategy, and combines the switching strategy execution driven by the mill's CNC center. This enables the switching control and management of the high and low pressure systems of the mill's thin oil station, allowing the mill to obtain suitable lubrication conditions under different operating conditions. It makes the high and low pressure switching control more precise and reliable, achieving the technical effect of improving the timing and mode matching of high and low pressure switching, optimizing the mill's lubrication effect and operational stability, and meeting the industrial real-time and precise control requirements in grinding scenarios.

[0103] Example 2, as Figure 2 As shown, based on the same inventive concept as in Embodiment 1 above, this application provides a high-low pressure switching control system for a mill thin oil station, the system comprising: The state curve acquisition module 1 is used to determine the state curve by monitoring the mechanical condition of the mill, wherein the monitoring elements include at least bearing clearance friction and oil circuit characteristics.

[0104] The target switching condition triggering module 2 is used to trigger the reflection layer port according to the state curve, perform a two-stage judgment based on the state transition probability and the curve slope, and trigger the target switching condition. The target switching condition includes a switching mode and a switching state. The switching mode is a quasi-switching mode or a safe and conservative mode, and the switching state is a high-voltage-low-voltage or low-voltage-high-voltage switching.

[0105] The switching strategy acquisition module 3 is used to execute the directional triggering and switching drive decision of the switching controller according to the target switching conditions, determine the switching strategy, and drive the switching strategy to perform switching control management between the high-pressure system and the low-pressure system according to the mill CNC center.

[0106] Decision chain acquisition module 4, wherein the decision chain acquisition module 4 is used to take pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation as the decision chain.

[0107] Furthermore, the state curve acquisition module 1 is used to perform the following steps: The oil circuit features include at least oil film thickness and oil viscosity; based on the monitoring elements, multi-state axes are determined and a state coordinate system is constructed; with the start of the mill, multi-position monitoring based on the monitoring elements is triggered, and monitoring is integrated in the state coordinate system to synchronize the state curves, wherein the state curves are updated in time with the monitoring data.

[0108] Furthermore, the switching strategy acquisition module 3 is used to perform the following steps: The state curves are stored in the mill data center; a reflective layer port and a switching controller are established and embedded in the mill CNC center; through communication network access, the interaction between the state curve storage library and the reflective layer port is established, and a dual-path interaction between the switching controller and the high-pressure system and the low-pressure system is established. The high-pressure system consists of a high-pressure oil pump, a high-pressure filter, a high-pressure pipeline, and a high-pressure control valve.

[0109] Furthermore, the switching strategy acquisition module 3 is used to perform the following steps: Historical switching records are retrieved to mine Markov state chains, with the state transition probability based on high and low voltage driving as the mining benchmark. Based on the Markov state chain, judgment condition transformation is performed to construct the reflection layer port, wherein the judgment condition includes mode dimension and high and low voltage switching dimension.

[0110] Furthermore, the switching strategy acquisition module 3 is used to perform the following steps: Based on the Markov state chain, a first slope threshold group and a second slope threshold group are determined, wherein the first slope threshold is a high slope threshold, and the thresholds within the group correspond one-to-one with the monitored elements; by mapping the first slope threshold with the quasi-switching mode, and the second slope threshold with the safe and conservative mode, a first judgment condition for the mode dimension is determined; based on the Markov state chain, a second judgment condition for the high-low voltage switching dimension is determined, wherein a preset probability of state transition based on the monitoring element transition is used as a condition; based on the first judgment condition and the second judgment condition, the reflection layer port is constructed.

[0111] Furthermore, the switching strategy acquisition module 3 is used to perform the following steps: Using pre-pressurization-valve body synchronization-flow transfer-steady-state regulation-post-steady-state compensation as the fifth-order decision logic, the historical switching records are used for first-level supervised training to determine the slow-response branch; for the slow-response branch, lightweight black-box migration training is performed to determine the fast-response branch; the slow-response branch and the fast-response branch are run in parallel, and the branch output ports are connected to the high-pressure system and the low-pressure system to determine the switching controller.

[0112] Furthermore, the target switching condition triggering module 2 is used to perform the following steps: Based on the mill's operating conditions, a sampling interval is set; based on the sampling interval, a front-end sensor array is set up to drive automated collection of monitoring elements and coordinate arrangement, and to update the state curve.

[0113] Furthermore, the target switching condition triggering module 2 is used to perform the following steps: The reflection layer port is triggered to locate the upper and lower nodes of the updated state curve, wherein the lower node is a real-time update node; according to the second judgment condition, a state transition probability determination is performed; if the transition from the upper node to the lower node does not meet the preset probability, the switching decision is terminated; if the preset probability is met, the target switching state is determined, wherein the target switching state is high voltage-low voltage or low voltage-high voltage; based on the upper and lower nodes, the target switching mode is determined by slope calculation and threshold matching; the target switching state and target switching mode are added to the target switching conditions.

[0114] Furthermore, the switching strategy acquisition module 3 is used to perform the following steps: Based on the target switching mode, branch-oriented triggering is executed to activate the target response branch; guided by the target switching state, logical decisions for high-low voltage switching are made based on the target response branch to determine the switching strategy.

[0115] The high-low pressure switching control system for the mill thin oil station provided in this embodiment of the invention can execute the high-low pressure switching control method for the mill thin oil station provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0116] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0117] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A high-low pressure switching control method for a mill thin oil station, characterized in that, The method includes: By monitoring the mechanical condition of the mill, the condition curve is determined. The monitoring elements include at least bearing clearance friction and oil circuit characteristics. According to the state curve, the reflective layer port is triggered to perform a two-stage determination based on the state transition probability and the slope of the curve, triggering the target switching condition. The target switching condition includes a switching mode and a switching state. The switching mode is a quasi-switching mode or a safe and conservative mode, and the switching state is a high-voltage-low-voltage or low-voltage-high-voltage switching. Based on the target switching conditions, the switching controller performs directional triggering and switching drive decision-making to determine the switching strategy. According to the mill CNC center, the switching strategy is driven to perform switching control management between the high-pressure system and the low-pressure system. The decision chain is pre-pressurization-valve body synchronization-flow transfer-steady-state regulation-post-steady-state compensation.

2. The high / low pressure switching control method for the mill thin oil station as described in claim 1, characterized in that, The oil circuit characteristics include at least oil film thickness and oil viscosity; Based on the monitoring elements, determine the multi-state axes and construct a state coordinate system; When the mill starts, multi-position monitoring based on monitoring elements is triggered, and the monitoring is integrated in the state coordinate system to synchronize the state curve, wherein the state curve is updated in time with the monitoring data.

3. The high / low pressure switching control method for the mill thin oil station as described in claim 1, characterized in that, The state curves are stored in the mill data center; Establish a reflective layer port and a switching controller, and embed them in the mill CNC center; Through communication network access, an interaction is established between the state curve repository and the reflective layer port, and a dual-path interaction is established between the switching controller and the high-voltage system and the low-voltage system. The high-voltage system consists of a high-pressure oil pump, a high-pressure filter, a high-pressure pipeline, and a high-pressure control valve.

4. The high / low pressure switching control method for the mill thin oil station as described in claim 3, characterized in that, Before triggering the reflection layer port, the construction of the reflection layer port includes: Historical switching records are retrieved to mine Markov state chains, with the state transition probability based on high and low voltage driving as the mining benchmark. Based on the Markov state chain, a judgment condition transformation is performed to construct the reflection layer port, wherein the judgment condition includes a mode dimension and a high / low voltage switching dimension.

5. The high / low pressure switching control method for the mill thin oil station as described in claim 4, characterized in that, Perform condition transformation to construct the reflection layer port, including: Based on the Markov state chain, a first slope threshold group and a second slope threshold group are determined, wherein the first slope threshold is a high slope threshold, and the thresholds in the group correspond one-to-one with the monitoring elements. By mapping the first slope threshold to the quasi-switching mode, and the second slope threshold to the safe and conservative mode, the first judgment condition of the mode dimension is determined; Based on the Markov state chain, a second judgment condition for the high-low voltage switching dimension is determined, wherein the condition is based on a preset probability of state transition based on the transition of monitoring elements. Based on the first and second judgment conditions, the reflection layer port is constructed.

6. The high / low pressure switching control method for the mill thin oil station as described in claim 4, characterized in that, Before executing the directed triggering of the switching controller, the construction of the switching controller includes: Using pre-pressurization-valve body synchronization-flow transfer-steady-state adjustment-post-steady-state compensation as the fifth-order decision logic, and based on the historical switching records, the first supervised training is performed to determine the slow response branch; For the slow response branch, perform lightweight black-box transfer training to determine the fast response branch; The slow-response branch and the fast-response branch are parallelized, and the branch output port is connected to the high-voltage system and the low-voltage system to determine the switching controller.

7. The high / low pressure switching control method for the mill thin oil station as described in claim 5, characterized in that, The conditions that trigger target switching include: The sampling interval is set according to the mill's operating conditions; Based on the sampling interval, a front-end sensor array is set up to drive automated collection and coordinate arrangement of monitoring elements, and update the state curve.

8. The high / low pressure switching control method for the mill thin oil station as described in claim 7, characterized in that, The conditions that trigger target switching include: Trigger the reflection layer port to locate the upper and lower nodes of the updated state curve, wherein the lower node is a real-time update node; Based on the second judgment condition, a state transition probability determination is performed. If the transition from the upper node to the lower node does not meet the preset probability, the switching decision is terminated. If the preset probability is met, the target switching state is determined, wherein the target switching state is high voltage-low voltage or low voltage-high voltage; Based on the upper and lower nodes, the target switching mode is determined by slope calculation and threshold matching; Add the target switching state and target switching mode to the target switching conditions.

9. The high / low pressure switching control method for the mill thin oil station as described in claim 8, characterized in that, Execute the directional triggering and switching drive decisions of the switching controller to determine the switching strategy, including: Based on the target switching mode, execute branch-oriented triggering to activate the target response branch; Guided by the target switching state, the switching strategy is determined by making logical decisions on high-low voltage switching based on the target response branch.

10. A high-low pressure switching control system for a mill thin oil station, characterized in that, For implementing the high / low pressure switching control method for a mill thin oil station according to any one of claims 1-9, the system comprises: The state curve acquisition module is used to determine the state curve by monitoring the mechanical condition of the mill. The monitoring elements include at least bearing clearance friction and oil circuit characteristics. The target switching condition triggering module is used to trigger the reflection layer port according to the state curve, perform a two-stage judgment based on the state transition probability and the curve slope, and trigger the target switching condition. The target switching condition includes a switching mode and a switching state. The switching mode is a quasi-switching mode or a safe and conservative mode, and the switching state is a high-voltage-low-voltage or low-voltage-high-voltage switching. The switching strategy acquisition module is used to execute the directional triggering and switching drive decision of the switching controller according to the target switching conditions, determine the switching strategy, and drive the switching strategy to perform switching control management between the high-pressure system and the low-pressure system according to the mill CNC center. The decision chain acquisition module is used to establish a decision chain based on pre-pressurization, valve body synchronization, flow transfer, steady-state regulation, and post-steady-state compensation.

Images