A fluid delivery stability real-time intervention method for high speed rotating machinery

Abstract

The application provides a fluid delivery stability real-time intervention method for a high-speed rotating machine, and relates to the technical field of process control, and comprises the following steps: acquiring a current output intensity value of a driving force source driving the high-speed rotating machine and a current actual rotating speed of the high-speed rotating machine in real time; calculating a real-time driving efficiency ratio, which represents the actual rotating speed driven by unit driving force intensity; determining whether the high-speed rotating machine enters a rotating speed instability critical state caused by abnormal increase of fluid load according to a change trend of the real-time driving efficiency ratio relative to a preset reference; if it is determined that the rotating speed instability critical state is entered, a cooperative intervention operation is performed, which comprises: increasing the output intensity of the driving force source to increase the rotating speed, and reducing the supply rate of the fluid delivered in unit time; after the cooperative intervention operation is performed, the real-time driving efficiency ratio is continuously monitored, and the fluid supply rate is gradually restored to a set value after the real-time driving efficiency ratio returns to a normal range.

Description

Technical Field

[0001] This invention belongs to the field of process control technology, specifically relating to a real-time intervention method for fluid transport stability in high-speed rotating machinery. Background Technology

[0002] In many industrial applications, high-speed rotating machinery is widely used to transport various fluid media to target areas, such as centrifugal pumps in the chemical industry, turbo compressors in the energy industry, and high-speed atomizers in the surface treatment industry. During operation, the stability and reliability of these devices depend heavily on the dynamic balance between fluid load and mechanical drive capability.

[0003] When the fluid load changes abruptly, such as a sudden surge in the supply rate or a sustained increase in the viscosity of the medium, the drive end of high-speed rotating machinery will experience an abnormally increased resistance torque. If this imbalance is not identified and addressed in time, it may lead to a sudden drop in mechanical speed, a decrease in delivery efficiency, or even a serious malfunction where the delivered fluid flows back to the drive end—manifesting as cavitation in centrifugal pumps, surge in compressors, and backflow in atomizers.

[0004] To address such issues, existing technologies mostly employ passive protection through mechanical structural improvements. For example, reflux channels or vents are installed inside the equipment to drain excess fluid or backflow media when a fault occurs. These solutions are essentially remedial measures with inherent limitations: first, they cannot fundamentally eliminate the risk of failure; once a fault occurs, it may still cause irreversible damage to precision drive components; second, the fluid discharged from the vent can contaminate non-target areas, compromising the precision of process boundaries, which is unacceptable in high-precision operating scenarios. Therefore, how to achieve proactive identification and intervention in the critical precursor stage of abnormally increased fluid load is a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, a real-time intervention method for fluid transport stability in high-speed rotating machinery is provided, wherein the high-speed rotating machinery is driven by a driving force source and is used to transport fluid to a target area, comprising the following steps:

[0006] The current output intensity value of the driving force source driving the high-speed rotating machinery and the current actual rotation speed of the high-speed rotating machinery are obtained in real time.

[0007] Based on the current output intensity value of the driving force source and the current actual rotation speed, the real-time driving efficiency ratio is calculated. The driving efficiency ratio represents the actual rotation speed that can be driven by a unit driving force intensity.

[0008] Based on the changing trend of the real-time drive efficiency ratio relative to the preset benchmark, it is determined whether the high-speed rotating machinery has entered a critical state of speed instability caused by an abnormal increase in fluid load.

[0009] If it is determined that the speed instability critical state has been entered, a coordinated intervention operation is performed. The coordinated intervention operation includes: increasing the output intensity of the driving force source to increase the speed and reducing the supply rate of the fluid being transported per unit time.

[0010] After performing the coordinated intervention operation, the real-time drive efficiency ratio is continuously monitored, and after the real-time drive efficiency ratio recovers to the normal range, the fluid supply rate is gradually restored to the set value.

[0011] According to the technical solution provided in this application, determining whether the high-speed rotating machinery has entered a critical state of speed instability based on the changing trend of the real-time drive efficiency ratio includes the following steps:

[0012] Obtain the instantaneous rate of change of the real-time drive efficiency ratio over multiple consecutive sampling periods;

[0013] Set a first short-time threshold and a second long-time threshold, wherein the absolute value of the second long-time threshold is less than the absolute value of the first short-time threshold;

[0014] If any of the following conditions are met, the system is considered to have entered a critical state of speed instability:

[0015] The instantaneous rate of change in a single sampling period is lower than the first short-time threshold;

[0016] Within a series of consecutive sampling periods, the cumulative decrease in the instantaneous rate of change exceeds the second long-term threshold.

[0017] According to the technical solution provided in this application, obtaining the first short-time threshold and the second long-time threshold includes the following steps:

[0018] When the high-speed rotating machinery is in an unloaded operating state without fluid load, a standard speed-driving force relationship curve is tested and established under different driving force intensities, and a set of standard driving efficiency ratios are calculated based on the curve.

[0019] During operation, the current driving force intensity and the current actual speed are acquired in real time, and the expected value of the standard speed under the current driving force intensity is calculated based on the standard speed-driving force relationship curve.

[0020] Based on the deviation between the current actual speed and the expected value of the standard speed, calculate the performance degradation coefficient that characterizes the current load performance of the high-speed rotating machinery;

[0021] Based on the performance degradation coefficient, a set of preset basic thresholds are scaled in real time to dynamically generate the first short-term threshold and the second long-term threshold suitable for the current working conditions.

[0022] According to the technical solution provided in this application, the step of dynamically generating a first short-term threshold and a second long-term threshold suitable for the current operating condition by scaling a set of preset basic thresholds in real time based on the performance degradation coefficient includes the following steps:

[0023] If the performance degradation coefficient is less than or equal to the first performance threshold, then the first adjustment strategy is adopted: keep the second long-term threshold unchanged, and make the absolute value of the first short-term threshold decrease as the performance degradation coefficient increases;

[0024] If the performance degradation coefficient is greater than the first performance threshold, a second adjustment strategy is adopted: the absolute values ​​of both the first short-term threshold and the second long-term threshold increase as the performance degradation coefficient increases.

[0025] According to the technical solution provided in this application, the following steps are also included:

[0026] Real-time monitoring of pressure values ​​in the fluid supply pipeline;

[0027] When it is determined that the speed instability critical state has been entered, the fluid pipeline pressure change characteristics within the set time window before and after the determination time are simultaneously acquired.

[0028] Based on the fluid pipeline pressure change characteristics, identify the main triggering factors for the current instability risk:

[0029] If the fluid pipeline pressure exhibits a characteristic of instantaneous rise followed by a sudden drop before the determination time, the main cause is determined to be a sudden change in the fluid supply rate.

[0030] If the fluid pipeline pressure shows a steady and gradual increase before the determination time, the main cause is determined to be an increase in fluid viscosity.

[0031] According to the technical solution provided in this application, the execution of the collaborative intervention operation includes the following steps:

[0032] If the primary cause is a sudden change in fluid supply rate, then the first cooperative control strategy is executed: the control command to reduce the fluid supply rate is configured to have a shorter response time and a larger adjustment range compared to the control command to increase the driving force intensity.

[0033] If the primary cause is increased fluid viscosity, a second collaborative control strategy is executed: a control command to increase the driving force intensity is configured to have a shorter response time and a larger adjustment range compared to a control command to decrease the fluid supply rate.

[0034] According to the technical solution provided in this application, the following steps are also included:

[0035] After the real-time drive efficiency ratio returns to the normal range for the first time, the dynamic evaluation phase is initiated.

[0036] During the dynamic evaluation phase, the dynamic response characteristics of the system to minor disturbances are obtained;

[0037] Based on the dynamic response characteristics, it is determined whether the system has fully recovered from the critical state of speed instability, and based on the determination result, it is determined whether the fluid supply rate can be restored to the set value.

[0038] According to the technical solution provided in this application, determining whether the system has fully recovered from the critical state of speed instability based on the dynamic response characteristics includes the following steps:

[0039] From the dynamic response features, at least two unrelated dynamic feature parameters are extracted to form the dynamic feature vector after the intervention and recovery.

[0040] The dynamic feature vector is compared with the pre-stored benchmark feature vector of normal health status collected under the same process parameters;

[0041] Calculate the matching degree or deviation degree between the dynamic feature vector and the baseline feature vector in the feature space;

[0042] If the matching degree reaches or exceeds the preset recovery threshold, or the deviation degree is lower than the preset abnormal threshold, then it is determined that the system has fully recovered from the critical state of speed instability.

[0043] According to the technical solution provided in this application, the dynamic characteristic parameters include any two of the following: the response recovery time constant to changes in fluid supply rate, the attenuation rate of the oscillation component at the inherent pulsation frequency of the driving force source, or the trend stability index of the driving efficiency ratio during the evaluation stage.

[0044] According to the technical solution provided in this application, the following steps are also included:

[0045] The cumulative number of times the speed instability critical state is triggered within a unit of time or a single work task cycle is counted;

[0046] If the cumulative number of times exceeds the preset frequency warning threshold, a warning signal is generated and output. The warning signal is used to prompt the inspection of abnormal fluid conditions or the health status of high-speed rotating machinery.

[0047] Compared with the prior art, the beneficial effects of this application are as follows:

[0048] I. Achieved proactive prevention and fundamental suppression of speed instability risk: This method constructs a key derived parameter, the drive efficiency ratio, by real-time monitoring of the driving force intensity and actual speed of high-speed rotating machinery. This parameter is extremely sensitive to changes in fluid load, and can accurately determine whether the system has entered a critical state within tens to hundreds of milliseconds before mechanical speed instability and fluid backflow actually occur. This transforms the control logic from reactive remediation to proactive prevention, fundamentally avoiding the occurrence of faults.

[0049] Second, it significantly improves the stability and reliability of the fluid transport process: By intervening at the critical point, this method eliminates sudden speed drops and fluid backflow caused by load imbalance, thus avoiding the risks of drive component damage and environmental pollution that cannot be eliminated in traditional mechanical venting schemes. This enables high-speed rotating machinery to maintain excellent transport stability under long-term, high-load operation, making it particularly suitable for high-end manufacturing scenarios with stringent requirements for process continuity and equipment reliability.

[0050] Third, it provides a rapid and coordinated optimization intervention method: After determining the critical state, this method performs a coordinated intervention operation of "increasing driving force and reducing supply". Through the dual effects of rapidly enhancing driving capability (pull) and reducing fluid load at the source (push), it can efficiently and smoothly pull the system away from the critical point and eventually restore it to a stable operating state. This mechanism avoids over-adjustment or oscillation that may be caused by adjusting a single parameter, and ensures that the intervention process itself does not introduce new unstable factors. Attached Figure Description

[0051] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0052] Figure 1 A flowchart illustrating the steps of the real-time intervention method for fluid transport stability in high-speed rotating machinery provided in this application. Detailed Implementation

[0053] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0054] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0055] Example 1

[0056] As mentioned in the background section, this application proposes a real-time intervention method for fluid transport stability in high-speed rotating machinery, whereby the high-speed rotating machinery is driven by a power source and is used to transport fluid to a target area, such as... Figure 1 As shown, it includes the following steps:

[0057] S1. Real-time acquisition of the current output intensity value of the driving force source driving the high-speed rotating machinery, and the current actual rotation speed of the high-speed rotating machinery;

[0058] S2. Based on the current output intensity value of the driving force source and the current actual rotation speed, calculate the real-time driving efficiency ratio, whereby the driving efficiency ratio characterizes the actual rotation speed that can be driven by a unit driving force intensity.

[0059] S3. Based on the trend of the change of the real-time drive efficiency ratio relative to the preset benchmark, determine whether the high-speed rotating machinery has entered the critical state of speed instability caused by abnormal increase in fluid load.

[0060] S4. If it is determined that the speed instability critical state has been entered, a coordinated intervention operation is performed. The coordinated intervention operation includes: increasing the output intensity of the driving force source to increase the speed and reducing the supply rate of the fluid being transported per unit time.

[0061] S5. After performing the coordinated intervention operation, continuously monitor the real-time drive efficiency ratio, and after the real-time drive efficiency ratio recovers to the normal range, gradually restore the fluid supply rate to the set value.

[0062] Specifically, the following detailed description uses a specific application of this method in the field of electrostatic spraying as an example. In this embodiment, the high-speed rotating machinery is a turbine motor of a high-speed electrostatic rotary cup, the driving force source is driving air pressure, the fluid is paint, the target area is the workpiece surface, and the critical state of speed instability is the critical state of paint re-return. Those skilled in the art should understand that this specific embodiment is only used to explain the present invention and is not intended to limit the scope of protection thereto.

[0063] S1. Obtain the output strength value of the driving force source and the actual rotational speed of the machine:

[0064] This step aims to obtain the basic feedback signals of the drive state of high-speed rotating machinery. Taking this embodiment as an example, the system obtains the current drive air pressure value supplied to the turbine motor in real time (as a specific form of the output intensity value of the drive force source) through a high-response pressure sensor installed on the drive air circuit, and simultaneously obtains the current actual rotational speed value of the turbine motor spindle in real time through a high-precision encoder or Hall sensor. These two parameters are the basic feedback signals of the control system.

[0065] S2. Calculate the real-time drive efficiency ratio:

[0066] The main control unit (e.g., a PLC or dedicated motion controller) calculates the real-time drive efficiency ratio based on the acquired current output intensity value of the driving force source and the current actual rotational speed. In this embodiment, the real-time drive efficiency ratio is defined as the ratio of the actual rotational speed value to the driving air pressure value, i.e., η = RPM / P_drive. This ratio comprehensively reflects the efficiency of the turbine motor in converting air pressure energy into rotational mechanical energy, and its value is affected by mechanical friction, bearing condition, and most importantly, fluid load (i.e., the state of the coating liquid film inside the cup head). Under stable operating conditions, this ratio will remain within a relatively stable reference range.

[0067] S3. Determine the critical state of speed instability:

[0068] The system determines whether high-speed rotating machinery has entered a critical state of speed instability due to an abnormal increase in fluid load based on the real-time trend of the drive efficiency ratio relative to a preset benchmark. The core logic of the determination is to monitor the downward trend of the drive efficiency ratio. Taking this embodiment as an example, when the coating liquid film inside the cup head begins to thicken abnormally, the fluid load torque increases, causing the speed to decrease under the same drive air pressure, thereby reducing the drive efficiency ratio. By calculating the rate of change of the drive efficiency ratio in real time or comparing it with a dynamic threshold, the system can identify this risk precursor tens to hundreds of milliseconds before the liquid film thickens enough to trigger paint re-return.

[0069] S4. Perform collaborative intervention procedures:

[0070] If the system is determined to be in a critical state of speed instability, the controller immediately executes a coordinated intervention. This operation consists of two synchronous or nearly synchronous sub-actions: First, a command is sent to the control element of the drive force source (such as a pneumatic proportional valve or a high-speed switching valve) to instantaneously increase the output intensity of the drive force source, aiming to rapidly increase the drive torque and pull the speed back to the target value, thereby thinning the thickened fluid load layer through enhanced mechanical action; Second, a command is sent to the metering element of the fluid supply system (such as a metering gear pump or a servo valve) to instantaneously reduce the supply rate of the delivered fluid per unit time, aiming to reduce the amount of fluid entering the machine from the source and directly alleviate the fluid load. This coordinated action of "increasing drive force and reducing supply" can efficiently pull the system away from the critical point. Taking this embodiment as an example, the drive air pressure is increased to increase the turbine motor speed, while the paint discharge flow rate is reduced.

[0071] S5. Recovery Monitoring and Gradual Recovery:

[0072] After the coordinated intervention, the system did not immediately return to its original state, but continuously monitored the real-time drive efficiency ratio. When the drive efficiency ratio recovered and stabilized within the pre-calibrated normal range for a preset stable duration (e.g., 500 milliseconds), it indicated that the system had truly escaped risk. At this point, the controller gradually restored the fluid supply rate to its original setpoint before the intervention using a gentle ramp function, avoiding secondary shocks caused by a step-like recovery of the supply rate. Taking this embodiment as an example, after the drive efficiency ratio recovered and stabilized within the normal range for a preset duration, the paint discharge flow rate was gradually restored to its pre-intervention setpoint.

[0073] The technical effect achieved by this implementation method is to proactively prevent the risk of speed instability during fluid transport. Its technical principle lies in using the drive efficiency ratio, a derived parameter, as a core monitoring indicator. This parameter is extremely sensitive to changes in the fluid load inside the machinery and can provide a clear early warning signal before instability actually occurs. Compared to existing technologies that can only passively release fluid after a failure occurs, this method, through early detection of a decrease in the drive efficiency ratio and combined with millisecond-level coordinated intervention of drive force and supply rate, can fundamentally prevent instability. This not only protects the precision components of high-speed rotating machinery but, more importantly, eliminates fluid backflow, environmental pollution, and process defects caused by instability, significantly improving stability and yield in high-quality operating scenarios.

[0074] In a preferred embodiment, determining whether the high-speed rotating machinery has entered a critical state of speed instability based on the changing trend of the real-time drive efficiency ratio includes the following steps:

[0075] Obtain the instantaneous rate of change of the real-time drive efficiency ratio over multiple consecutive sampling periods;

[0076] Set a first short-time threshold and a second long-time threshold, wherein the absolute value of the second long-time threshold is less than the absolute value of the first short-time threshold;

[0077] If any of the following conditions are met, the system is considered to have entered a critical state of speed instability:

[0078] The instantaneous rate of change in a single sampling period is lower than the first short-time threshold;

[0079] Within a series of consecutive sampling periods, the cumulative decrease in the instantaneous rate of change exceeds the second long-term threshold.

[0080] Taking this embodiment as an example: The system first acquires the instantaneous rate of change of the real-time drive efficiency ratio over multiple consecutive high-speed sampling periods. The instantaneous rate of change dη / dt can be obtained by calculating the difference between the η value of the current sampling period and the η value of the previous sampling period, and then dividing by the sampling period time. It reflects the instantaneous speed of change in the efficiency ratio.

[0081] Simultaneously, the system sets two judgment thresholds: a first short-time threshold and a second long-time threshold. The first short-time threshold is a negative value with a large absolute value, used to detect rapid and sharp drops in the drive efficiency ratio, typically corresponding to sudden situations such as sudden blockage of the fluid supply or the entry of large particles. The second long-time threshold is also a negative value, but its absolute value is set to be smaller than that of the first short-time threshold. It is used to detect slow but continuous downward trends in the drive efficiency ratio, typically corresponding to gradual problems such as a slow increase in fluid viscosity or slight wear of mechanical parts.

[0082] The decision logic uses an "OR" condition; intervention is triggered when either condition is met. Condition 1: If the instantaneous rate of change in a single sampling period is lower than the first short-term threshold, it means the driving efficiency ratio has experienced a sharp decline in a very short time, and the system determines it to be a sudden critical state. Condition 2: If, over multiple consecutive sampling periods (e.g., 10 periods), the absolute value of the cumulative decrease after summing the instantaneous rate of change (negative values) in each period exceeds the second long-term threshold, it means the driving efficiency ratio exhibits a clear and continuous deterioration trend, though not drastic, and the system determines it to be a gradual critical state.

[0083] The technical effect achieved by this implementation is to improve the robustness and reliability of determining the critical state of speed instability. Its technical principle lies in employing a dual-channel parallel monitoring strategy. Single threshold determination is susceptible to signal noise interference or slow detection risks. By setting a high threshold for sudden, rapid changes (the first short-term threshold) and a low threshold for sustained, gradual changes (the second long-term threshold), two risk patterns at different time scales can be covered simultaneously. This ensures that the system can react swiftly to sudden faults and provide reliable early warnings for slowly accumulating risks, significantly reducing the probability of false alarms and missed alarms, making the entire early warning system's perception capability more comprehensive and accurate.

[0084] In a preferred embodiment, obtaining the first short-time threshold and the second long-time threshold includes the following steps:

[0085] When the high-speed rotating machinery is in an unloaded operating state without fluid load, a standard speed-driving force relationship curve is tested and established under different driving force intensities, and a set of standard driving efficiency ratios are calculated based on the curve.

[0086] During operation, the current driving force intensity and the current actual speed are acquired in real time, and the expected value of the standard speed under the current driving force intensity is calculated based on the standard speed-driving force relationship curve.

[0087] Based on the deviation between the current actual speed and the expected value of the standard speed, calculate the performance degradation coefficient that characterizes the current load performance of the high-speed rotating machinery;

[0088] Based on the performance degradation coefficient, a set of preset basic thresholds are scaled in real time to dynamically generate the first short-term threshold and the second long-term threshold suitable for the current working conditions.

[0089] Taking this embodiment as an example: the dynamic acquisition of the threshold begins with a calibration phase. After the high-speed electrostatic rotary cup is installed or during routine maintenance, testing is conducted under dry operation with the turbine motor under no-coating load. The system controls the drive air pressure to change in steps from low to high, recording the turbine speed corresponding to each stable air pressure point, thereby establishing a unique "standard speed-air pressure relationship curve" for this rotary cup. Based on this curve, a set of ideal standard drive efficiency ratios η_standard can be calculated, representing the optimal performance benchmark of the device when it is brand new and under no-load.

[0090] During normal spraying operations, the system acquires the current driving air pressure P_current and the current actual rotational speed RPM_current in real time. Based on the standard curve established during calibration, the expected standard rotational speed RPM_expected, which should be present when the equipment is in a healthy dry operating state under the current air pressure P_current, can be interpolated.

[0091] Subsequently, the performance degradation coefficient K, which characterizes the current load performance of the turbine motor, is calculated. K can be calculated using the formula K = (RPM_expected - RPM_current) / RPM_expected, or by using other normalized calculation methods that reflect the deviation between the actual and ideal speeds. This coefficient K comprehensively reflects the decrease in mechanical efficiency caused by long-term wear and tear, as well as the impact of the current coating load. The larger the K value, the further the actual performance deviates from the ideal state, and the more strained the system is.

[0092] Finally, based on the calculated performance degradation coefficient K, the system scales a set of preset base thresholds (i.e., the first short-term threshold base value T1_base and the second long-term threshold base value T2_base) in real time. The scaling model can be linear, for example: T1_current = T1_base × (1 + α×K), T2_current = T2_base × (1 + β×K), where α and β are scaling factors. In this way, when the device performance degrades (K increases), the judgment threshold will be relaxed accordingly (i.e., the absolute value of its negative value becomes larger), making the judgment conditions adapt to the current actual capability of the device and avoiding excessively frequent false alarms due to device aging.

[0093] The technical effect achieved by this implementation is to realize the adaptability of the judgment threshold, ensuring the effectiveness of the early warning system throughout the entire equipment lifecycle. Its technical principle is to establish a dynamic calibration model based on the equipment's own performance benchmark. Because high-speed rotating machinery wears down with use, its no-load performance gradually degrades. Using a fixed judgment threshold may be too lenient (false alarms) for a new device, and too strict (false alarms) for an older device. By acquiring the equipment fingerprint through no-load operation calibration and calculating the performance degradation coefficient in real time during operation, the threshold can be intelligently relaxed as the equipment ages, thus always maintaining a match between the judgment sensitivity and the actual equipment performance. This solves the engineering problem that fixed-threshold systems cannot be stably applied in the long term.

[0094] In a preferred embodiment, the step of dynamically generating a first short-term threshold and a second long-term threshold suitable for the current operating conditions by scaling a set of preset base thresholds in real time based on the performance degradation coefficient includes the following steps:

[0095] If the performance degradation coefficient is less than or equal to the first performance threshold, then the first adjustment strategy is adopted: keep the second long-term threshold unchanged, and make the absolute value of the first short-term threshold decrease as the performance degradation coefficient increases;

[0096] If the performance degradation coefficient is greater than the first performance threshold, a second adjustment strategy is adopted: the absolute values ​​of both the first short-term threshold and the second long-term threshold increase as the performance degradation coefficient increases.

[0097] Taking this embodiment as an example: The system presets a key performance threshold, namely the first performance threshold K_th. This threshold is determined through experiments or statistical analysis and is used to distinguish between mild and severe performance degradation states of the device. During the actual scaling process, the system first compares the real-time calculated performance degradation coefficient K with the first performance threshold K_th.

[0098] If K ≤ K_th, it indicates that the equipment performance degradation is within a mild range, and the system adopts the first adjustment strategy. Under this strategy, the second long-term threshold (used to detect gradual risks) remains unchanged. This is because under mild degradation, the system's response characteristics to slow changes do not change significantly, and the baseline for gradual risks remains unchanged. However, for the first short-term threshold (used to detect sudden risks), its absolute value decreases as K increases. This means that the system becomes more sensitive to sudden speed drops. The principle is that mild wear may slightly reduce the system's instantaneous overload capacity, therefore, it is necessary to tighten the warning line for sudden risks to compensate for this slight decrease in capacity and ensure that sudden risks can be detected in a timely manner.

[0099] If K > K_th, it indicates that the equipment performance has significantly degraded, entering a severely degraded state. In this case, the system adopts a second adjustment strategy. Under this strategy, the absolute values ​​of both the first short-term threshold and the second long-term threshold increase with the increase of K. This means that the judgment conditions for sudden and gradual risks are relaxed simultaneously. The principle is that when equipment performance degrades severely, normal fluid load fluctuations are more likely to cause a decrease in the drive efficiency ratio. If the thresholds are not relaxed at this time, the system will remain in a false alarm state. Therefore, it is necessary to simultaneously increase the absolute values ​​of the two thresholds (i.e., reduce trigger sensitivity) to suppress noise signals generated by poor basic performance, while still being able to capture real and significant risk trends.

[0100] The technical effect achieved by this implementation is to realize more refined and intelligent threshold management. Its technical principle lies in recognizing that the impact of performance degradation on the sensitivity of different types of risk warnings is non-linear. In the early stages of degradation, it mainly affects the system's transient response capability, therefore, the instantaneous threshold is strategically tightened. In the later stages of degradation, it affects the overall load capacity of the system, therefore, all thresholds need to be relaxed simultaneously to avoid false alarms. This differentiated strategy of tightening first and then loosening, compared to simple linear scaling, can more accurately match the characteristics of the equipment at different health stages. It ensures that the equipment is sufficiently sensitive when new and sufficiently stable when old, thus achieving the optimal balance of risk warning accuracy, reliability, and process stability throughout its entire service life.

[0101] In a preferred embodiment, the following steps are also included:

[0102] Real-time monitoring of pressure values ​​in the fluid supply pipeline;

[0103] When it is determined that the speed instability critical state has been entered, the fluid pipeline pressure change characteristics within the set time window before and after the determination time are simultaneously acquired.

[0104] Based on the fluid pipeline pressure change characteristics, identify the main triggering factors for the current instability risk:

[0105] If the fluid pipeline pressure exhibits a characteristic of instantaneous rise followed by a sudden drop before the determination time, the main cause is determined to be a sudden change in the fluid supply rate.

[0106] If the fluid pipeline pressure shows a steady and gradual increase before the determination time, the main cause is determined to be an increase in fluid viscosity.

[0107] Taking this embodiment as an example: During operation, in addition to monitoring the drive parameters, the system also monitors the pressure value in the pipeline in real time through a high-frequency pressure sensor installed on the fluid supply pipeline (in this embodiment, the paint supply pipeline). This pressure value directly reflects the flow resistance of the fluid from the supply pump to the high-speed rotating machinery (in this embodiment, the rotary cup head).

[0108] When the system determines that it has entered a critical state of speed instability (in this embodiment, a critical state of paint reversion), the controller will simultaneously lock and retrieve historical fluid pipeline pressure data within a set time window before and after the determination time (e.g., 200 milliseconds before the determination to 100 milliseconds after the determination). The focus of the analysis is the pressure change characteristics before the determination time.

[0109] The system performs feature analysis on this pressure data curve, primarily identifying two preset modes. The first mode is a sudden rise followed by a sharp drop: the pressure rapidly climbs to a peak within a very short time (e.g., 10-50 milliseconds), then quickly drops, possibly even below the initial pressure. This characteristic usually corresponds to abrupt changes in the fluid supply rate, such as air bubbles in the fluid causing a brief interruption followed by recovery, or a momentary pulse-like over-spray in the feeding system, causing instantaneous overload within the machinery. The second mode is a steady, gradual rise: the pressure exhibits a relatively stable but continuously rising trend over a timescale of several hundred milliseconds or even longer, without sharp spikes. This characteristic usually corresponds to an increase in fluid viscosity, such as a decrease in fluid temperature or solvent evaporation leading to increased viscosity, resulting in a continuous and gradual increase in flow resistance.

[0110] Based on the identified characteristics of pressure changes, the system categorized the main triggers for the current instability risk into one of the two types mentioned above. This diagnostic result provides a crucial basis for taking more targeted intervention measures.

[0111] The technical effect achieved by this implementation method is to realize the root cause diagnosis of speed instability risk, upgrading intervention measures from general to targeted. The technical principle lies in the fact that instability risks from different physical causes will leave distinctly different traces on the fluid pipeline pressure parameter. Sudden changes in supply rate are essentially instantaneous blockage and release of flow, inevitably causing shock-like pressure fluctuations; while increased viscosity is essentially a slow increase in flow resistance, manifested as a steady rise in pressure. By capturing and analyzing these characteristic patterns in real time, the system can distinguish risks with similar surface symptoms (both involving a decrease in drive efficiency ratio) but different underlying causes. This diagnostic capability is a key manifestation of the intelligence of this method; it points the way to implementing the most effective intervention strategy, avoiding slow recovery or poor results due to mismatch between intervention measures and the underlying cause, and significantly improving the accuracy and efficiency of the entire active intervention system.

[0112] In a preferred embodiment, the execution of the coordinated intervention operation includes the following steps:

[0113] If the primary cause is a sudden change in fluid supply rate, then the first cooperative control strategy is executed: the control command to reduce the fluid supply rate is configured to have a shorter response time and a larger adjustment range compared to the control command to increase the driving force intensity.

[0114] If the primary cause is increased fluid viscosity, a second collaborative control strategy is executed: a control command to increase the driving force intensity is configured to have a shorter response time and a larger adjustment range compared to a control command to decrease the fluid supply rate.

[0115] Taking this embodiment as an example: After the controller determines the critical state of speed instability and completes the cause identification, it no longer executes a single intervention action, but instead calls one of two preset strategies based on the diagnostic results.

[0116] If the primary cause is identified as a sudden change in fluid supply rate (in this embodiment, a sudden change in paint flow rate), the first collaborative control strategy is executed. The core idea of ​​this strategy is rapid source cutoff to assist in speed stabilization. In implementation, the control command to reduce the fluid supply rate is given the highest priority: its control loop (e.g., commands to servo valves or metering pumps) is configured to have an extremely short response time (e.g., a delay of less than 20 milliseconds from command issuance to the start of flow rate change), and the adjustment range is set to be relatively large (e.g., immediately reducing to 30%-50% of the original supply rate), aiming to cut off the excessive fluid supply causing overload as quickly as possible. Simultaneously, the control command to increase the driving force intensity (in this embodiment, driving air pressure) is issued synchronously, but its configured response time can be slightly longer (e.g., a smooth increase via a proportional valve), and the adjustment range is relatively small (e.g., an increase of 5%-10%). Its main purpose is to maintain sufficient rotational speed to ensure normal operation after a sudden drop in supply rate, avoiding process interruption due to the sudden drop in supply rate.

[0117] If the primary cause is identified as increased fluid viscosity (in this embodiment, increased paint viscosity), a second collaborative control strategy is executed. The core idea of ​​this strategy is to significantly increase speed while simultaneously reducing load. In implementation, the control command to increase the driving force intensity is given the highest priority: it requires a rapid, step-up increase in driving force intensity to a higher value (e.g., 15%-25%) to generate stronger mechanical action to overcome the significant shear resistance caused by the high-viscosity fluid. The response time of this command is also required to be very short. Conversely, the control command to reduce the fluid supply rate is slightly delayed (e.g., 50-100 milliseconds) and the adjustment is smaller (e.g., reduced to 80%-90% of the original supply rate). Its purpose is to supplementarily and gradually reduce the load after the increased speed has established an advantage, thus collaboratively consolidating a stable state.

[0118] The technical effect achieved by this implementation is to dynamically optimize the primary and secondary relationship and intensity ratio of the two control dimensions in the collaborative intervention based on different causes, thereby achieving the fastest and most stable recovery. Its technical principle lies in grasping the main contradiction of system imbalance under different inducing factors. For sudden changes in the supply rate, the main contradiction is excessive input; therefore, the primary means to resolve this contradiction is rapid throttling. For increased viscosity, the main contradiction is insufficient shear force; therefore, the primary means to resolve this contradiction is strong acceleration. By mapping the causal diagnosis results to differentiated control strategies, this method achieves precise customization of intervention actions. This avoids the problems that may arise from a one-size-fits-all intervention (for example, simply drastically reducing the supply rate for increased viscosity may lead to even greater fluid accumulation inside the machinery due to excessively low flow rates), ensuring that the system can quickly return to stability in the most reasonable way regardless of the disturbance, greatly improving the intervention success rate and process robustness.

[0119] In a preferred embodiment, the following steps are also included:

[0120] After the real-time drive efficiency ratio returns to the normal range for the first time, the dynamic evaluation phase is initiated.

[0121] During the dynamic evaluation phase, the dynamic response characteristics of the system to minor disturbances are obtained;

[0122] Based on the dynamic response characteristics, it is determined whether the system has fully recovered from the critical state of speed instability, and based on the determination result, it is determined whether the fluid supply rate can be restored to the set value.

[0123] Taking this embodiment as an example: After the collaborative intervention operation is executed, the system continuously monitors the real-time drive efficiency ratio. When the drive efficiency ratio first recovers and enters the preset normal range, the system does not immediately begin to restore the supply rate, but instead initiates a dynamic evaluation phase. The purpose of this phase is to assess whether the system has the stability to withstand secondary disturbances.

[0124] During the dynamic evaluation phase, the system acquires its dynamic response characteristics using one of two non-disruptive methods. The first method utilizes inherent process gaps: the controller monitors the trajectory and operational commands of the equipment. When the equipment performs an idle stroke (movement without operation) or transitions at the edge of the target area, the fluid's set supply rate drops stepwise from the process value to zero or a low value. The system actively uses this inevitable, significant step change in the supply rate as a natural and powerful test signal. The system accurately records the timing of the supply rate change and simultaneously monitors the response curve of the drive efficiency ratio to this disturbance, particularly the depth of the drop, the overshoot, and the speed and stability of recovery to the normal value. For example, in this embodiment, when the painting robot performs an idle stroke or transitions at the workpiece edge, the paint flow rate setpoint changes stepwise, and the system uses this opportunity to acquire response characteristics.

[0125] The second method is to monitor inherent pulsations: the system uses spectrum analysis to identify and continuously acquire the oscillation component in the real-time drive efficiency ratio signal that corresponds to the inherent rotational frequency or harmonic frequency of the metering element (such as a gear pump) in the fluid supply system. When the system is healthy, the amplitude of this oscillation in the drive efficiency ratio caused by pump pulsations is small and stable; when the fluid state inside the system is still unstable or the flow field has not fully recovered, the system's gain on this frequency disturbance will increase, manifested as a significant increase in the amplitude of the oscillation component. Taking this embodiment as an example, the system continuously acquires the amplitude of the oscillation component of the real-time drive efficiency ratio at the inherent pulsation frequency of the paint pump.

[0126] The system takes the response characteristics (time-domain response curve or frequency-domain oscillation amplitude) obtained by any of the above methods as input, and uses the built-in evaluation algorithm (such as comparing with the health baseline curve, calculating characteristic values, etc.) to determine whether the system has fully recovered from the critical state of speed instability, and dynamically decides whether to end the waiting and allow the fluid supply rate to recover to the set value.

[0127] The technical effect achieved by this implementation method is to realize intelligent and safe recovery processes, replacing fixed delays with dynamic assessments. Its technical principle is that true system stability recovery is not only reflected in the return of static parameters (average drive efficiency ratio) to normal ranges, but also in the recovery of system dynamic characteristics (disturbance resistance and damping characteristics). By actively utilizing the inevitable supply rate steps in the process as test signals, or passively monitoring the system's ability to suppress inherent pulsations, this method can perform a health check on the system without interfering with normal operational quality. This avoids the inefficient practice of waiting a fixed, conservative period regardless of the severity of risk in traditional methods, and also prevents secondary risks caused by prematurely restoring the supply rate before the system is truly stable. It ensures that every recovery decision is based on evidence and is safe and reliable, thereby potentially shortening recovery time and increasing production cycle time while ensuring safety.

[0128] In a preferred embodiment, determining whether the system has fully recovered from the critical state of speed instability based on the dynamic response characteristics includes the following steps:

[0129] From the dynamic response features, at least two unrelated dynamic feature parameters are extracted to form the dynamic feature vector after the intervention and recovery.

[0130] The dynamic feature vector is compared with the pre-stored benchmark feature vector of normal health status collected under the same process parameters;

[0131] Calculate the matching degree or deviation degree between the dynamic feature vector and the baseline feature vector in the feature space;

[0132] If the matching degree reaches or exceeds the preset recovery threshold, or the deviation degree is lower than the preset abnormal threshold, then it is determined that the system has fully recovered from the critical state of speed instability.

[0133] Taking this embodiment as an example: The system extracts at least two independent parameters from the dynamic response characteristics, each describing the system's dynamic behavior from different dimensions, to construct a dynamic feature vector after the current recovery. For example, if a supply rate step test is used (in this embodiment, a step change in paint flow rate), the response recovery time constant (describing the speed at which the system recovers from a disturbance to a steady state) and the overshoot percentage (describing the severity of oscillations in the system response) can be extracted; if pump pulsation monitoring is used, the amplitude of the oscillation component at a specific frequency and the oscillation decay rate can be extracted.

[0134] Meanwhile, the system pre-stores one or more baseline feature vector libraries. These baseline vectors are the average value of feature vectors obtained through multiple sampling and statistical analysis when the system is in a recognized normal and healthy state under the same or highly similar operating process parameters (such as fluid type, target rotation speed, and basic supply rate). They represent the health fingerprint under that operating condition. Taking this embodiment as an example, the baseline feature vectors are feature vectors of a normal and healthy state collected under the same spraying process parameters.

[0135] During the evaluation, the system compares the dynamic feature vector acquired in real time with the pre-stored baseline feature vector under the corresponding working condition. The core of the comparison is to calculate the similarity between the two in the mathematical feature space. This can be measured by calculating the Euclidean distance (deviation), with a smaller distance indicating a closer approximation of the healthy state; or by calculating the cosine similarity (matching degree), with a higher similarity indicating a closer approximation.

[0136] The system presets a recovery threshold (e.g., similarity must reach 90%) or an anomaly threshold (e.g., Euclidean distance must not exceed a certain value). If the calculated matching degree reaches or exceeds the recovery threshold, or the deviation is lower than the anomaly threshold, the system's dynamic characteristics are determined to be fully consistent with the health benchmark, meaning the system has fully recovered from the critical state of speed instability. Otherwise, the recovery is deemed insufficient, the current state is maintained, and the evaluation period is extended.

[0137] The technical effect achieved by this implementation is to upgrade recovery judgment from subjective, qualitative observation to objective, quantitative, and scientific assessment. Its technical principle is the application of pattern recognition and statistical decision theory. Individual feature parameters are easily affected by random factors, while vectors composed of multiple features can more comprehensively characterize the system's state. By quantitatively comparing with a benchmark feature vector, the fuzzy concept of full recovery gains a clear and repeatable mathematical standard. This method significantly improves the accuracy and consistency of state assessment and reduces the risk of misjudgment. It makes recovery decisions no longer based on empirical guesswork but on precise calculations based on data, representing a significant advancement in the intelligent and precise direction of this active intervention method. It lays a core data processing foundation for building a high-speed rotating machinery control system with self-learning and adaptive capabilities.

[0138] In a preferred embodiment, the dynamic characteristic parameters include any two of the following: the response recovery time constant to changes in the fluid supply rate, the attenuation rate of the oscillation component at the inherent pulsation frequency of the driving force source, or the trend stability index of the driving efficiency ratio during the evaluation phase.

[0139] Taking this embodiment as an example: This implementation method clarifies the specific options for extracting dynamic feature parameters from system response characteristics. These parameters quantify the dynamic behavior of the system from different physical dimensions.

[0140] The first optional dynamic characteristic parameter is the response recovery time constant to changes in the fluid supply rate. When tested using the first method (utilizing a step in the idle supply rate), after obtaining the response curve of the drive efficiency versus the decrease in the supply rate, the system fits the recovery segment of the curve using a first-order or second-order system model. The recovery time constant (e.g., the time required for a first-order system to reach 63.2% of its steady-state value) is extracted. This parameter directly reflects the magnitude of the system's internal damping and the speed at which it recovers from disturbances. A healthy, robust system has a shorter recovery time constant, meaning it can quickly quell disturbances; conversely, if the system is still in a vulnerable state, the recovery process will be sluggish, and the time constant will be longer. In this embodiment, for example, this parameter is the response recovery time constant to changes in the coating flow rate step.

[0141] The second optional parameter is the attenuation rate of the oscillation component at the inherent pulsation frequency of the driving force source. When using the second method (monitoring inherent pulsation), the system separates the oscillation component corresponding to the characteristic frequency of the driving force source (a paint pump in this embodiment) from the real-time driving efficiency ratio signal through bandpass filtering or Fourier transform. The system further analyzes the envelope of this oscillation signal within the evaluation time period and calculates its amplitude attenuation rate (e.g., a percentage decrease per second). The attenuation rate characterizes the system's ability to suppress periodic disturbances. When the fluid state is stable and the system damping characteristics are good, any small forced oscillation will be rapidly attenuated; if the internal fluid state of the machinery is still unstable, the oscillation may continue or even amplify, resulting in a low attenuation rate. In this embodiment, for example, this parameter is the attenuation rate of the oscillation component at the pump's inherent pulsation frequency.

[0142] The third optional parameter is the trend stability index of the driving efficiency ratio during the evaluation phase. This parameter is independent of specific test stimuli, but rather involves statistical analysis of the time series of the driving efficiency ratio itself during the evaluation phase. The system can calculate the standard deviation of the driving efficiency ratio over this period, or calculate the standard deviation of the residuals after linear fitting, or calculate its coefficient of variation (the ratio of the standard deviation to the mean). The trend stability index quantifies the degree of macroscopic fluctuation in the driving efficiency ratio. Even if the average driving efficiency ratio is within the normal range, if its fluctuation is severe (poor stability index), it indicates that the system has not reached a true steady state and may still have unresolved disturbances or be in metastable equilibrium.

[0143] In practice, the system selects any two or all of the above parameters and combines them to form the dynamic feature vector after the current recovery. For example, one feature vector could be [recovery time constant = 120ms, pulsation decay rate = 0.8 / s], and another vector could be [pulsation decay rate = 0.8 / s, trend stability index (standard deviation) = 0.015]. These parameters are usually unrelated due to their different physical meanings, and can comprehensively characterize the health status of the system from multiple independent perspectives such as response speed, damping characteristics, and steady-state fluctuations.

[0144] In a preferred embodiment, the following steps are also included:

[0145] The cumulative number of times the speed instability critical state is triggered within a unit of time or a single work task cycle is counted;

[0146] If the cumulative number of times exceeds the preset frequency warning threshold, a warning signal is generated and output. The warning signal is used to prompt the inspection of abnormal fluid conditions or the health status of high-speed rotating machinery.

[0147] Taking this embodiment as an example: During system operation, an event counter and a timer or task cycle marker are maintained. Each time a coordinated intervention operation is determined and triggered for a critical state of speed instability (paint return critical state in this embodiment), the event counter is incremented by one.

[0148] The statistical period can be set in two main modes: the first is a fixed time window, such as counting the cumulative number of events in the past hour or the past shift (8 hours); the second is to follow the production task cycle, such as counting the cumulative number of events in the entire task cycle from the start of the current task to the end.

[0149] The system has one or more preset frequency warning thresholds. These thresholds are set based on historical production data, process expert experience, or statistical analysis. For example, for a certain stable mass production process, under healthy equipment and normal fluid conditions, the average number of interventions triggered per hour may only be 0-1. Therefore, the warning threshold can be set to "3 times per hour" or "5 times within a single task cycle".

[0150] During operation, the system compares the cumulative number of triggers with a preset frequency warning threshold in real time or periodically (e.g., at the end of each statistical period). If the cumulative number exceeds the threshold, it is determined to be an abnormally frequent occurrence, and the system will automatically generate a warning signal. This warning signal can be output through a pop-up alert box on the Human-Machine Interface (HMI), by sending information to the Manufacturing Execution System (MES), or by triggering an audible and visual alarm.

[0151] The generated warning signals contain clear directions. Their purpose is not to replace real-time intervention, but rather to guide deeper investigations. The warning signals are used to prompt operators or maintenance personnel to check for two types of potential problems: first, abnormal fluid conditions, such as whether the fluid is expired, whether the viscosity deviates significantly from process standards, whether incompatible substances have been mixed in, or whether it contains excessive air bubbles; second, the health condition of high-speed rotating machinery, such as whether bearings are severely worn, whether the dynamic balance is abnormal, or whether there is partial blockage in the fluid channels. In this embodiment, for example, the warning signal is used to prompt the inspection of abnormal coating conditions or the health condition of the rotary cup turbine motor.

[0152] This guides maintenance to shift from treating symptoms to identifying the root cause, achieving a leap from passive response to proactive prevention, and further improving the reliability of equipment operation and maintenance efficiency.

[0153] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A real-time intervention method for fluid transport stability in high-speed rotating machinery, wherein the high-speed rotating machinery is driven by a driving force source and is used to transport fluid to a target area, characterized in that, Includes the following steps: The current output intensity value of the driving force source driving the high-speed rotating machinery and the current actual rotation speed of the high-speed rotating machinery are obtained in real time. Based on the current output intensity value of the driving force source and the current actual rotation speed, the real-time driving efficiency ratio is calculated. The driving efficiency ratio represents the actual rotation speed that can be driven by a unit driving force intensity. Based on the changing trend of the real-time drive efficiency ratio relative to the preset benchmark, it is determined whether the high-speed rotating machinery has entered a critical state of speed instability caused by an abnormal increase in fluid load. If it is determined that the speed instability critical state has been entered, a coordinated intervention operation is performed. The coordinated intervention operation includes: increasing the output intensity of the driving force source to increase the speed and reducing the supply rate of the fluid being transported per unit time. After performing the coordinated intervention operation, the real-time drive efficiency ratio is continuously monitored, and after the real-time drive efficiency ratio recovers to the normal range, the fluid supply rate is gradually restored to the set value. Determining whether the high-speed rotating machinery has entered a critical state of speed instability based on the changing trend of the real-time drive efficiency ratio includes the following steps: Obtain the instantaneous rate of change of the real-time drive efficiency ratio over multiple consecutive sampling periods; Set a first short-time threshold and a second long-time threshold, wherein the absolute value of the second long-time threshold is less than the absolute value of the first short-time threshold; If any of the following conditions are met, the system is considered to have entered a critical state of speed instability: The instantaneous rate of change in a single sampling period is lower than the first short-time threshold; Within a series of consecutively set sampling periods, the cumulative decrease in the instantaneous rate of change exceeds the second long-term threshold. The acquisition of the first short-time threshold and the second long-time threshold includes the following steps: When the high-speed rotating machinery is in an unloaded operating state without fluid load, a standard speed-driving force relationship curve is tested and established under different driving force intensities, and a set of standard driving efficiency ratios are calculated based on the curve. During operation, the current driving force intensity and the current actual speed are acquired in real time, and the expected value of the standard speed under the current driving force intensity is calculated based on the standard speed-driving force relationship curve. Based on the deviation between the current actual speed and the expected value of the standard speed, calculate the performance degradation coefficient that characterizes the current load performance of the high-speed rotating machinery; Based on the performance degradation coefficient, a set of preset basic thresholds are scaled in real time to dynamically generate the first short-term threshold and the second long-term threshold suitable for the current working conditions.

2. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 1, characterized in that, The step of scaling a set of preset base thresholds in real time based on the performance degradation coefficient to dynamically generate a first short-term threshold and a second long-term threshold suitable for the current operating conditions includes the following steps: If the performance degradation coefficient is less than or equal to the first performance threshold, then the first adjustment strategy is adopted: keep the second long-term threshold unchanged, and make the absolute value of the first short-term threshold decrease as the performance degradation coefficient increases; If the performance degradation coefficient is greater than the first performance threshold, a second adjustment strategy is adopted: the absolute values ​​of both the first short-term threshold and the second long-term threshold increase as the performance degradation coefficient increases.

3. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 2, characterized in that, It also includes the following steps: Real-time monitoring of pressure values ​​in the fluid supply pipeline; When it is determined that the speed instability critical state has been entered, the fluid pipeline pressure change characteristics within the set time window before and after the determination time are simultaneously acquired. Based on the fluid pipeline pressure change characteristics, identify the main triggering factors for the current instability risk: If the fluid pipeline pressure exhibits a characteristic of instantaneous rise followed by a sudden drop before the determination time, the main cause is determined to be a sudden change in the fluid supply rate. If the fluid pipeline pressure shows a steady and gradual increase before the determination time, the main cause is determined to be an increase in fluid viscosity.

4. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 3, characterized in that, The execution of the collaborative intervention operation includes the following steps: If the primary cause is a sudden change in fluid supply rate, then the first cooperative control strategy is executed: the control command to reduce the fluid supply rate is configured to have a shorter response time and a larger adjustment range compared to the control command to increase the driving force intensity. If the primary cause is increased fluid viscosity, a second collaborative control strategy is executed: a control command to increase the driving force intensity is configured to have a shorter response time and a larger adjustment range compared to a control command to decrease the fluid supply rate.

5. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 1, characterized in that, It also includes the following steps: After the real-time drive efficiency ratio returns to the normal range for the first time, the dynamic evaluation phase is initiated. During the dynamic evaluation phase, the dynamic response characteristics of the system to minor disturbances are obtained; Based on the dynamic response characteristics, it is determined whether the system has fully recovered from the critical state of speed instability, and based on the determination result, it is determined whether the fluid supply rate can be restored to the set value.

6. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 5, characterized in that, The determination of whether the system has fully recovered from the critical state of speed instability based on the dynamic response characteristics includes the following steps: From the dynamic response features, at least two unrelated dynamic feature parameters are extracted to form the dynamic feature vector after the intervention and recovery. The dynamic feature vector is compared with the pre-stored benchmark feature vector of normal health status collected under the same process parameters; Calculate the matching degree or deviation degree between the dynamic feature vector and the baseline feature vector in the feature space; If the matching degree reaches or exceeds the preset recovery threshold, or the deviation degree is lower than the preset abnormal threshold, then it is determined that the system has fully recovered from the critical state of speed instability.

7. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 6, characterized in that, The dynamic characteristic parameters include any two of the following: the response recovery time constant to changes in fluid supply rate, the attenuation rate of the oscillation component at the inherent pulsation frequency of the driving force source, or the trend stability index of the driving efficiency ratio during the evaluation phase.

8. The real-time intervention method for fluid transport stability in high-speed rotating machinery according to claim 1, characterized in that, It also includes the following steps: The cumulative number of times the speed instability critical state is triggered within a unit of time or a single work task cycle is counted; If the cumulative number of times exceeds the preset frequency warning threshold, a warning signal is generated and output. The warning signal is used to prompt the inspection of abnormal fluid conditions or the health status of high-speed rotating machinery.

Images