A water intake pump ship remote monitoring system and method

Abstract

The application provides a water intake pump ship remote monitoring system and method, which comprises a data acquisition unit: multi-dimensional environmental parameter data around a water intake port of a water intake pump ship is acquired in real time; and a blockage state analysis unit: based on the acquired environmental information, a real-time blockage state of the water intake port of the water intake pump ship is analyzed. The water intake pump ship remote monitoring system and method provided in the application embodiment uses a particle swarm optimization long short-term memory network algorithm, uses the global super parameter optimization advantage of a particle swarm algorithm, fits the dynamic coupling relationship between a water intake port blockage working condition, water body environmental parameters and water pump unit cavitation failure, realizes prediction of a water pump cavitation critical power threshold and failure occurrence rate, and determines the blockage grade and cavitation risk, formulates a stepwise unit power regulation strategy, under the premise of ensuring water intake operation efficiency, suppresses occurrence of water pump unit cavitation, and avoids problems such as impeller cavitation wear and unit operation shock.

Description

Technical Field

[0001] This application relates to the field of water intake technology, and in particular to a remote monitoring system and method for a water intake pumping vessel. Background Technology

[0002] A water intake pumping vessel is a mobile water intake structure that is equipped with a water pump unit, control equipment, and water pipeline. It can float on the surface of rivers, lakes, and reservoirs. Relying on the buoyancy of the hull, it automatically rises and falls with the water level. With the anchoring system for positioning and flexible pipelines connecting to the onshore pipe network, it can achieve stable and continuous raw water extraction and transportation. It is widely used in municipal water supply, industrial production, farmland irrigation, and emergency water intake scenarios.

[0003] Existing water intake pumps rely on a fixed-power pump unit operating continuously. The impeller rotation generates negative pressure to draw water into the pump body, which is then pressurized to complete the water delivery. During long-term water intake, aquatic plants, branches, silt, and floating debris in the water easily accumulate and adhere to the filter screen and inlet of the intake port, causing gradual blockage. This reduces the inlet flow area and continuously increases the inlet resistance. However, the pump unit's output power cannot adaptively adjust operating parameters according to the inlet blockage. When the inlet is obstructed, the local pressure in the pump chamber and impeller inlet drops below the water's saturation vaporization pressure, leading to water vaporization and the generation of numerous bubbles, resulting in pump cavitation. As the pump unit continues to operate, these bubbles repeatedly impact the impeller and pump casing wall with the water flow, causing metal surface erosion and wear, impeller deformation and damage, and shortening the pump's service life. Summary of the Invention

[0004] This application aims to at least partially solve one of the technical problems in the aforementioned technologies.

[0005] Therefore, one objective of this application is to propose a remote monitoring system and method for water intake pump vessels, which can predict the critical power threshold and failure rate of water pump cavitation, and formulate a stepped power adjustment strategy based on the blockage level and cavitation risk. While ensuring water intake efficiency, it can suppress water pump cavitation, avoid impeller cavitation wear, unit operation vibration and other failures, and improve the stability and service life of water pump operation.

[0006] To achieve the above objectives, the first aspect of this application proposes a remote monitoring system for a water intake pumping vessel, comprising: a data acquisition unit for real-time acquisition of multi-dimensional environmental parameter data around the water intake port of the water intake pumping vessel; a blockage status analysis unit for analyzing the real-time blockage status of the water intake port of the water intake pumping vessel based on the acquired environmental information; and a deep analysis unit comprising a data preprocessing module, a cavitation prediction module, and a strategy formulation module. The data preprocessing module acquires the raw data output by the data acquisition unit, performs data filtering and cleaning on the raw data, and standardizes the cleaned effective data. The cavitation prediction module uses a particle swarm optimization long short-term memory network algorithm based on the acquired effective data to determine the critical cavitation power threshold and cavitation failure rate of the pumping unit. The strategy formulation module acquires the blockage status results and cavitation prediction results output by the blockage status analysis unit and formulates a power adjustment strategy for the pumping unit.

[0007] In addition, the remote monitoring system for a water intake pumping vessel proposed in this application may also have the following additional technical features:

[0008] Specifically, it also includes a system management unit: used for parameter setting, operation and maintenance, and access control of the remote monitoring system.

[0009] Specifically, the system management unit includes: a data security storage module: which encrypts and stores data using encryption technology, establishes an access control mechanism to restrict user access to sensitive data, and backs up data to ensure rapid recovery in case of data loss or damage; and a settings module: which configures and manages various parameters of the remote monitoring system, and sets the system's control parameters and notification content.

[0010] Specifically, the real-time acquisition of multi-dimensional environmental parameter data around the water intake port of the water intake pump vessel includes: acquiring water pressure parameters around the water intake port using a pressure sensor; acquiring water temperature parameters around the water intake port using a temperature sensor; acquiring water quality parameters around the water intake port using a water quality detection sensor; acquiring water impurity parameters around the water intake port using an impurity monitoring sensor; and acquiring the operating power parameters of the water pump unit using a unit power acquisition instrument.

[0011] Specifically, the blockage status analysis unit includes: a feature extraction module for receiving multidimensional environmental parameter data collected by the data acquisition unit, and filtering, calibrating, aligning, and preprocessing the multidimensional environmental parameter data; a quantization calculation module for determining the pressure difference change at the water intake port of the water intake pumping vessel, the water impurity accumulation rate per unit time, and the water turbidity time-series change coefficient based on the acquired multidimensional environmental parameter data; and a judgment module for determining the actual blockage coverage area of ​​the water intake port at different monitoring time nodes based on the results output by the quantization calculation module, and determining the real-time blockage degree and operational blockage status of the water intake port at different times.

[0012] Specifically, the prediction steps of the particle swarm optimization long short-term memory network algorithm include: S1, acquiring processed multi-dimensional environmental parameter data, combining the blockage status data of the water intake port of the water intake pumping vessel and the historical operation data of the water pumping unit to construct a sample dataset, and dividing the dataset into a training set for model training and a validation set for model parameter calibration; S2, inputting the training set into the long short-term memory network basic model, using the particle swarm optimization algorithm to drive particle iterative optimization, updating the optimal position of individual particles and the global optimal position of the population in real time, and obtaining the optimal hyperparameter combination; S3, using the optimal hyperparameter combination to reconstruct and solidify the long short-term memory network model structure, obtaining the particle swarm optimization... S4. The preprocessed multidimensional environmental parameter data is input into the calibrated particle swarm optimization long short-term memory network cavitation fault time-series prediction model. Through the model's autonomous learning and fitting of the dynamic coupling relationship between the water intake port blockage state, multidimensional environmental parameters of the water body, and the operating power of the water pump unit, the critical power threshold for cavitation faults of the water pump unit under the current operating conditions and the probability of cavitation faults of the water pump unit are determined. S5. The cavitation critical power threshold and cavitation fault incidence rate data output by the model are verified, outlier removal is performed, and standardization and desensitization processing is carried out to obtain the cavitation prediction results.

[0013] Specifically, in S2, the hyperparameters for iterative optimization of the particle swarm algorithm include particle swarm size, upper and lower limits of particle flight speed, particle position value range, individual learning factor, global learning factor, and inertia weight coefficient.

[0014] Specifically, in step S3, calibrating the prediction model after the validation set has been reconstructed includes: inputting multidimensional environmental parameter data from the validation set into the prediction model, obtaining the cavitation prediction results output by the model, comparing the prediction results with the corresponding actual cavitation fault data in the validation set, and correcting the model parameters based on the comparison results.

[0015] Specifically, the strategy for adjusting the operating power of the water pump unit includes: increasing the unit's operating power when the water intake port is slightly blocked and the cavitation failure rate is below a preset safety threshold; gradually reducing the unit's operating power when the water intake port is moderately blocked and the cavitation failure rate is within the warning range; and reducing the unit's operating power when the water intake port is severely blocked and the cavitation failure rate is above the danger threshold.

[0016] The second aspect of this application proposes a monitoring method for a remote monitoring system of a water intake pumping vessel, comprising the following steps: Step 1: Real-time acquisition of water quality parameters, water pressure parameters, water temperature parameters, water impurity parameters, and pump unit operating power parameters around the water intake port of the water intake pumping vessel via a data acquisition unit; Step 2: Filtering, calibration, alignment, and pre-processing of the acquired data; based on the pre-processed data, analyzing the blockage state of the water intake port of the water intake pumping vessel at different times; Step 3: Constructing a time-series prediction model using a particle swarm optimization long short-term memory network algorithm; utilizing model iterative training and parameter calibration to fit the dynamic coupling relationship between the blockage state of the water intake port, water environment parameters, and pump unit operating power, and determining the current cavitation critical power threshold and cavitation failure rate of the pump unit; Step 4: Receiving the blockage state analysis results and cavitation prediction results, and generating a corresponding pump unit power adjustment strategy; Step 5: Completing full system operation status monitoring, fault self-checking, access control, and full-process data archiving and storage through a system management unit, realizing remote monitoring of the water intake pumping vessel.

[0017] Compared with the prior art, this application has the following beneficial effects: The remote monitoring system and method for water intake pumping vessels of this application, by setting up a data acquisition unit, collects water environment parameters and pump unit operating parameters in real time, and combines them with a blockage state analysis unit to quantitatively determine the degree of blockage at the water intake port, ensuring the accuracy of the blockage state identification at the water intake port. At the same time, a particle swarm optimization long short-term memory network algorithm is adopted, which utilizes the global hyperparameter optimization advantage of the particle swarm algorithm to fit the dynamic coupling relationship between the blockage condition at the water intake port, water environment parameters and cavitation faults of the pump unit, so as to predict the critical power threshold and failure rate of the pump cavitation. Combined with the blockage level and cavitation risk, a stepped unit power adjustment strategy is formulated to suppress the occurrence of pump unit cavitation while ensuring the efficiency of water intake operation, avoid impeller cavitation wear, unit operation vibration and other fault problems, reduce the pressure of manual operation and maintenance and the probability of equipment failure, and improve the stability and operating efficiency of water intake pumping vessels in long-term water intake operations.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0020] Figure 1 This is a schematic diagram of a remote monitoring system and method for a water intake pumping vessel according to an embodiment of this application;

[0021] Figure 2This is a flowchart of a remote monitoring system and method for a water intake pumping vessel according to one embodiment of this application. Detailed Implementation

[0022] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0023] The following describes a remote monitoring system and method for a water intake pumping vessel according to an embodiment of this application, with reference to the accompanying drawings.

[0024] like Figure 1 As shown in the figure, a remote monitoring system for a water intake pumping vessel according to an embodiment of this application may include a data acquisition unit for real-time acquisition of multi-dimensional environmental parameter data around the water intake port of the water intake pumping vessel; a blockage status analysis unit for analyzing the real-time blockage status of the water intake port of the water intake pumping vessel based on the acquired environmental information; and a deep analysis unit including a data preprocessing module, a cavitation prediction module, and a strategy formulation module. The data preprocessing module is used to acquire the raw acquired data output by the data acquisition unit, remove and clean the raw data, and standardize the cleaned effective data. The cavitation prediction module uses a particle swarm optimization long short-term memory network algorithm based on the acquired effective data to determine the critical cavitation power threshold and cavitation failure rate of the pumping unit. The strategy formulation module acquires the blockage status results and cavitation prediction results output by the blockage status analysis unit and formulates a power adjustment strategy for the pumping unit.

[0025] As illustrated by the example, through the coordinated operation of multiple units, the monitoring, prediction, and control of the blockage status of the water intake port of the water intake pumping vessel and the cavitation failure of the pumping unit are realized. It is possible to predict the cavitation risk of the unit in advance and formulate the optimal power adjustment strategy based on the actual blockage condition of the water intake port, thereby improving the stability and safety of the water intake operation of the water intake pumping vessel.

[0026] In one embodiment of this application, a system management unit is further included: used for parameter setting, operation and maintenance, and access control of the remote monitoring system. The system management unit includes: a data security storage module: which encrypts and stores data using encryption technology, establishes an access control mechanism to restrict user access to sensitive data, and backs up the data to ensure rapid recovery in case of data loss or damage; and a settings module: used to configure and manage various parameters of the remote monitoring system, and set system control parameters and notification content.

[0027] As illustrated by the example, the data security storage module can prevent core data such as monitoring data, operating parameters, and control logs from being leaked, tampered with, or lost during transmission and storage, ensuring the integrity, security, and traceability of system data. At the same time, the settings module can update the system control thresholds and alarm notification mechanisms according to different water intake conditions, equipment operating status, and maintenance needs, ensuring the stable operation of the monitoring system.

[0028] In one embodiment of this application, multi-dimensional environmental parameter data around the water intake port of the water intake pumping vessel are collected in real time, including: collecting water pressure parameters around the water intake port using a pressure sensor; collecting water temperature parameters around the water intake port using a temperature sensor; collecting water quality parameters around the water intake port using a water quality detection sensor; collecting water impurity parameters around the water intake port using an impurity monitoring sensor; and collecting the operating power parameters of the water pumping unit using a unit power acquisition instrument.

[0029] For example, by collecting data on the dynamic changes in water pressure fluctuations, water temperature variations, water quality, impurity accumulation, and unit power load at the water intake port, raw data support is provided for subsequent quantitative analysis of water intake port blockage, prediction of pump unit cavitation risk, and formulation of power regulation strategies, ensuring the accuracy of monitoring and analysis by the monitoring system.

[0030] In one embodiment of this application, the blockage status analysis unit includes: a feature extraction module for receiving multidimensional environmental parameter data collected by the data acquisition unit, and filtering, calibrating, aligning, and preprocessing the multidimensional environmental parameter data; a quantization calculation module for determining the pressure difference change at the water intake port of the water intake pumping vessel, the water impurity accumulation rate per unit time, and the water turbidity time-series change coefficient based on the acquired multidimensional environmental parameter data; and a judgment module for determining the actual blockage coverage area of ​​the water intake port at different monitoring time nodes based on the results output by the quantization calculation module, and determining the real-time blockage degree and operational blockage status of the water intake port at different times.

[0031] To illustrate, the feature extraction module removes noise interference, data deviation, and timing misalignment from the original sensor data, ensuring the validity and consistency of the input data. Furthermore, the quantification module transforms the abstract changes in the water environment and the clogging trend into quantifiable numerical indicators. Then, the judgment module combines multi-dimensional quantification parameters to determine the clogging coverage area, clogging level, and dynamic operating status, providing a basis for subsequent prediction of cavitation risk of water pump units and formulation of power adaptive adjustment strategies.

[0032] In one embodiment of this application, the prediction steps of the particle swarm optimization long short-term memory network algorithm include: S1, acquiring processed multi-dimensional environmental parameter data, combining the blockage status data of the water intake port of the water intake pumping vessel and the historical operation data of the water pumping unit to construct a sample dataset, and dividing the dataset into a training set for model training and a validation set for model parameter calibration; S2, inputting the training set into the long short-term memory network basic model, using the particle swarm optimization algorithm to drive particle iterative optimization, updating the optimal position of individual particles and the global optimal position of the population in real time, and obtaining the optimal hyperparameter combination; S3, using the optimal hyperparameter combination to reconstruct and solidify the long short-term memory network model structure, obtaining the particle... The particle swarm optimization long short-term memory network cavitation fault time-series prediction model is calibrated by inputting the validation set into the reconstructed prediction model. S4: Preprocessed multidimensional environmental parameter data is input into the calibrated particle swarm optimization long short-term memory network cavitation fault time-series prediction model. Through model autonomous learning and fitting the dynamic coupling relationship between the water intake port blockage state, multidimensional water environmental parameters, and the pump unit operating power, the critical power threshold for cavitation faults and the probability of cavitation faults in the pump unit under the current operating conditions are determined. S5: The cavitation critical power threshold and cavitation fault incidence rate data output by the model are verified, outlier removal is performed, and standardization and desensitization processing are carried out to obtain the cavitation prediction results.

[0033] In one embodiment of this application, in S2, the hyperparameters for iterative optimization of the particle swarm optimization algorithm include particle swarm size, upper and lower limits of particle flight speed, particle position value range, individual learning factor, global learning factor, and inertia weight coefficient.

[0034] In one embodiment of this application, in step S3, the calibration of the prediction model after the validation set is reconstructed includes: inputting multidimensional environmental parameter data from the validation set into the prediction model, obtaining the cavitation prediction results output by the model, comparing the prediction results with the corresponding actual cavitation fault data in the validation set, and correcting the parameters of the model based on the comparison results.

[0035] As illustrated by the example, the model's hyperparameters are adaptively and iteratively optimized by leveraging the global optimization capability of the particle swarm optimization algorithm. This improves the model's ability to learn from the coupled changes in complex aquatic environments, dynamic blockage conditions, and unit operating states. Simultaneously, by combining validation set data comparison and calibration with output data cleaning and optimization mechanisms, the model's prediction error is reduced, and the stability and robustness of the predicted cavitation critical power threshold and failure rate are improved. This enables the model to adapt to the actual operating conditions of the water intake pump vessel and provides predictive data support for the formulation of subsequent pump unit power regulation strategies.

[0036] In one embodiment of this application, a power adjustment strategy for the water pump unit is formulated, including: increasing the unit's operating power when the water intake port is slightly blocked and the cavitation failure rate is lower than a preset safety threshold; gradually reducing the unit's operating power when the water intake port is moderately blocked and the cavitation failure rate is within the warning range; and reducing the unit's operating power when the water intake port is severely blocked and the cavitation failure rate is higher than the danger threshold.

[0037] For example, by matching the degree of blockage at the water intake port with the incidence of cavitation failure, the power is appropriately increased when there is slight blockage and no risk of cavitation, thereby improving the efficiency of water intake and supply. When there is moderate blockage and potential cavitation risk, a step-by-step power reduction method is adopted to avoid sudden power changes that could cause unit operation oscillations. When there is severe blockage and a high risk of cavitation, the power is reduced in time to suppress cavitation and avoid problems such as impeller cavitation, equipment wear, and unit failure shutdown, thereby improving the operating efficiency of the water intake pumping vessel and the operating safety of the pumping unit.

[0038] like Figure 2 As shown, a monitoring method for a remote monitoring system of a water intake pumping vessel includes the following steps: Step 1: Real-time acquisition of water quality parameters, water pressure parameters, water temperature parameters, water impurity parameters, and pump unit operating power parameters around the water intake port of the water intake pumping vessel via a data acquisition unit; Step 2: Filtering, calibration, alignment, and pre-processing of the acquired data; based on the pre-processed data, analyzing the blockage status of the water intake port of the water intake pumping vessel at different times; Step 3: Constructing a time-series prediction model using a particle swarm optimization long short-term memory network algorithm; utilizing model iterative training and parameter calibration to fit the dynamic coupling relationship between the blockage status of the water intake port, water environment parameters, and pump unit operating power, determining the current cavitation critical power threshold and cavitation failure rate of the pump unit; Step 4: Receiving the blockage status analysis results and cavitation prediction results, and generating corresponding pump unit power adjustment strategies; Step 5: Completing full system operation status monitoring, fault self-checking, access control, and full-process data archiving and storage through a system management unit, realizing remote monitoring of the water intake pumping vessel.

[0039] In summary, the remote monitoring system and method for a water intake pumping vessel proposed in this application, by setting up a data acquisition unit to collect water environment parameters and pump unit operating parameters in real time, and combining it with a blockage status analysis unit to quantitatively determine the degree of blockage at the water intake port, ensures the accuracy of the blockage status identification. Simultaneously, a particle swarm optimization long short-term memory network algorithm is employed, utilizing the global hyperparameter optimization advantage of the particle swarm algorithm to fit the dynamic coupling relationship between the water intake port blockage condition, water environment parameters, and pump unit cavitation faults. This enables the prediction of the critical power threshold for pump cavitation and the failure rate. Furthermore, by combining the blockage level and cavitation risk, a stepped unit power adjustment strategy is formulated. While ensuring water intake operation efficiency, this system suppresses the occurrence of pump unit cavitation, avoids impeller cavitation wear, unit operating vibration, and other fault problems, reduces manual maintenance pressure and equipment failure probability, and improves the stability and operating efficiency of the water intake pumping vessel during long-term water intake operations.

[0040] In the description of this specification, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0041] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0042] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A remote monitoring system for a water intake pumping vessel, characterized in that, include: Data acquisition unit: Real-time acquisition of multi-dimensional environmental parameter data around the water intake port of the water intake pumping vessel; Blockage Status Analysis Unit: Based on the acquired environmental information, analyze the real-time blockage status of the water intake port of the water intake pumping vessel; The deep analysis unit includes a data preprocessing module, a cavitation occurrence prediction module, and a strategy formulation module. Data preprocessing module: Used to acquire the raw data output by the data acquisition unit, remove and clean the raw data, and standardize the cleaned effective data; Cavitation occurrence prediction module: Based on the acquired effective data, the particle swarm optimization long short-term memory network algorithm is used to determine the critical power threshold for cavitation and the cavitation failure rate of the water pump unit; Strategy formulation module: Obtain the blockage status results and cavitation prediction results output by the blockage status analysis unit, and formulate the water pump unit operating power adjustment strategy.

2. The remote monitoring system for a water intake pumping vessel according to claim 1, characterized in that, It also includes a system management unit: used for parameter setting, operation and maintenance, and access control of the remote monitoring system.

3. The remote monitoring system for a water intake pumping vessel according to claim 2, characterized in that, The system management unit includes: Data security storage module: Encrypts data for storage and transmission using encryption technology, establishes access control mechanisms to restrict user access to sensitive data, and backs up data to ensure rapid recovery in case of data loss or damage; Configuration module: Used to configure and manage various parameters of the remote monitoring system, and to set the system's control parameters and notification content.

4. The remote monitoring system for a water intake pumping vessel according to claim 3, characterized in that, The real-time acquisition of multi-dimensional environmental parameter data around the water intake port of the water intake pumping vessel includes: The water pressure parameters around the water intake port are collected using a pressure sensor. Temperature sensors are used to collect water temperature parameters around the water intake port. Water quality parameters around the water intake port are collected using water quality detection sensors. Impurity monitoring sensors are used to collect water impurity parameters around the water intake port. The operating power parameters of the water pump unit are collected using a unit power acquisition instrument.

5. The remote monitoring system for a water intake pumping vessel according to claim 4, characterized in that, The blockage status analysis unit includes: Feature extraction module: used to receive multidimensional environmental parameter data collected by the data acquisition unit, and to filter, calibrate, align and preprocess the multidimensional environmental parameter data; Quantitative calculation module: Based on the acquired multidimensional environmental parameter data, determine the pressure difference change at the water intake port of the water intake pumping vessel, the accumulation rate of impurities in the water per unit time, and the time-series variation coefficient of water turbidity; Judgment Module: Based on the output of the quantization calculation module, the actual blockage coverage area of ​​the water intake port at different monitoring time nodes is determined, and the real-time blockage degree and operational blockage status of the water intake port at different times are determined.

6. The remote monitoring system for a water intake pumping vessel according to claim 5, characterized in that, The prediction steps of the particle swarm optimization long short-term memory network algorithm include: S1. Obtain the processed multidimensional environmental parameter data, combine the water intake port blockage status data of the water intake pumping vessel and the historical operation data of the water pumping unit, construct a sample dataset, and divide the dataset into samples to obtain a training set for model training and a validation set for model parameter calibration. S2. Input the training set into the long short-term memory network basic model, use the particle swarm optimization algorithm to drive the particle iterative optimization, update the optimal position of individual particles and the global optimal position of the population in real time, and obtain the optimal hyperparameter combination. S3. Using the optimal hyperparameter combination, reconstruct and solidify the long short-term memory network model structure to obtain a particle swarm optimization long short-term memory network cavitation fault time-series prediction model. Input the validation set into the reconstructed prediction model for calibration. S4. Input the preprocessed multidimensional environmental parameter data into the calibrated particle swarm optimization long short-term memory network cavitation fault time-series prediction model. Through the model's autonomous learning and fitting of the dynamic coupling relationship between the water intake port blockage state, multidimensional environmental parameters of the water body and the operating power of the water pump unit, determine the critical power threshold for cavitation faults of the water pump unit under the current operating conditions and the probability of cavitation faults of the water pump unit. S5. The cavitation critical power threshold and cavitation failure rate data output by the model are verified, outliers are removed, and standardized and desensitized to obtain the cavitation prediction results.

7. A remote monitoring system for a water intake pumping vessel according to claim 6, characterized in that, In S2, the hyperparameters for iterative optimization of the particle swarm optimization algorithm include particle swarm size, upper and lower limits of particle flight speed, particle position value range, individual learning factor, global learning factor, and inertia weight coefficient.

8. A remote monitoring system for a water intake pumping vessel according to claim 7, characterized in that, In step S3, the calibration of the prediction model after the verification set is reconstructed includes: inputting multidimensional environmental parameter data from the verification set into the prediction model, obtaining the cavitation prediction results output by the model, comparing the prediction results with the corresponding actual cavitation fault data in the verification set, and correcting the parameters of the model based on the comparison results.

9. A remote monitoring system for a water intake pumping vessel according to claim 8, characterized in that, The aforementioned strategy for adjusting the operating power of the water pump unit includes: When the water intake port is slightly blocked and the cavitation failure rate is lower than the preset safety threshold, the unit's operating power is increased. When the water intake port is moderately blocked and the cavitation failure rate is in the warning range, the unit's operating power is reduced in stages. When the water intake port is severely blocked, the cavitation failure rate is higher than the danger threshold, reducing the unit's operating power.

10. The monitoring method of the remote monitoring system for a water intake pumping vessel according to any one of claims 1-9, characterized in that, Includes the following steps: Step 1: Collect water quality parameters, water pressure parameters, water temperature parameters, water impurity parameters, and water pump unit operating power parameters in real time around the water intake port of the water intake pumping vessel through the data acquisition unit; Step 2: The collected data is filtered, calibrated, aligned and cleaned preprocessed. Based on the preprocessed data, the blockage status of the water intake port of the water intake pumping vessel at different times is analyzed. Step 3: Construct a time-series prediction model using a particle swarm optimization long short-term memory network algorithm. Utilize model iterative training and parameter calibration to fit the dynamic coupling relationship between the water intake port blockage status, water environment parameters, and pump unit operating power, and determine the current cavitation critical power threshold and cavitation failure rate of the pump unit. Step 4: Receive the blockage status analysis results and cavitation prediction results, and generate the corresponding water pump unit power regulation strategy; Step 5: Through the system management unit, complete the monitoring of the entire system's operating status, fault self-check, access control, and full-process data archiving and storage to realize remote monitoring of the water intake pumping vessel.

Images