Intelligent monitoring method and system for safe state of efflux pump

Abstract

The invention discloses an intelligent monitoring method and system for the safety state of an efflux pump, and relates to the technical field of industrial equipment safety, the system collects working condition data of the efflux pump in real time through a sensor group installed on the efflux pump, and pre-processes the working condition data to construct a fluid state data set and a load evolution data set; medium disturbance evaluation is carried out by calculating a medium disturbance leading edge index VPI and a medium disturbance response threshold TH, when shear disturbance is evaluated to be formed, a load evolution analysis instruction is executed, a load lag evolution index LEI is calculated based on structure response data, and then a joint instability risk index UDR is calculated in combination, so that the shear disturbance is evaluated to be formed. And performing comprehensive instability response evaluation with a set first joint instability boundary threshold UA and a set second joint instability boundary threshold UB, and executing a grading safety adjustment strategy according to a comprehensive instability response evaluation result. According to the system, early recognition, dynamic evaluation and closed-loop response of the abnormal state of the outer conveying pump can be achieved, and the safety and stability of operation of the conveying system are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial equipment safety, in particular to an intelligent monitoring method and system for safety state of an external delivery pump. BACKGROUND

[0002] As the core equipment of long-distance delivery pipe network, the stable operation of the external delivery pump is directly related to the continuity and safety of the entire delivery system. In continuous process such as petroleum and chemical industry, the external delivery pump often operates in high-load and high-viscosity conditions, and its working state is easily affected by factors such as changes in the physical properties of the delivery medium, scheduling fluctuations, and pressure disturbances. The traditional state monitoring method mainly relies on single-point threshold judgment of mechanical temperature rise, current and voltage, etc., and is not easy to effectively model and warn the linkage mechanism of complex disturbance propagation trend and load response, and is difficult to adapt to the operation scene of multiple disturbances and high dynamic coupling. Therefore, it is necessary to introduce an intelligent monitoring mechanism for the safety state of the external delivery pump, which has dynamic recognition ability, can fuse multi-dimensional working condition characteristics and perform risk deduction, so as to realize the whole process modeling and intervention response of the disturbance front formation, load response lag and instability trend amplification, and improve the system robustness and intelligent defense capability of the external delivery pump operation.

[0003] The current industrial site often uses dispersed sensing and manual experience judgment or simple alarm logic to monitor the pump body. This kind of means has significant defects in the face of complex patterns such as "nonlinear disturbance, mechanical lag and coupled instability". First, it is not easy to identify the formation and propagation path of the disturbance wave in advance, such as viscosity pressure wave caused by sudden change of medium physical properties; second, it lacks a dynamic tracking mechanism for quantitatively expressing and tracking the load response lag phenomenon caused by disturbance, resulting in early load fluctuation signals being ignored; third, it lacks a unified index to evaluate and warn the linkage amplification mechanism between disturbance evolution and load response, and it is not easy to form a systematic prevention and control chain. For example, in a complex scheduling scenario, if high-density dispersant medium is introduced, the traditional system can only respond when the pump body vibrates violently or the load overflows, and the best intervention opportunity has been missed. Therefore, it is necessary to build a systematic evaluation framework based on disturbance recognition, load lag perception and risk fusion to fill the gap in the current monitoring mechanism for the whole process cognition of "formation, propagation, response and instability". SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides an intelligent monitoring method and system for safety state of an external delivery pump, which solves the problems in the background art.

[0005] To achieve the above purpose, the present application realizes the following technical scheme: an intelligent monitoring system for safety state of an external delivery pump, comprising a data acquisition module, a disturbance recognition module, a load lag recognition module, a joint risk assessment module and a control execution module. The data acquisition module is used to collect the operating condition data of the external pump in real time based on the sensor group installed on the external pump, and transmit it to the intelligent monitoring system for preprocessing to obtain fluid state data group and load evolution data group. The disturbance identification module is used to calculate the medium disturbance front index VPI based on the fluid state data set, and to evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, the load evolution analysis command is executed. The load hysteresis identification module is used to execute load evolution analysis instructions, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. The joint risk assessment module is used to calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and to conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. The control execution module is used to receive early warning commands from the comprehensive instability response assessment in real time and to perform safety adjustments through the intelligent monitoring system.

[0006] Preferably, the data acquisition module includes a working condition acquisition unit and a data synchronization unit; The operating condition acquisition unit is used to collect the operating condition data of the external pump in real time based on the sensor group installed at each position of the external pump; The sensor group includes a piezoresistive pressure sensor, an online rotary viscometer, an ultrasonic flow meter, a current transformer, a resistance thermometer, and a piezoelectric accelerometer. The piezoresistive pressure sensor is installed on the side wall of the intake pipe near the inlet impeller inlet and on the high-pressure side of the pump outlet section near the outlet check valve to collect the instantaneous inlet air pressure pr and the instantaneous outlet air pressure pc in real time. The online rotary viscometer is installed on the side wall of the main fluid flow section at the pump inlet to collect the fluid viscosity in real time (uf). The ultrasonic flow meter is installed on the inner wall of the pipe section 0.5m before the pump inlet to collect the inlet flow velocity vc in real time. The current transformer is used to be installed between the three-phase busbars of the pump's main power supply input line to collect the motor load current Ir in real time; The resistance thermometer is embedded inside the spindle bushing to collect the spindle housing temperature Ts in real time. The piezoelectric accelerometer is used for adhesive mounting on the outer shell panel of the pump body bearing seat in the vertical direction to collect vibration acceleration ar in real time.

[0007] Preferably, the data synchronization unit is used to establish a communication connection between the sensor group and the intelligent monitoring system based on the wireless network, and transmit the working condition data collected by the sensor group to the intelligent monitoring system in real time for preprocessing to obtain the fluid state data group and the load evolution data group. The preprocessing includes denoising, timestamp alignment, filling in missing values, and dimensionless processing; The denoising process preserves the transient structural features of the operating data and removes high-frequency noise through wavelet denoising technology. The timestamp alignment performs interpolation and prediction to fill in missing time points in the operating data through linear interpolation resampling technology. The missing value filling process completes the operating data affected by sampling interruption and communication packet loss through missing value interpolation technology. The dimensionless processing removes the dimensional influence of the operating data through the Max-Min method. The fluid state data set includes the instantaneous air pressure pr at the pump inlet, the instantaneous air pressure pc at the pump outlet, the flow velocity vc at the pipe inlet, and the fluid viscosity uf. The load evolution data set includes motor load current Ir, spindle housing temperature Ts, and vibration acceleration ar.

[0008] Preferably, the disturbance identification module includes a disturbance analysis unit and a disturbance evaluation unit; The disturbance analysis unit is used to perform correlation calculations based on the fluid state data set, analyze the viscosity and pressure difference coupled disturbance front formed by the medium disturbance wave in the pipeline, and generate the medium disturbance front index VPI, specifically: In the formula, log represents the logarithmic function.

[0009] Preferably, the disturbance assessment unit is used to evaluate the maximum value of the medium disturbance front index VPI (Very Highest Precipitation Index) based on the state where the disturbance front has not formed within the past 30 days. max and median VPI med The dynamic medium disturbance response threshold TH is established as follows: Then, the medium disturbance is evaluated by comparing it with the real-time acquired medium disturbance front index VPI. The specific evaluation scheme is as follows: When the medium disturbance front index VPI ≤ the medium disturbance response threshold TH, it indicates that the disturbance front has not formed, and normal monitoring should be maintained at this time; When the media disturbance front index VPI is greater than the media disturbance response threshold TH, it indicates that shear disturbance has been formed and there is an abnormal propagation trend of the media. At this time, the load evolution analysis command is executed.

[0010] Preferably, the load hysteresis identification module is used to execute load evolution analysis instructions when the medium disturbance assessment indicates that shear disturbance has been formed; The load evolution analysis command is used to perform correlation calculations based on the load evolution data set to obtain the Load Lag Evolution Index (LEI). This index is used to identify abnormal load responses of the export pumps caused by the ingress of disturbance fronts and to extract cumulative lag evolution characteristics. Specifically: In the formula, t represents the current time, ∆t represents the monitoring time window, Ir(τ), Ts(τ) and ar(τ) represent the motor load current, spindle housing temperature and vibration acceleration at time τ, respectively, and dτ represents the time calculus symbol.

[0011] Preferably, the joint risk assessment module includes an instability risk analysis unit and a load response assessment unit; The instability risk analysis unit is used to perform correlation calculations based on the media disturbance frontier index (VPI) and the load hysteresis evolution index (LEI), amplifying the disturbance propagation trend and the load hysteresis evolution in a linked manner, and determining the current state conditions and operating performance of the pump cavity in a dynamic imbalance state, so as to form a joint instability risk index (UDR), specifically: UDR = (VPI + 1). 2 *log(1+LEI), where log represents the logarithmic function.

[0012] Preferably, the load response assessment unit is used to sort the joint instability risk index UDR over the past 30 days from smallest to largest, and preset the Q75 quantile as the first joint instability boundary threshold UA and the Q95 quantile as the second joint instability boundary threshold UB using the quantile method. Then, it performs a comprehensive instability response assessment with the real-time acquired joint instability risk index UDR. The specific assessment scheme is as follows: When the combined instability risk index UDR is less than or equal to the first combined instability boundary threshold UA, it indicates that the pump body response is stable, and the first early warning command is executed at this time. When the first joint instability boundary threshold UA < the joint instability risk index UDR < the second joint instability boundary threshold UB, it indicates that there is a load response imbalance, and the second early warning instruction is executed at this time; When the combined instability risk index UDR is greater than or equal to the second combined instability boundary threshold UB, it indicates that the pump body is in an unstable state, and the third early warning command is executed at this time.

[0013] Preferably, the control execution module is used to receive early warning instructions from the comprehensive instability response assessment in real time, and to perform safety adjustments through the intelligent monitoring system, as follows; The first warning instruction initiates a 3-minute trend tracking window and performs iterative analysis through the disturbance identification module. The second early warning instruction issues a yellow warning to remind the dispatch terminal and the main control room to reduce the power of the external pump by 10% and freeze the downstream dispatch for 30 minutes, and performs iterative analysis through the disturbance identification module; The third early warning command issues a red warning to force the dispatch terminal and main control room to switch to the backup pump, start the bypass circulation and cooling redundancy equipment, and notify maintenance personnel to carry out equipment maintenance.

[0014] A method for intelligent monitoring of the safety status of an external pump includes the following steps: S1. Based on the sensor group installed on the external pump, the operating condition data of the external pump is collected in real time and transmitted to the intelligent monitoring system for preprocessing to obtain the fluid state data group and load evolution data group. S2. Calculate the medium disturbance front index VPI based on the fluid state data set, and evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, execute the load evolution analysis command. S3. Execute the load evolution analysis command, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. S4. Calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. S5 receives early warning instructions from the comprehensive instability response assessment in real time and performs safety adjustments through the intelligent monitoring system.

[0015] This invention provides an intelligent monitoring method and system for the safety status of an external pump. It has the following beneficial effects: (1) The system's data acquisition module realizes multi-dimensional real-time perception of the key operating conditions of the external pump, covering the fluid side's transport status parameters such as pressure, viscosity, and flow rate, as well as the mechanical load response characteristics of the structure side such as current, temperature, and vibration. Based on the sensor groups installed at various locations of the external pump, the system collects the operating condition data of the external pump in real time. The data is then transmitted to the intelligent monitoring system for noise reduction, timestamp alignment, missing value filling, and dimensionless processing, respectively constructing fluid state data groups and load evolution data groups. This provides a high-quality input source for subsequent disturbance identification and load assessment, avoiding the impact of data distortion or offset on the judgment results. This module completes the basic task of collecting physical quantities into usable analytical data, laying the foundation for the first gate of the system's intelligent monitoring.

[0016] (2) The system's disturbance identification module constructs the medium disturbance front index VPI through fluid state data sets. For the first time, the system proposes to reveal the intensity of fluid disturbances by using the logarithmic mapping relationship between pressure difference and viscosity, reflecting the development trend of transport physical anomalies. It also evaluates the medium disturbances with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbances have formed, the system executes the load evolution analysis command. The load hysteresis identification module calculates the load hysteresis evolution index LEI based on the load evolution data sets, accumulating the differential characteristics of current, temperature rise, and vibration in a time integral manner to accurately characterize the structural layer performance of the disturbance hysteresis response. The two form a disturbance and response transmission chain through mathematical correlation logic, realizing asynchronous mapping identification from medium behavior to structural load, effectively solving the problem of ambiguity in the correlation between fluid disturbances and mechanical instability, establishing a cross-domain coupled calculation model, and completing the time-series identification task of state evolution trend.

[0017] (3) The joint risk assessment module of the system further calculates the joint instability risk index UDR as the core fusion index based on the media disturbance frontier index VPI and the load hysteresis evolution index LEI. By strengthening the linkage amplification effect between disturbance and hysteresis through exponential and logarithmic functions, a discrimination mechanism for the dynamic imbalance state of the pump chamber at the system level is constructed. The system sets the first joint instability boundary threshold UA and the second joint instability boundary threshold UB based on the quantile method, realizing comprehensive risk identification from disturbance evolution to pump body dynamic imbalance. The control execution module triggers different early warning and operation strategies according to the early warning instructions of the comprehensive instability response assessment. When the pump body response is stable, the trend tracking window is started for continuous monitoring. When the load response is unbalanced, the power is automatically reduced and the scheduling plan is frozen to interrupt possible instability evolution. When the pump body is in an unstable state, a red early warning is automatically issued, and strong intervention measures such as forced standby pump switching, cooling bypass start-up, and maintenance notification are executed. This module directly transforms the assessment results into multi-level dynamic control behavior, effectively blocking the instability propagation path and ensuring the system's stable operation and the immediate response capability of safe redundancy deployment. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the intelligent monitoring system for the safety status of an external pump according to the present invention; Figure 2 This is a schematic diagram illustrating the steps of an intelligent monitoring method for the safety status of an external pump according to the present invention; Figure 3 This is a schematic diagram illustrating the operating principle of an intelligent monitoring system for the safety status of an external pump according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1 Please see Figure 1 This invention provides an intelligent monitoring system for the safety status of an external pump. To achieve the above objectives, this invention is implemented through the following technical solutions: including a data acquisition module, a disturbance identification module, a load hysteresis identification module, a joint risk assessment module, and a control execution module; The data acquisition module is used to collect the operating condition data of the external pump in real time based on the sensor group installed on the external pump, and transmit it to the intelligent monitoring system for preprocessing to obtain fluid state data group and load evolution data group. The disturbance identification module is used to calculate the medium disturbance front index VPI based on the fluid state data set, and to evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, the load evolution analysis command is executed. The load hysteresis identification module is used to execute load evolution analysis instructions, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. The joint risk assessment module is used to calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and to conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. The control execution module is used to receive early warning commands from the comprehensive instability response assessment in real time and to perform safety adjustments through the intelligent monitoring system.

[0021] In this embodiment, the data acquisition module deploys multiple types of industrial sensors to acquire real-time operating data of the external pump. This data is then processed by the intelligent monitoring system via a wireless network for noise reduction, timestamp alignment, missing value filling, and dimensionless processing, forming structured fluid state data sets and load evolution data sets. This significantly improves data integrity and the stability of the analytical foundation. Compared to traditional methods relying solely on single temperature or current monitoring methods, this module achieves richer-dimensional state perception and lateral correlation capabilities, ensuring high-precision data input. The disturbance identification module utilizes the fluid state data sets to construct a nonlinear pressure difference and viscosity coupling formula, forming the medium disturbance front index VPI, which accurately reflects the propagation trend of shear disturbances caused by flow hindrance or fluid discontinuity. This VPI is then compared with the constructed medium disturbance response threshold TH to dynamically assess whether a disturbance front has formed. When the medium disturbance front index VPI exceeds the medium disturbance response threshold TH, the load evolution analysis command is immediately executed, initiating the load hysteresis identification process. The load hysteresis identification module executes load evolution analysis commands and constructs the Load Hysteresis Evolution Index (LEI) from the load evolution data set to assess cumulative anomalies in the structural response. This module is highly sensitive to potential dynamic mismatches between disturbances and loads, filling the technical blind spot of existing monitoring technologies that cannot capture load hysteresis chain responses induced by shear waves, and identifying risk sources from both the medium transfer and structural response dimensions. The joint risk assessment module constructs a joint instability risk index (UDR) by nonlinearly amplifying and coupling the media disturbance frontier index (VPI) and the load hysteresis evolution index (LEI). This UDR is then used to comprehensively assess the instability response, combining it with the constructed first joint instability boundary threshold (UA) and second joint instability boundary threshold (UB). The current pump state is categorized into three levels: stable pump response, load response imbalance, and pump instability. The assessment results are transmitted to the control execution module in real time, triggering different levels of operation. The system executes the first warning command, initiating a three-minute trend tracking window to continuously monitor disturbances; executes the second warning command, automatically reducing pump power by 10%, freezing scheduling for 30 minutes, and re-evaluating; and triggers the third warning command, forcibly activating the standby pump, bypass circulation, and redundant cooling equipment, and notifying on-duty personnel to perform emergency maintenance. Compared to the limitations of traditional fixed alarm threshold systems, which are not easy to effectively identify and intervene in multivariate coupled instability, this system not only constructs a dynamic evaluation system based on adaptive evolution of operating history, but also realizes hierarchical control, active intervention and response closed loop at the regulation level, which significantly improves the stability, safety and accident response speed of pump operation, and is especially suitable for continuous delivery guarantee in extreme working conditions such as high temperature and high pressure, and complex viscous media transmission.

[0022] Example 2 Please refer to Figure 1 and Figure 3Specifically: the data acquisition module includes a working condition acquisition unit and a data synchronization unit; The operating condition acquisition unit is used to collect the operating condition data of the external pump in real time based on the sensor group installed at each position of the external pump; The sensor group includes a piezoresistive pressure sensor, an online rotary viscometer, an ultrasonic flow meter, a current transformer, a resistance thermometer, and a piezoelectric accelerometer. The piezoresistive pressure sensor is installed on the side wall of the inlet pipe near the inlet impeller inlet and on the high-pressure side of the pump outlet section near the outlet check valve. It collects the instantaneous gas pressure pr at the pump inlet and the instantaneous gas pressure pc at the pump outlet in real time, which represent the real-time gas pressure values ​​at the pump inlet and pump outlet, respectively, reflecting the suction pressure level of the medium before entering the pump chamber and the back pressure state maintained when the pump discharges the medium. The online rotary viscometer is installed on the side wall of the main fluid flow section at the pump inlet to collect the fluid viscosity uf in real time, which represents the magnitude of the internal friction force of the fluid and is a physical parameter of flow resistance. The ultrasonic flow meter is installed on the inner wall of the pipe section 0.5m before the pump inlet to collect the inlet flow velocity vc in real time, which represents the volume flow intensity of fluid passing through a unit cross section per unit time. The current transformer is installed between the three-phase busbars of the pump's main power supply input line to collect the motor load current Ir in real time, representing the electrical signal characteristics of the pump's total load. The resistance thermometer is embedded inside the spindle bushing to collect the spindle housing temperature Ts in real time, which indicates the heat accumulation during mechanical movement and reflects the thermal effects caused by friction and load. The piezoelectric accelerometer is used to be adhesively installed on the outer shell panel in the vertical direction of the pump body bearing seat to collect vibration acceleration ar in real time, which represents the intensity of mechanical vibration caused by unbalanced forces acting on the equipment during operation.

[0023] The data synchronization unit is used to establish a communication connection between the sensor group and the intelligent monitoring system based on the wireless network, and transmit the working condition data collected by the sensor group to the intelligent monitoring system in real time for preprocessing to obtain fluid state data group and load evolution data group. The preprocessing includes denoising, timestamp alignment, filling in missing values, and dimensionless processing; The denoising process preserves the transient structural features of the operating data and removes high-frequency noise through wavelet denoising technology. The timestamp alignment performs interpolation and prediction to fill in missing time points in the operating data through linear interpolation resampling technology. The missing value filling process completes the operating data affected by sampling interruption and communication packet loss through missing value interpolation technology. The dimensionless processing removes the dimensional influence of the operating data through the Max-Min method. The fluid state data set includes the instantaneous air pressure pr at the pump inlet, the instantaneous air pressure pc at the pump outlet, the flow velocity vc at the pipe inlet, and the fluid viscosity uf. The load evolution data set includes motor load current Ir, spindle housing temperature Ts, and vibration acceleration ar.

[0024] In this embodiment, by deploying piezoresistive pressure sensors, online rotary viscometers, ultrasonic flow meters, current transformers, resistance thermometers, and piezoelectric accelerometers at key locations on the pump body, the system can acquire real-time operating data of the external pump, including instantaneous inlet air pressure pr, instantaneous outlet air pressure pc, pipe inlet flow velocity vc, fluid viscosity uf, motor load current Ir, spindle housing temperature Ts, and vibration acceleration ar, covering complete operating parameters from fluid properties to mechanical response. Simultaneously, with the help of a data synchronization unit, the operating data is wirelessly collected in real-time to the intelligent monitoring system for noise reduction, timestamp alignment, missing value filling, and dimensionless processing, forming fluid state data sets and load evolution data sets, ensuring stable data quality, clear structure, and time-series alignment. Compared with traditional monitoring methods based on only a single signal, this module significantly improves the dimensional completeness, timeliness, and noise suppression capabilities of data perception, providing reliable and rich data support for subsequent disturbance identification and load risk analysis, thus laying the foundation for intelligent pump status identification and proactive early warning.

[0025] Example 3 Please refer to Figure 1 and Figure 3 Specifically: the disturbance identification module includes a disturbance analysis unit and a disturbance evaluation unit; The disturbance analysis unit is used to perform correlation calculations based on the fluid state data set, analyze the viscosity and pressure difference coupled disturbance front formed by the medium disturbance wave in the pipeline, and form the medium disturbance front index VPI, which reflects the physical propagation trend of medium anomalies, specifically: In the formula, log represents the logarithmic function. This represents the pressure difference required to maintain the current conveying state at a unit flow rate, reflecting the flow resistance intensity of the medium and revealing flow anomalies. log(1+uf) is used to perform nonlinear mapping on fluid viscosity changes, enhancing the ability to identify high viscosity change segments. Using a logarithmic function for feature enhancement can improve the response to low-frequency disturbances. The integral square root is used to perform nonlinear compression on the results.

[0026] The disturbance assessment unit is used to evaluate the maximum value of the medium disturbance front index VPI (VPI) based on the state where the disturbance front has not formed within the past 30 days. max and median VPI med The dynamic medium disturbance response threshold TH is established as follows: Then, the medium disturbance is evaluated by comparing it with the real-time acquired medium disturbance front index VPI. The specific evaluation scheme is as follows: When the medium disturbance front index VPI ≤ the medium disturbance response threshold TH, it indicates that the disturbance front has not formed, and normal monitoring should be maintained at this time; When the media disturbance front index VPI is greater than the media disturbance response threshold TH, it indicates that shear disturbance has been formed and there is an abnormal propagation trend of the media. At this time, the load evolution analysis command is executed.

[0027] In this embodiment, the disturbance analysis unit constructs the medium disturbance front index VPI based on the coupling relationship of the fluid state data group, which comprehensively reflects the propagation trend of shear disturbance caused by increased viscosity or sudden pressure difference in the fluid in the pipeline. By performing logarithmic mapping on the fluid viscosity and coupling it with the unit pressure difference ratio, the sensitivity of identification of high viscosity and low speed segment is improved. At the same time, the introduction of square root nonlinear compression makes the abnormal features physically interpretable.

[0028] The physical meaning of the formula is to measure the physical propagation trend of the shear disturbance wavefront caused by abnormal pipeline transportation under high viscosity transportation conditions. This represents the pressure difference required to maintain pumping at a unit flow rate. It is a measure of fluid flow resistance. Compared to directly using the pressure difference, adding the inlet flow velocity vc as the denominator essentially standardizes the pressure difference, avoiding misjudgments due to flow velocity changes. This reflects a pressure difference correction logic based on the flow velocity perspective. log(1+uf) is used to perform nonlinear amplification mapping on changes in medium viscosity, enhancing the response sensitivity in high-viscosity sections. The product of log(1+uf) is combined into a trend expression, so that the output can maintain the ability to amplify sudden anomalies and also have the effect of suppressing the threshold of disturbance persistence. The overall meaning of the square root compression term is used to nonlinearly compress the output range, reducing the abruptness of the high-value medium disturbance front index VPI in the judgment of system anomalies.

[0029] The disturbance assessment unit constructs a dynamic response threshold TH based on the median and maximum values ​​of historical VPIs, and then performs a media disturbance assessment with the real-time acquired media disturbance front index VPI to intelligently determine whether a disturbance front has formed, thereby triggering subsequent responses. Compared with traditional fixed threshold methods or single-index monitoring methods, this module has advantages such as dynamic adaptation, feature enhancement, and physical coupling expression. It not only significantly improves the system's ability to identify low-frequency, slowly varying disturbances, but also strengthens the early response capability to potential pump instability trends. It constructs a primary judgment closed loop from fluid fluctuation identification to structural response linkage, providing accurate and reliable pre-trigger basis for subsequent early warning and control adjustment of the entire system. Specific implementation examples are shown in Table 1.

[0030] Table 1: Examples of Calculation of the Medium Disturbance Frontier Index (VPI); Example 4

[0031] Please refer to Figure 1 and Figure 3 Specifically: the load hysteresis identification module is used to execute load evolution analysis instructions when the medium disturbance assessment indicates that shear disturbance has been formed; The load evolution analysis command is used to perform correlation calculations based on the load evolution data set to obtain the Load Lag Evolution Index (LEI). This index is used to identify abnormal load responses of the export pumps caused by the ingress of disturbance fronts and to extract cumulative lag evolution characteristics. Specifically: In the formula, t represents the current time, ∆t represents the monitoring time window, Ir(τ), Ts(τ), and ar(τ) represent the motor load current, spindle housing temperature, and vibration acceleration at time τ, respectively, and dτ represents the time calculus symbol. Let be the first derivative of the motor load current, representing the rate of change of the motor load current at time τ. The first derivative of the spindle housing temperature represents the rate of change of the spindle housing temperature at time τ. The overall trend is superimposed using an integral form to form a cumulative expression of hysteresis.

[0032] In this embodiment, when the medium disturbance assessment indicates that shear disturbance has formed, the load hysteresis identification module executes a load evolution analysis command. It calls upon the motor load current Ir, spindle housing temperature Ts, and vibration acceleration ar from the load evolution data set to construct the load hysteresis evolution index LEI. This index comprehensively characterizes the delayed and cumulative changes in the pump structure response after disturbance intrusion. This approach not only establishes a logical loop between disturbance formation and abnormal load response but also achieves the fusion extraction of multi-dimensional features such as load response hysteresis, abnormal energy consumption, and abnormal vibration. It effectively overcomes the delayed identification problem caused by traditional methods that rely solely on instantaneous threshold judgment. This enhances the early warning, dynamic tracking, and hysteresis modeling capabilities for pump instability precursors, significantly improving the system's robustness and response accuracy under complex operational disturbance scenarios.

[0033] The purpose of the physical meaning of the formula is to quantify whether the abnormal load response exhibited by the external pump after the medium disturbance front is introduced has a time-delay accumulation trend. It represents the rate of change of the motor load current and is a real-time dynamic indicator reflecting the conversion of electrical energy into mechanical load. The rate of change of the spindle housing temperature represents a medium-to-slow response signal to changes in structural frictional load or energy consumption. ar(τ) represents the vibration acceleration, which is the response intensity of the pump body's mechanical structure under the influence of unbalanced disturbances and is a high-frequency feedback signal. The coupling and integration of these three factors indicates whether the electrical, thermal, and structural dynamic responses form a time-series linkage resonance characteristic within the entire monitoring window period. This reflects whether the disturbance can accumulate, amplify, and form a structural instability trend within a certain time range along the electrical, thermal, and vibration paths. The absolute value is used to avoid the cancellation of positive and negative signs, thus accurately reflecting the total intensity of the change. The cumulative integration is performed over the entire monitoring window Δt, and the trend accumulation of the three-factor linkage response in the time dimension is used to construct the total amount of the time-series response intensity density function, i.e., whether the linkage response is continuous rather than sporadic within this time period. Specific implementation examples are shown in Table 2.

[0034] Table 2: Example Table of Load Hysteresis Evolution Index (LEI) Calculation; Example 5

[0035] Please refer to Figure 1 and Figure 3 Specifically: the joint risk assessment module includes an instability risk analysis unit and a load response assessment unit; The instability risk analysis unit is used to perform correlation calculations based on the media disturbance frontier index (VPI) and the load hysteresis evolution index (LEI), amplifying the disturbance propagation trend and the load hysteresis evolution in a linked manner, and determining the current state conditions and operating performance of the pump cavity in a dynamic imbalance state, so as to form a joint instability risk index (UDR), specifically: UDR = (VPI + 1). 2 *log(1+LEI), where log represents the logarithmic function, (VPI+1) 2 The quadratic amplification effect strengthens the driving effect of the medium disturbance front index VPI on the instability of the entire system. log(1+LEI) strengthens the surge response to the load hysteresis evolution index LEI through the logarithmic function, preventing the sudden change in the judgment of system instability due to large fluctuations in the load hysteresis evolution index LEI.

[0036] The load response assessment unit is used to sort the joint instability risk index (UDR) over the past 30 days from smallest to largest, and preset the Q75 quantile as the first joint instability boundary threshold (UA) and the Q95 quantile as the second joint instability boundary threshold (UB) using the quantile method. Then, it performs a comprehensive instability response assessment with the real-time acquired joint instability risk index (UDR). The specific assessment scheme is as follows: When the combined instability risk index UDR is less than or equal to the first combined instability boundary threshold UA, it indicates that the pump body response is stable, and the first early warning command is executed at this time. When the first joint instability boundary threshold UA < the joint instability risk index UDR < the second joint instability boundary threshold UB, it indicates that there is a load response imbalance, and the second early warning instruction is executed at this time; When the combined instability risk index UDR is greater than or equal to the second combined instability boundary threshold UB, it indicates that the pump body is in an unstable state, and the third early warning command is executed at this time.

[0037] In this embodiment, the instability risk analysis unit uses the medium disturbance frontier index VPI and the load hysteresis evolution index LEI as inputs to construct the joint instability risk index UDR, and uses exponential amplification and logarithmic suppression mechanisms to realize the linkage amplification expression of disturbance trend and structural hysteresis. The physical meaning of the formula reflects the intensity of disturbance fluctuations caused by changes in flow velocity, pressure difference, and viscosity of the medium before it enters the pump chamber. (VPI+1) 2 Emphasizing the accelerating effect of the increasing disturbance trend on system pressure is significant because, in the early stages of disturbance formation, if the medium disturbance front index VPI is not high but a propagation front has already formed, sufficient attention needs to be paid to the system. When the medium disturbance front index VPI rises sharply, the squaring operation can significantly amplify its impact, thus reflecting the driving force of the disturbance trend on the entire instability chain. The logarithmic function of log(1+LEI) achieves a smooth and enhanced response to the load hysteresis evolution index LEI, preventing misleading judgments by high-frequency abnormal fluctuations and forming a smooth enhancement. The medium disturbance front index VPI reflects the pressure gradient and viscosity changes experienced by the fluid at the pump inlet. These changes are often precursors or manifestations of internal shear disturbances in the fluid, and are the physical driving force for potential future unsteady fluctuations within the pump chamber. The load hysteresis evolution index LEI represents the degree of asynchronous response of the mechanical structure to external disturbances. The increase in the medium disturbance front index VPI is the root cause of the surge in the load hysteresis evolution index LEI. The two constitute a dynamic "cause and effect" relationship chain, therefore they are logically strongly correlated.

[0038] The load response assessment unit, based on the historical percentile ranking of the Joint Instability Risk Index (UDR), sets Q75 and Q95 as the first joint instability boundary threshold UA and the second joint instability boundary threshold UB, establishing a three-stage response mechanism. It triggers control measures such as trend tracking, power reduction, and forced switching according to the real-time level of the UDR. This module avoids the drawbacks of traditional single-indicator misjudgments or fixed-threshold delayed responses, possessing advantages such as strong adaptability, sensitive judgment, and progressive response. It can significantly improve the stable operation capability, early warning accuracy, and control closed-loop performance of the external pumping system under high-disturbance conditions, strengthening the ability to capture early signs of system instability and proactively intervene. Specific implementation examples are shown in Table 3.

[0039] Table 3: Implementation Examples of the Joint Instability Risk Index (UDR) Example 6

[0040] Please refer to Figure 1 and Figure 3 Specifically: the control execution module is used to receive early warning instructions from the comprehensive instability response assessment in real time, and to perform safety adjustments through the intelligent monitoring system, as follows; The first warning instruction initiates a 3-minute trend tracking window and performs iterative analysis through the disturbance identification module. The second early warning instruction issues a yellow warning to remind the dispatch terminal and the main control room to reduce the power of the external pump by 10% and freeze the downstream dispatch for 30 minutes, and performs iterative analysis through the disturbance identification module; The third early warning command issues a red warning to force the dispatch terminal and main control room to switch to the backup pump, start the bypass circulation and cooling redundancy equipment, and notify maintenance personnel to carry out equipment maintenance.

[0041] In this embodiment, the control execution module executes differentiated control strategies based on different early warning commands triggered by the comprehensive instability response assessment results, forming a progressive dynamic safety intervention mechanism from mild to severe. During the pump's stable response phase, a three-minute trend tracking window is initiated via the first early warning command, iteratively updating the disturbance potential in conjunction with the disturbance identification module. During the load imbalance phase, the second early warning command links the dispatch terminal and the main control room, automatically reducing the pump power by 10% and freezing downstream dispatch for 30 minutes, effectively suppressing the disturbance amplification process. When the pump is in an unstable state, the third early warning command is immediately executed, forcibly switching to the standby pump, activating the cooling redundancy and bypass system, and notifying maintenance, forming a multi-system coordinated emergency intervention. Through the implementation of this module, not only is an automatic execution closed loop from assessment to control achieved, but the dynamic steady-state recovery capability and emergency response efficiency of the pump system are also enhanced. Compared with traditional manual alarms and single load limiting methods, this significantly improves the system's adaptive control level, safe operation redundancy capability, and proactive pre-accident handling capability.

[0042] Example 7 Please refer to Figure 2 A method for intelligent monitoring of the safety status of an external pump includes the following steps: S1. Based on the sensor group installed on the external pump, the operating condition data of the external pump is collected in real time and transmitted to the intelligent monitoring system for preprocessing to obtain the fluid state data group and load evolution data group. S2. Calculate the medium disturbance front index VPI based on the fluid state data set, and evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, execute the load evolution analysis command. S3. Execute the load evolution analysis command, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. S4. Calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. S5 receives early warning instructions from the comprehensive instability response assessment in real time and performs safety adjustments through the intelligent monitoring system.

[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent monitoring system for the safety status of an external pump, characterized in that: It includes a data acquisition module, a disturbance identification module, a load hysteresis identification module, a joint risk assessment module, and a control execution module; The data acquisition module is used to collect the operating condition data of the external pump in real time based on the sensor group installed on the external pump, and transmit it to the intelligent monitoring system for preprocessing to obtain fluid state data group and load evolution data group. The disturbance identification module is used to calculate the medium disturbance front index VPI based on the fluid state data set, and to evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, the load evolution analysis command is executed. The load hysteresis identification module is used to execute load evolution analysis instructions, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. The joint risk assessment module is used to calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and to conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. The control execution module is used to receive early warning commands from the comprehensive instability response assessment in real time and to perform safety adjustments through the intelligent monitoring system.

2. The intelligent monitoring system for the safety status of an external pump according to claim 1, characterized in that: The data acquisition module includes a working condition acquisition unit and a data synchronization unit; The operating condition acquisition unit is used to collect the operating condition data of the external pump in real time based on the sensor group installed at each position of the external pump; The sensor group includes a piezoresistive pressure sensor, an online rotary viscometer, an ultrasonic flow meter, a current transformer, a resistance thermometer, and a piezoelectric accelerometer. The piezoresistive pressure sensor is installed on the side wall of the intake pipe near the inlet impeller inlet and on the high-pressure side of the pump outlet section near the outlet check valve to collect the instantaneous inlet air pressure pr and the instantaneous outlet air pressure pc in real time. The online rotary viscometer is installed on the side wall of the main fluid flow section at the pump inlet to collect the fluid viscosity in real time (uf). The ultrasonic flow meter is installed on the inner wall of the pipe section 0.5m before the pump inlet to collect the inlet flow velocity vc in real time. The current transformer is used to be installed between the three-phase busbars of the pump's main power supply input line to collect the motor load current Ir in real time; The resistance thermometer is embedded inside the spindle bushing to collect the spindle housing temperature Ts in real time. The piezoelectric accelerometer is used for adhesive mounting on the outer shell panel of the pump body bearing seat in the vertical direction to collect vibration acceleration ar in real time.

3. The intelligent monitoring system for the safety status of an external pump according to claim 2, characterized in that: The data synchronization unit is used to establish a communication connection between the sensor group and the intelligent monitoring system based on the wireless network, and transmit the working condition data collected by the sensor group to the intelligent monitoring system in real time for preprocessing to obtain fluid state data group and load evolution data group. The preprocessing includes denoising, timestamp alignment, filling in missing values, and dimensionless processing; The denoising process preserves the transient structural features of the operating data and removes high-frequency noise through wavelet denoising technology. The timestamp alignment performs interpolation and prediction to fill in missing time points in the operating data through linear interpolation resampling technology. The missing value filling process completes the operating data affected by sampling interruption and communication packet loss through missing value interpolation technology. The dimensionless processing removes the dimensional influence of the operating data through the Max-Min method. The fluid state data set includes the instantaneous air pressure pr at the pump inlet, the instantaneous air pressure pc at the pump outlet, the flow velocity vc at the pipe inlet, and the fluid viscosity uf. The load evolution data set includes motor load current Ir, spindle housing temperature Ts, and vibration acceleration ar.

4. The intelligent monitoring system for the safety status of an external pump according to claim 3, characterized in that: The disturbance identification module includes a disturbance analysis unit and a disturbance evaluation unit; The disturbance analysis unit is used to perform correlation calculations based on the fluid state data set, analyze the viscosity and pressure difference coupled disturbance front formed by the medium disturbance wave in the pipeline, and generate the medium disturbance front index VPI, specifically: In the formula, log represents the logarithmic function.

5. The intelligent monitoring system for the safety status of an external pump according to claim 4, characterized in that: The disturbance assessment unit is used to evaluate the maximum value of the medium disturbance front index VPI (VPI) based on the state where the disturbance front has not formed within the past 30 days. max and median VPI med The dynamic medium disturbance response threshold TH is established as follows: Then, the medium disturbance is evaluated by comparing it with the real-time acquired medium disturbance front index VPI. The specific evaluation scheme is as follows: When the medium disturbance front index VPI ≤ the medium disturbance response threshold TH, it indicates that the disturbance front has not formed, and normal monitoring should be maintained at this time; When the media disturbance front index VPI is greater than the media disturbance response threshold TH, it indicates that shear disturbance has been formed and there is an abnormal propagation trend of the media. At this time, the load evolution analysis command is executed.

6. The intelligent monitoring system for the safety status of an external pump according to claim 5, characterized in that: The load hysteresis identification module is used to execute load evolution analysis instructions when the medium disturbance assessment indicates that shear disturbance has been formed; The load evolution analysis command is used to perform correlation calculations based on the load evolution data set to obtain the Load Lag Evolution Index (LEI). This index is used to identify abnormal load responses of the export pumps caused by the ingress of disturbance fronts and to extract cumulative lag evolution characteristics. Specifically: In the formula, t represents the current time, ∆t represents the monitoring time window, Ir(τ), Ts(τ) and ar(τ) represent the motor load current, spindle housing temperature and vibration acceleration at time τ, respectively, and dτ represents the time calculus symbol.

7. The intelligent monitoring system for the safety status of an external pump according to claim 6, characterized in that: The joint risk assessment module includes an instability risk analysis unit and a load response assessment unit; The instability risk analysis unit is used to perform correlation calculations based on the media disturbance frontier index (VPI) and the load hysteresis evolution index (LEI), amplifying the disturbance propagation trend and the load hysteresis evolution in a linked manner, and determining the current state conditions and operating performance of the pump cavity in a dynamic imbalance state, so as to form a joint instability risk index (UDR), specifically: UDR = (VPI + 1). 2 *log(1+LEI), where log represents the logarithmic function.

8. The intelligent monitoring system for the safety status of an external pump according to claim 7, characterized in that: The load response assessment unit is used to sort the joint instability risk index (UDR) over the past 30 days from smallest to largest, and preset the Q75 quantile as the first joint instability boundary threshold (UA) and the Q95 quantile as the second joint instability boundary threshold (UB) using the quantile method. Then, it performs a comprehensive instability response assessment with the real-time acquired joint instability risk index (UDR). The specific assessment scheme is as follows: When the combined instability risk index UDR is less than or equal to the first combined instability boundary threshold UA, it indicates that the pump body response is stable, and the first early warning command is executed at this time. When the first joint instability boundary threshold UA < the joint instability risk index UDR < the second joint instability boundary threshold UB, it indicates that there is a load response imbalance, and the second early warning instruction is executed at this time; When the combined instability risk index UDR is greater than or equal to the second combined instability boundary threshold UB, it indicates that the pump body is in an unstable state, and the third early warning command is executed at this time.

9. The intelligent monitoring system for the safety status of an external pump according to claim 8, characterized in that: The control execution module is used to receive early warning commands from the comprehensive instability response assessment in real time, and to perform safety adjustments through the intelligent monitoring system, as follows; The first warning instruction initiates a 3-minute trend tracking window and performs iterative analysis through the disturbance identification module. The second early warning instruction issues a yellow warning to remind the dispatch terminal and the main control room to reduce the power of the external pump by 10% and freeze the downstream dispatch for 30 minutes, and performs iterative analysis through the disturbance identification module; The third early warning command issues a red warning to force the dispatch terminal and main control room to switch to the backup pump, start the bypass circulation and cooling redundancy equipment, and notify maintenance personnel to carry out equipment maintenance.

10. A method for intelligent monitoring of the safety status of an external pump, applied to an intelligent monitoring system for the safety status of an external pump as described in any one of claims 1-9, characterized in that: Includes the following steps: S1. Based on the sensor group installed on the external pump, the operating condition data of the external pump is collected in real time and transmitted to the intelligent monitoring system for preprocessing to obtain the fluid state data group and load evolution data group. S2. Calculate the medium disturbance front index VPI based on the fluid state data set, and evaluate the medium disturbance with the established medium disturbance response threshold TH. When the evaluation indicates that shear disturbance has been formed, execute the load evolution analysis command. S3. Execute the load evolution analysis command, calculate the load hysteresis evolution index LEI based on the load evolution data set, and analyze the abnormal load response of the export pump. S4. Calculate the joint instability risk index UDR based on the medium disturbance frontier index VPI and the load hysteresis evolution index LEI, and conduct a comprehensive instability response assessment with the established first joint instability boundary threshold UA and second joint instability boundary threshold UB. S5 receives early warning instructions from the comprehensive instability response assessment in real time and performs safety adjustments through the intelligent monitoring system.

Images