Submersible pump corrosion early warning and life prediction system based on big data

Abstract

The application relates to the technical field of equipment life prediction, and discloses a submersible pump corrosion early warning and life prediction system based on big data, which comprises a first detection module, a second detection module, a data processing module, a corrosion early warning module and a life prediction module; wherein the first detection module measures the surface parameter of a pump body of the submersible pump based on a detection strategy; the second detection module measures the operation parameter of the submersible pump based on the detection strategy; the data processing module is used for calculating the surface index and the operation index of the pump body; the corrosion early warning module performs corrosion early warning on the submersible pump based on the surface index and the operation index of the pump body, and generates the detection strategy; and the life prediction module is used for collecting environmental parameters and performing life prediction on the submersible pump in combination with the surface index and the operation index of the pump body. The application solves the problems that the existing technology is difficult to find and dynamically evaluate the corrosion problem of the submersible pump in the early stage, and provides technical support for safe operation of the submersible pump.

Description

Technical Field

[0001] This application relates to the technical field of equipment life prediction, specifically to a submersible pump corrosion early warning and life prediction system based on big data. Background Technology

[0002] With the widespread application of water resource development, drainage projects, and underwater transport systems, submersible pumps, as key hydraulic equipment, have been widely used in municipal, industrial, agricultural, and mining fields. However, submersible pumps still face many challenges in actual operation, among which the corrosion of the pump body is particularly prominent. As the core equipment in the transport system, the overall performance and lifespan of the submersible pump affect the operating efficiency and safety of the entire system. Due to long-term exposure to the complex and harsh underwater environment, the pump casing, pump shaft end, and other key components of the submersible pump are susceptible to the effects of corrosive media, sandy water flow, high-velocity erosion, and temperature fluctuations, leading to material corrosion, reduced structural strength, and even defects such as cracks and perforations. This not only reduces the working efficiency of the submersible pump but also poses potential risks of equipment failure and safety accidents.

[0003] Currently, the main methods for addressing corrosion issues in submersible pumps are regular inspections and post-failure repairs to ensure safe operation. Regular inspections rely heavily on manual experience or portable testing instruments. While they can detect some surface corrosion defects, they struggle to effectively identify corrosion in hidden areas or internal structures. Furthermore, such inspections often require equipment shutdown, impacting system efficiency. Post-failure repairs primarily involve replacing or repairing pumps that have already failed, but they lack early warning capabilities and are costly. In recent years, some online monitoring technologies have been applied to submersible pump systems, using various sensors to monitor the pump's operating status in real time. However, existing online monitoring systems generally suffer from the following problems: limited monitoring data, lacking comprehensive analysis of the pump corrosion process; difficulty in timely assessment of the degree and trend of corrosion, inability to dynamically adjust corrosion identification strategies, and poor preventative maintenance effectiveness.

[0004] For example, patent application CN117370887A discloses a method, device, equipment, and medium for predicting the remaining life of a water pump. This method involves acquiring parameters of the water pump under test that characterize its current health status, and using these parameters as input to a health status prediction model trained using a training sample set consisting of multiple training sample groups corresponding to various test conditions. Since each training sample in the training sample set uses the corresponding sample water pump parameters under the corresponding test condition as feature data, and the current health status value of these parameters under a unified chemical condition across multiple different test conditions as classification label data, the health status prediction model outputs the current health status value of the water pump under test under the unified chemical condition. Therefore, based on the total health status value under the unified chemical condition, the current health status value of the water pump under test under the unified chemical condition, and the maintenance time of the unit health status value, the exact remaining life of the water pump under test can be accurately predicted.

[0005] For example, patent application CN120180951A discloses a hydraulic design data modeling method for well submersible pumps, which includes five steps: sensor acquisition, hydraulic data enhancement, hydraulic performance prediction, intelligent material matching, and model output. Specifically, it includes using physical information generative adversarial networks combined with spatiotemporal kriging interpolation algorithms to enhance three-dimensional flow field data; constructing a multimodal graph neural network that integrates physical control losses to achieve dynamic prediction of performance indicators such as pump efficiency, head, and net positive suction head; and evaluating the life, cost, and processability of candidate materials based on a corrosion-wear coupling model and reinforcement learning algorithms, intelligently recommending the optimal combination, and providing systematic support for the design of efficient, low-cost, and long-life well submersible pumps.

[0006] All of the above patents suffer from the problems raised in this background technology: it is difficult to judge the degree and development trend of corrosion in a timely manner, and it is impossible to dynamically adjust the corrosion identification strategy.

[0007] The information disclosed in this background section is intended only to enhance the understanding of the overall background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0008] The technical problem to be solved by this application is to overcome the defects of the prior art and provide a big data-based submersible pump corrosion early warning and life prediction system, which solves the problem of difficulty in early detection and dynamic assessment of submersible pump corrosion in the prior art and provides technical support for the safe operation of submersible pumps.

[0009] To solve the above-mentioned technical problems, this application provides the following technical solution:

[0010] A big data-based submersible pump corrosion early warning and life prediction system includes a first detection module, a second detection module, a data processing module, a corrosion early warning module, and a life prediction module.

[0011] The first detection module measures the surface parameters of the submersible pump body based on a detection strategy; the second detection module measures the operating parameters of the submersible pump based on a detection strategy.

[0012] The data processing module is used to process the pump body surface parameters and calculate the pump body surface indicators; the data processing module is also used to process the operating parameters and calculate the submersible pump's operating indicators.

[0013] The corrosion early warning module provides early warning of corrosion for submersible pumps based on the pump body surface indicators and operating indicators, and generates detection strategies based on the pump body surface indicators and operating indicators.

[0014] The life prediction module is used to collect environmental parameters and combine them with the pump body surface indicators and operating indicators to predict the life of the submersible pump.

[0015] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the first detection module includes a first detection unit and a first control unit; wherein the first detection unit is used to detect pump body surface parameters, and the first control unit is used to control the detection accuracy of pump body surface parameters;

[0016] The surface parameters include corrosion layer thickness data of the pump casing, corrosion layer thickness data of the pump shaft end, and pump body crack data; wherein, the corrosion layer thickness data of the pump casing and the corrosion layer thickness data of the pump shaft end are measured at multiple detection points using an ultrasonic thickness gauge, and the number of detection points is set by the first control unit; ultrasonic probes are arranged at multiple detection points on the pump body surface to record ultrasonic images, and the number and length of cracks on the pump body surface and inside are obtained through ultrasonic images, and the number of detection points is set by the first control unit.

[0017] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the second detection module includes a second detection unit and a second control unit; wherein the second detection unit is used to detect the operating parameters of the submersible pump, and the second control unit is used to control the detection cycle of the operating parameters.

[0018] The operating parameters include pump body vibration data, pump chamber pressure, pump shaft speed, pump shaft end temperature, and pump shaft torque. Pump body vibration data is collected using an accelerometer. Pump chamber pressure is collected by pressure sensors distributed inside the pump chamber or at the outlet, recording the location and pressure value of each sensor. Pump shaft torque is collected using a torque sensor; within any detection cycle, the second detection unit continuously collects m pump shaft torque values ​​at fixed time intervals, where m is a positive integer. Pump shaft end temperature is collected using a temperature sensor. Pump shaft speed is collected using a speed sensor, and the second detection unit monitors the pump shaft speed data in real time.

[0019] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the data processing module includes a first processing unit, which calculates pump body surface parameters based on pump body surface parameters. The surface parameters include pump casing corrosion layer thickness, pump casing surface roughness, pump shaft end corrosion layer thickness, and pump body crack index. The pump casing corrosion layer thickness is the average of corrosion layer thickness data at all detection points on the pump casing; the pump casing surface roughness is the variance of corrosion layer thickness data at all detection points on the pump casing; the pump shaft end corrosion layer thickness is the average of corrosion layer thickness data at all detection points on the pump shaft end; and the pump body crack index is the total length of all cracks in the pump body.

[0020] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the data processing module includes a second processing unit, which calculates operating indicators based on the submersible pump's operating parameters. These operating indicators include a vibration index, a pressure skew index, a torque fluctuation index, an overheat index, and a speed stability index. The vibration index is the maximum amplitude in the pump body vibration data. The pressure skew index is calculated by: calculating the pressure gradient at each location based on the pump chamber pressure values ​​collected by each pressure sensor; values ​​exceeding a set pressure gradient threshold are considered skewed pressure values, and the ratio of the number of skewed pressure values ​​to the total number of pressure values ​​is the pressure skew index. The overheat index is the maximum temperature at the pump shaft end. The speed stability index is the standard deviation of the pump shaft speed. The torque fluctuation index is calculated based on the difference between each pump shaft torque value and the mean of the pump shaft torque values.

[0021] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the corrosion early warning module includes an early warning unit and a strategy unit; the early warning unit is configured with an operational risk threshold range for each operational indicator; when the value of at least one operational indicator exceeds the corresponding operational risk threshold range, the strategy unit generates a first strategy instruction and sends the first strategy instruction to a first control unit; the first control unit responds to the first strategy instruction and generates a first adjustment strategy; the first adjustment strategy includes the detection accuracy of each pump body surface parameter; the first detection unit detects each pump body surface parameter of the submersible pump based on the first adjustment strategy.

[0022] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the early warning unit is further configured with a surface risk threshold range for each pump body surface indicator; when the value of at least one pump body surface indicator exceeds the corresponding surface risk threshold range, the strategy unit generates a second strategy instruction and sends the second strategy instruction to the second control unit; the second control unit responds to the second strategy instruction and generates a second adjustment strategy; the second adjustment strategy includes the detection cycle for each operating parameter; the second detection unit detects each operating parameter of the submersible pump based on the second adjustment strategy.

[0023] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the early warning unit is further configured with an early warning strategy for corrosion early warning of the submersible pump; the early warning strategy is as follows:

[0024] At the start of each detection cycle of the operating parameters, each pump body surface parameter is collected synchronously, and each pump body surface index for the corresponding detection cycle is calculated.

[0025] Each pump body surface index is obtained for the most recent M detection cycles, and the expansion rate of each pump body surface index is calculated. The expansion rate of any pump body surface index is calculated as follows: based on the increase of the pump body surface index in any two adjacent detection cycles; the average of the increase in the most recent M detection cycles is calculated and divided by the average cycle length of the M detection cycles to obtain the expansion rate of the corresponding pump body surface index.

[0026] The corrosion risk index of the submersible pump is calculated based on the expansion rate of each pump body surface index; the early warning unit is also equipped with a first early warning threshold; if the corrosion risk index is greater than the first early warning threshold, a corrosion early warning alarm for the submersible pump is generated.

[0027] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the life prediction module includes a data acquisition unit and a prediction model unit. The data acquisition unit is used to collect environmental parameters and preprocess the environmental parameters, pump surface indicators, and operating indicators. The environmental parameters include at least water temperature, water flow rate, and sediment content. The preprocessing includes data cleaning and normalization of the environmental parameters, pump surface indicators, and operating indicators, and encoding the environmental parameters, pump surface indicators, and operating indicators into feature vectors. The prediction model unit is equipped with a life prediction model, which is used to predict the life of the submersible pump. The input of the life prediction model is the feature vector of the environmental parameters, pump surface indicators, and operating indicators, and the output is the predicted value of the remaining life of the submersible pump.

[0028] As a preferred embodiment of the big data-based submersible pump corrosion early warning and life prediction system described in this application, the data acquisition unit is further configured with a risk threshold for each environmental parameter; if any environmental parameter is greater than the corresponding risk threshold, the data acquisition unit sends corrosion risk early warning information to the early warning unit; the early warning unit is further configured with a second early warning threshold, which is less than the first early warning threshold; in response to the corrosion risk early warning information, the early warning unit adjusts the value of the first early warning threshold to the second early warning threshold.

[0029] Compared with the prior art, the beneficial effects achieved by this application are as follows:

[0030] This application achieves comprehensive and real-time monitoring coverage by performing high-precision, multi-point measurements of the pump body's surface and operational parameters. When any surface or operational indicator is detected to be outside the normal range, the system automatically adjusts the detection frequency or accuracy of the relevant parameters, dynamically optimizing the monitoring strategy to ensure earlier detection of potential corrosion problems or operational anomalies. Through this feedback adjustment mechanism, the system not only effectively reduces unnecessary downtime for inspections but also makes maintenance work more scientific and targeted, thereby reducing the overall operating costs of the equipment.

[0031] This application, by monitoring key components of submersible pumps and combining big data analysis with intelligent corrosion early warning strategies, can identify corrosion risks and abnormal operating trends in advance, implement precise and effective preventive maintenance measures, effectively reduce safety risks caused by corrosion failure or sudden malfunctions, and ensure the stable operation of the submersible pump system. Attached Figure Description

[0032] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A schematic diagram of the structure of the big data-based submersible pump corrosion early warning and life prediction system provided in this application;

[0034] Figure 2 A functional diagram of the big data-based submersible pump corrosion early warning and life prediction system provided in this application. Detailed Implementation

[0035] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0036] This embodiment introduces a submersible pump corrosion early warning and life prediction system based on big data, referring to... Figure 1 The system includes a first detection module, a second detection module, a data processing module, a corrosion early warning module, and a lifespan prediction module; the functions of each module are as follows: Figure 2 As shown.

[0037] The first detection module measures the surface parameters of the submersible pump body based on a detection strategy.

[0038] The first detection module includes a first detection unit and a first control unit; wherein, the first detection unit is used to detect pump body surface parameters, and the first control unit is used to control the detection accuracy of pump body surface parameters;

[0039] The surface parameters include corrosion layer thickness data of the pump casing, corrosion layer thickness data of the pump shaft end, and pump body crack data; wherein, the corrosion layer thickness data of the pump casing and the corrosion layer thickness data of the pump shaft end are measured at multiple detection points using an ultrasonic thickness gauge, and the number of detection points is set by the first control unit; ultrasonic probes are arranged at multiple detection points on the pump body surface to record ultrasonic images, and the number and length of cracks on the pump body surface and inside are obtained through ultrasonic images, and the number of detection points is set by the first control unit.

[0040] The second detection module measures the operating parameters of the submersible pump based on the detection strategy;

[0041] The second detection module includes a second detection unit and a second control unit; wherein, the second detection unit is used to detect the operating parameters of the submersible pump, and the second control unit is used to control the detection cycle of the operating parameters;

[0042] The operating parameters include pump body vibration data, pump chamber pressure, pump shaft speed, pump shaft end temperature, and pump shaft torque. Pump body vibration data is collected using accelerometers. During operation, factors such as water flow impact and mechanical wear cause pump body vibration; the amplitude and frequency of these vibrations reflect the pump's health status. Monitoring these vibration parameters allows for the timely detection of potential faults, such as structural loosening, shaft imbalance, or component fatigue damage. Pump chamber pressure is collected by multiple pressure sensors distributed within the pump chamber or at the outlet, recording the location and corresponding pressure value of each sensor. The pressure distribution within the pump chamber is a key parameter for evaluating the submersible pump's operating status and hydraulic performance. By using pressure sensors positioned within the pump chamber or at the outlet, pressure changes under different operating conditions can be monitored in real time, accurately reflecting the forces acting on the pump body and the internal flow field, facilitating the detection of abnormal conditions such as blockage, cavitation, or fluid pulsation. Pump shaft torque is acquired based on a torque sensor. In any given detection cycle, the second detection unit continuously acquires m pump shaft torque values ​​at fixed time intervals, where m is a positive integer. Pump shaft torque reflects the submersible pump's ability to overcome water flow resistance and load variations. Real-time monitoring of pump shaft torque can effectively identify signs of abnormal load, decreased energy efficiency, or mechanical failure in the submersible pump. Pump shaft end temperature is acquired based on a temperature sensor.

[0043] The pump shaft end of a submersible pump is a critical transmission component, prone to heat generation during operation. Temperature changes at the end directly reflect mechanical friction, sealing conditions, and heat dissipation performance. Monitoring the pump shaft end temperature allows for the timely detection of potential faults such as overheating, seal failure, or insufficient lubrication. Pump shaft speed is acquired by a speed sensor, and a second detection unit monitors the pump shaft speed data in real time. The impeller of the submersible pump is connected to the motor via the pump shaft, and the pump shaft speed directly affects water output efficiency and operational stability. Real-time monitoring of the pump shaft speed and its stability effectively ensures the efficient and stable operation of the submersible pump, preventing performance degradation or system failure due to abnormal speed.

[0044] The data processing module is used to process the pump body surface parameters and calculate the pump body surface indicators; the data processing module is also used to process the operating parameters and calculate the submersible pump's operating indicators.

[0045] The data processing module includes a first processing unit, which calculates pump body surface parameters based on pump body surface parameters. These surface parameters include pump casing corrosion layer thickness, pump casing surface roughness, pump shaft end corrosion layer thickness, and pump body crack index. The pump casing corrosion layer thickness is the average of corrosion layer thickness data at all detection points on the pump casing. The pump casing surface roughness is the variance of corrosion layer thickness data at all detection points on the pump casing. The pump shaft end corrosion layer thickness is the average of corrosion layer thickness data at all detection points on the pump shaft end. The pump body crack index is the total length of all cracks in the pump body.

[0046] The data processing module further includes a second processing unit, which calculates operating indicators based on the submersible pump's operating parameters. These operating indicators include a vibration index, a pressure skew index, a torque fluctuation index, an overheat index, and a speed stability index. The vibration index is the maximum amplitude in the pump body vibration data. The pressure skew index is calculated by calculating the pressure gradient at each location based on the pump chamber pressure values ​​collected by each pressure sensor. Values ​​exceeding a set pressure gradient threshold are considered skewed pressure values, and the ratio of the number of skewed pressure values ​​to the total number of pressure values ​​is the pressure skew index. Those skilled in the art can set the specific value of the pressure gradient threshold based on experience or actual needs. The overheat index is the maximum temperature at the pump shaft end. The speed stability index is the standard deviation of the pump shaft speed. The torque fluctuation index is calculated based on the difference between each pump shaft torque value and the mean of the pump shaft torque values. The formula for calculating the torque fluctuation index is as follows:

[0047] ;

[0048] Where F represents the torque ripple index; This represents the i-th pump shaft torque value collected by the second detection unit; This represents the average of the m pump shaft torque values ​​collected by the second detection unit; α is an adjustment coefficient, set by those skilled in the art based on actual needs; the above formula uses linear weighting, giving higher weight to the measured pump shaft torque values ​​that are more recent in time;

[0049] The corrosion early warning module provides early warning of corrosion for submersible pumps based on the pump body surface indicators and operating indicators, and generates detection strategies based on the pump body surface indicators and operating indicators.

[0050] The corrosion early warning module includes an early warning unit and a strategy unit. The early warning unit is configured with an operational risk threshold range for each operational indicator. When the value of at least one operational indicator exceeds the corresponding operational risk threshold range, the strategy unit generates a first strategy instruction and sends the first strategy instruction to a first control unit. The first control unit responds to the first strategy instruction and generates a first adjustment strategy. The first adjustment strategy includes the detection accuracy of each pump body surface parameter. The first detection unit detects each pump body surface parameter of the submersible pump based on the first adjustment strategy. Those skilled in the art can specifically set the operational risk threshold range based on experience or actual needs.

[0051] The first control unit is also configured with a basic detection strategy for pump body surface parameters, which also includes the detection accuracy for each pump body surface parameter. During normal system operation, the first detection unit executes the basic detection strategy for pump body surface parameters; when the first control unit receives a first strategy instruction, the first detection unit switches to execute a first adjustment strategy. In the first adjustment strategy, the detection accuracy for each pump body surface parameter is higher than the detection accuracy of the corresponding pump body surface parameter in the basic detection strategy. In this embodiment, the preferred basic detection strategy for pump body surface parameters is as follows: the number of detection points for pump casing corrosion layer thickness is 20; the number of detection points for pump shaft end corrosion layer thickness is 15; and the number of detection points for pump body crack data is 25. The first adjustment strategy is as follows: the number of detection points for pump casing corrosion layer thickness is increased to 40; the number of detection points for pump shaft end corrosion layer thickness is increased to 30; and the number of detection points for pump body crack data is increased to 50. When operational indicators abnormally exceed the set operational risk threshold range, the system enhances its ability to detect defects such as corrosion and cracks on the surface of the submersible pump by improving the detection accuracy of pump body surface parameters, thereby improving the system's safety and operational stability.

[0052] The early warning unit is also configured with a surface risk threshold range for each pump body surface indicator. When the value of at least one pump body surface indicator exceeds the corresponding surface risk threshold range, the strategy unit generates a second strategy instruction and sends the second strategy instruction to the second control unit. The second control unit responds to the second strategy instruction and generates a second adjustment strategy. The second adjustment strategy includes the detection cycle for each operating parameter. The second detection unit detects each operating parameter of the submersible pump based on the second adjustment strategy. Those skilled in the art can specifically set the surface risk threshold range based on experience or actual needs.

[0053] The second control unit is also configured with a basic detection strategy for operating parameters, which also includes a detection cycle for each operating parameter. During normal system operation, the second detection unit executes the basic detection strategy for operating parameters; when the second control unit receives a second strategy instruction, the second detection unit switches to execute a second adjustment strategy. In the second adjustment strategy, the detection cycle for each operating parameter is shorter than the detection cycle for the corresponding operating parameter in the basic detection strategy, thereby increasing the detection frequency of operating parameters. In this embodiment, the preferred basic detection strategy for operating parameters is as follows: the detection cycle for any operating parameter is 14 days; the second adjustment strategy is as follows: the detection cycle for any operating parameter is shortened to 7 days. When surface indicators abnormally exceed the set surface risk threshold range, the system significantly increases the detection frequency of operating parameters by shortening the detection cycle, so as to promptly detect abnormal changes in the operating status of the submersible pump, such as vibration, pressure, speed, temperature, and torque, providing early warning of potential faults and ensuring the safe and reliable operation of the equipment.

[0054] The early warning unit is also equipped with an early warning strategy for corrosion early warning of submersible pumps; the early warning strategy is as follows:

[0055] At the start of each detection cycle of the operating parameters, each pump body surface parameter is collected synchronously, and each pump body surface index for the corresponding detection cycle is calculated.

[0056] Each pump surface index for the most recent M detection cycles is obtained, and the expansion rate of each pump surface index is calculated. The expansion rate of any pump surface index is calculated as follows: based on the increase of the pump surface index in any two adjacent detection cycles; that is, subtract the pump surface index corresponding to the previous detection cycle from the pump surface index corresponding to the latter detection cycle; calculate the average of the increase in the most recent M detection cycles, and divide it by the average cycle length of the M detection cycles to obtain the expansion rate of the corresponding pump surface index.

[0057] The corrosion risk index of the submersible pump is calculated based on the expansion rate of each pump body surface index. In this embodiment, the expansion rate of each pump body surface index is normalized and weighted and summed to obtain the corrosion risk index of the submersible pump. In the weighted summation, the weight value of the expansion rate of any pump body surface index is set by those skilled in the art based on actual needs.

[0058] The early warning unit is also equipped with a first early warning threshold; if the corrosion risk index is greater than the first early warning threshold, a corrosion early warning alarm for the submersible pump is generated. Those skilled in the art can set the specific value of the first early warning threshold based on experience or actual needs.

[0059] The life prediction module is used to collect environmental parameters and combine them with the pump body surface indicators and operating indicators to predict the life of the submersible pump.

[0060] The lifespan prediction module includes a data acquisition unit and a prediction model unit. The data acquisition unit collects environmental parameters and preprocesses the environmental parameters, pump surface indicators, and operational indicators. The environmental parameters include water temperature, water flow rate, and sediment content. The preprocessing includes data cleaning and normalization of the environmental parameters, pump surface indicators, and operational indicators, and encoding the environmental parameters, pump surface indicators, and operational indicators into feature vectors. The prediction model unit is equipped with a lifespan prediction model, which is used to predict the lifespan of the submersible pump. The input of the lifespan prediction model is the feature vector of the environmental parameters, pump surface indicators, and operational indicators, and the output is the predicted value of the remaining lifespan of the submersible pump.

[0061] The lifespan prediction model is a trained neural network model, and its training method includes:

[0062] Collect historical data, each of which contains a set of environmental parameters, pump body surface indicators, operating indicators, and the corresponding submersible pump lifespan.

[0063] The environmental parameters, pump body surface indicators, and operating indicators in the historical data are preprocessed, including data cleaning, normalization, and feature encoding operations.

[0064] Historical data is divided into training and test sets. Prediction model units are trained based on the training set, and the prediction accuracy of the model is evaluated using the test set.

[0065] Adjust the hyperparameters of the model based on the test results, optimize the performance of the lifetime prediction model, until the model reaches the set prediction accuracy standard;

[0066] Once the prediction accuracy meets the application requirements, the trained life prediction model is deployed in the system to predict the life of submersible pumps.

[0067] The data acquisition unit is also configured with a risk threshold for each environmental parameter; if any environmental parameter exceeds the corresponding risk threshold, the data acquisition unit sends a corrosion risk warning to the early warning unit; the early warning unit is also configured with a second early warning threshold, which is less than the first early warning threshold; in response to the corrosion risk warning, the early warning unit adjusts the value of the first early warning threshold to the second early warning threshold. Those skilled in the art can set the specific values ​​of the risk threshold for each environmental parameter and the second early warning threshold based on experience or actual needs.

[0068] This application introduces a dynamic early warning threshold adjustment mechanism based on environmental parameters, further improving the sensitivity and adaptability of the submersible pump corrosion early warning mechanism. This mechanism is mainly implemented through the collaboration of a data acquisition unit and an early warning unit, forming an adaptive corrosion early warning strategy that is linked to changes in environmental risk. It not only considers the deterioration trend of the pump body itself, i.e., the expansion trend of corrosion, but also fully responds to the potential risk of increased corrosion brought about by changes in the external environment. This enables dynamic adjustment of the corrosion early warning threshold, improves the system's response speed to environmental degradation, enhances the sensitivity and robustness of the corrosion early warning mechanism, and improves the reliability of submersible pump operation and the foresight of maintenance.

[0069] The system described in this embodiment enables real-time monitoring and prediction of the remaining life of key components of a submersible pump, such as the pump casing and pump shaft end. It comprehensively assesses the corrosion status and operational health of the submersible pump, promptly identifies potential corrosion defects or performance degradation trends, and facilitates the early implementation of corresponding maintenance or replacement measures to ensure the long-term safe, stable, and efficient operation of the submersible pump system.

[0070] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. A submersible pump corrosion early warning and life prediction system based on big data, characterized in that, It includes a first detection module, a second detection module, a data processing module, a corrosion early warning module, and a life prediction module; The first detection module measures the surface parameters of the submersible pump body based on a detection strategy. The first detection module includes a first detection unit and a first control unit; wherein, the first detection unit is used to detect pump body surface parameters, and the first control unit is used to control the detection accuracy of pump body surface parameters; The second detection module measures the operating parameters of the submersible pump based on the detection strategy; The second detection module includes a second detection unit and a second control unit; wherein, the second detection unit is used to detect the operating parameters of the submersible pump, and the second control unit is used to control the detection cycle of the operating parameters; The data processing module is used to process the pump body surface parameters and calculate the pump body surface indicators; the data processing module is also used to process the operating parameters and calculate the submersible pump's operating indicators. The data processing module includes a first processing unit, which calculates pump body surface parameters based on pump body surface parameters. The data processing module further includes a second processing unit, which calculates operating indicators based on the operating parameters of the submersible pump. The corrosion early warning module provides early warning of corrosion for submersible pumps based on the pump body surface indicators and operating indicators, and generates detection strategies based on the pump body surface indicators and operating indicators. The corrosion early warning module includes an early warning unit and a strategy unit; the early warning unit is configured with an operational risk threshold range for each operational indicator; when the value of at least one operational indicator exceeds the corresponding operational risk threshold range, the strategy unit generates a first strategy instruction and sends the first strategy instruction to a first control unit; the first control unit responds to the first strategy instruction and generates a first adjustment strategy; the first adjustment strategy includes the detection accuracy of each pump body surface parameter; the first detection unit detects each pump body surface parameter of the submersible pump based on the first adjustment strategy; The early warning unit is also configured with a surface risk threshold range for each pump body surface indicator; when the value of at least one pump body surface indicator exceeds the corresponding surface risk threshold range, the strategy unit generates a second strategy instruction and sends the second strategy instruction to the second control unit; the second control unit responds to the second strategy instruction and generates a second adjustment strategy; the second adjustment strategy includes the detection cycle of each operating parameter; the second detection unit detects each operating parameter of the submersible pump based on the second adjustment strategy; The life prediction module is used to collect environmental parameters and combine them with the pump body surface indicators and operating indicators to predict the life of the submersible pump.

2. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 1, characterized in that, The surface parameters include corrosion layer thickness data of the pump casing, corrosion layer thickness data of the pump shaft end, and pump body crack data; wherein, the corrosion layer thickness data of the pump casing and the corrosion layer thickness data of the pump shaft end are measured at multiple detection points using an ultrasonic thickness gauge, and the number of detection points is set by the first control unit; ultrasonic probes are arranged at multiple detection points on the pump body surface to record ultrasonic images, and the number and length of cracks on the pump body surface and inside are obtained through ultrasonic images, and the number of detection points is set by the first control unit.

3. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 1, characterized in that, The operating parameters include pump body vibration data, pump chamber pressure, pump shaft speed, pump shaft end temperature, and pump shaft torque. Pump body vibration data is collected using an accelerometer. Pump chamber pressure is collected by pressure sensors distributed inside the pump chamber or at the outlet, recording the location and pressure value of each sensor. Pump shaft torque is collected using a torque sensor; within any detection cycle, the second detection unit continuously collects m pump shaft torque values ​​at fixed time intervals, where m is a positive integer. Pump shaft end temperature is collected using a temperature sensor. Pump shaft speed is collected using a speed sensor, and the second detection unit monitors the pump shaft speed data in real time.

4. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 3, characterized in that, The surface parameters include pump casing corrosion layer thickness, pump casing surface roughness, pump shaft end corrosion layer thickness, and pump body crack index; wherein, the pump casing corrosion layer thickness is the mean value of corrosion layer thickness data at all detection points on the pump casing; the pump casing surface roughness is the variance of corrosion layer thickness data at all detection points on the pump casing; the pump shaft end corrosion layer thickness is the mean value of corrosion layer thickness data at all detection points on the pump shaft end; and the pump body crack index is the total length of all cracks in the pump body.

5. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 3, characterized in that, The operating indicators include vibration index, pressure skew index, torque fluctuation index, overheat index, and speed stability index. The vibration index is the maximum amplitude in the pump body vibration data. The pressure skew index is calculated as follows: based on the pump chamber pressure values ​​collected by each pressure sensor, the pressure gradient at each location is calculated; values ​​exceeding a set pressure gradient threshold are considered skewed pressure values, and the ratio of the number of skewed pressure values ​​to the total number of pressure values ​​is the pressure skew index. The overheat index is the maximum temperature at the pump shaft end. The speed stability index is the standard deviation of the pump shaft speed. The torque fluctuation index is calculated based on the difference between each pump shaft torque value and the mean of the pump shaft torque values.

6. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 5, characterized in that, The early warning unit is also equipped with an early warning strategy for corrosion early warning of submersible pumps; the early warning strategy specifically includes: At the start of each detection cycle of the operating parameters, each pump body surface parameter is collected synchronously, and each pump body surface index for the corresponding detection cycle is calculated. Each pump body surface index is obtained for the most recent M detection cycles, and the expansion rate of each pump body surface index is calculated. The expansion rate of any pump body surface index is calculated as follows: based on the growth amount of the pump body surface index in any two adjacent detection cycles, the average value of the growth amount in the most recent M detection cycles is calculated, and then divided by the average cycle length of the M detection cycles to obtain the expansion rate of the corresponding pump body surface index. The corrosion risk index of the submersible pump is calculated based on the expansion rate of each pump body surface index. The early warning unit is also equipped with a first early warning threshold; If the corrosion risk index is greater than the first warning threshold, a corrosion warning alarm for the submersible pump is generated.

7. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 6, characterized in that, The lifespan prediction module includes a data acquisition unit and a prediction model unit. The data acquisition unit is used to collect environmental parameters and preprocess the environmental parameters, pump body surface indicators, and operating indicators. The environmental parameters include at least water temperature, water flow rate, and sediment content. The preprocessing includes data cleaning and normalization of the environmental parameters, pump body surface indicators, and operating indicators, and encoding the environmental parameters, pump body surface indicators, and operating indicators into feature vectors. The prediction model unit is equipped with a lifespan prediction model, which is used to predict the lifespan of the submersible pump. The input of the lifespan prediction model is the feature vector of the environmental parameters, pump body surface indicators, and operating indicators, and the output is the predicted value of the remaining lifespan of the submersible pump.

8. The submersible pump corrosion early warning and life prediction system based on big data as described in claim 7, characterized in that, The data acquisition unit is also configured with a risk threshold for each environmental parameter; if any environmental parameter is greater than the corresponding risk threshold, the data acquisition unit sends a corrosion risk warning to the warning unit; the warning unit is also configured with a second warning threshold, which is less than the first warning threshold. In response to the corrosion risk warning information, the early warning unit adjusts the value of the first early warning threshold to the second early warning threshold.

Images