A method for constructing a hydraulic simulation model of a gate hoist

Abstract

This invention provides a method for constructing a hydraulic simulation model of a gate hoist, belonging to the field of hydraulic system simulation technology. This method first plots the load-time variation curve of the gate hoist through theoretical calculations and actual working condition surveys, and designs a proportional relief valve control strategy. Then, a hydraulic system test platform is built and debugged, and multi-dimensional experimental data is collected. After data quality assessment, outlier handling, and dataset partitioning, the simulation model's parameters are optimized and trained using the training set. Finally, the model undergoes multi-index testing and characteristic matching analysis, and closed-loop iterative optimization is performed until the model meets performance requirements. This invention achieves high-precision simulation of nonlinear dynamic loads, provides rich data sources for model training through multi-dimensional experimental data collection, and the constructed simulation model exhibits excellent accuracy, generalization, and robustness. It can also be extended to fault condition simulation, providing technical support for the design, debugging, and fault handling of gate hoist hydraulic systems.

Description

Technical Field

[0001] This invention relates to the field of hydraulic system simulation technology, specifically to a method for constructing a hydraulic simulation model of a gate hoist. Background Technology

[0002] The dynamic load characteristics of the hydraulic system of a gate hoist are complex, with nonlinear operating conditions such as transient impacts during start-up and braking, and fluctuations in load. The accuracy of its simulation model directly affects the design, commissioning, operation, and maintenance of the hydraulic system. Existing dynamic load simulation methods for hydraulic system simulation models often rely on engineering experience to select 2-3 steady-state load points from the entire life cycle operation scenario, ignoring the complex dynamic operating conditions. They construct a static loading loop through relief valves and throttle valves, collecting only macroscopic parameters such as pressure and flow rate, and then using algorithms such as linear interpolation and least squares fitting to generate a "dynamic load curve" as the load input source for the simulation model.

[0003] The existing technology has obvious drawbacks: the experimental data is one-sided and cannot describe the complex behavior of dynamic systems; it performs simple linear fitting on nonlinear dynamic loads and ignores the action characteristics of hydraulic components; the vibration and pressure shock generated by the elasticity of components such as oil, pipelines, piston pump cylinder and distribution plate under nonlinear dynamic loads cannot be reflected by a single static load simulation, resulting in a low degree of matching between the constructed simulation model and the actual hydraulic system, and the simulation results have limited reference value. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for constructing a hydraulic simulation model of a gate hoist, which solves the problems of traditional static load simulation failing to reflect the complex dynamic behavior of hydraulic systems, having limited experimental data dimensions, and low simulation model accuracy. This method achieves high-precision simulation of nonlinear dynamic loads and constructs a hydraulic simulation model with excellent accuracy, generalization, and robustness.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for constructing a hydraulic simulation model of a gate hoist includes the following steps: S1. Through theoretical calculations and actual working condition surveys, plot the curves of load change with time under different working conditions of the gate hoist, analyze the inherent characteristics of the hydraulic system, and design the control strategy of the proportional relief valve so that the proportional relief valve can adjust the opening degree according to the real-time changes in load and the system operating status. S2. Build a test platform for the hydraulic system of the gate hoist, select and calibrate high-precision detection components, debug the hydraulic circuit, and ensure the stable operation of the test platform; S3. Design a test plan for normal operating conditions and conduct the test, while simultaneously collecting the conventional and characteristic parameters of the hydraulic system; S4. Evaluate the quality of the collected experimental data, use a statistical learning outlier detection algorithm to process outliers, and divide the processed data into training and test sets. S5. Optimize and train the parameters of the hydraulic simulation model based on the training set; S6. Perform multi-index tests on the trained model for accuracy, generalization and robustness. Perform static and dynamic characteristic matching analysis between the model output results and the actual hydraulic system test data. If the model does not meet the predetermined performance requirements, backtrack and optimize the data acquisition, data processing and model training links in sequence, re-execute the corresponding steps and retrain and test the model to form a closed loop iterative optimization until a hydraulic simulation model suitable for the gate opening and closing mechanism hydraulic system is obtained.

[0006] Preferably, in step S1, the inherent characteristics of the hydraulic system include the output characteristics of the hydraulic pump, the response characteristics of the hydraulic cylinder, and the hydraulic resistance characteristics of the pipeline.

[0007] Preferably, in step S2, the detection element includes a pressure sensor, a flow sensor, a displacement sensor, and a temperature sensor. The output signal of the detection element is calibrated using standard metrology equipment to ensure that the measurement accuracy of the detection element within the test measurement range meets the preset requirements. The hydraulic circuit debugging includes checking the installation sealing of the hydraulic components and testing the starting performance of the hydraulic pump and the smoothness of the hydraulic cylinder movement.

[0008] Preferably, in step S3, the conventional parameters include pressure, flow rate, and hydraulic oil temperature, and the characteristic parameters include piston pump vibration acceleration and piston pump outlet high-frequency pressure.

[0009] Preferably, in step S4, the data quality assessment includes data integrity and data accuracy verification, the outlier handling includes identifying and removing invalid outlier data, and the data processing also includes standardizing the data.

[0010] Preferably, in step S6, the accuracy test uses one or more error evaluation indicators such as root mean square error and mean absolute error to evaluate the degree of agreement between the model output and the actual experimental data; the generalization test introduces experimental data that has not participated in model training to verify the model's ability to adapt to unfamiliar data; the robustness test observes the changes in model performance by adding random noise or interference signals to the test data.

[0011] Preferably, in step S6, when the model fails to meet the predetermined performance requirements, the specific process of backtracking optimization is as follows: First, analyze whether there are problems such as incomplete coverage of experimental conditions or omission of data acquisition parameters during the data acquisition stage. If so, supplement the experiment and re-acquire the data. Next, check whether there are any improper handling of outliers in the data processing process. If so, optimize the data processing flow. Then, analyze the scientific nature of the parameter adjustment strategy and the division of the training set and test set during the model training process, and adjust the model training strategy accordingly.

[0012] Preferably, the test plan also includes a fault condition simulation and emergency strategy training step: a fault injection module is added to the test plan to simulate common sudden faults in the hydraulic system, system response data under fault conditions is collected, and emergency control strategies of the hydraulic system are trained based on the response data, so that the simulation model has the ability to simulate fault conditions and provide technical support for emergency handling.

[0013] Preferably, the common sudden failures include one or more of the following: distribution plate wear, pump cavitation, proportional relief valve jamming, and sensor failure; the system response data includes parameter data corresponding to sudden pressure drop and flow interruption.

[0014] Preferably, the emergency control strategy includes a strategy for temporarily controlling the hydraulic system based on simulation model predictions when sensors fail.

[0015] Beneficial effects of this invention: 1. This invention achieves high-precision reproduction of the nonlinear dynamic load of a gate hoist. By combining theoretical calculations with actual working condition surveys, a complete load-time variation curve is plotted to capture the dynamic characteristics of the load. Combined with a customized control strategy using a proportional relief valve, the system pressure is dynamically adjusted to ensure that the pressure output is highly matched with the load demand. This effectively reproduces the elastic vibration and pressure impact effects of components such as oil, pipelines, and piston pump cylinders under nonlinear loads, solving the core defects of traditional static load simulation.

[0016] 2. This invention constructs a multi-dimensional data acquisition system for conventional parameters and characteristic parameters, breaking through the limitations of traditional experiments that only collect conventional parameters such as pressure and flow rate. The newly added characteristic parameters such as piston pump vibration acceleration and high-frequency pressure can uncover the implicit dynamic characteristics of the hydraulic system, providing richer and more realistic data sources for simulation model training, and avoiding insufficient simulation accuracy due to missing data dimensions.

[0017] 3. This invention adopts a closed-loop iterative optimization approach, forming a complete optimization closed loop from data acquisition and data processing to model training and testing. Through multi-index testing of accuracy, generalization and robustness and characteristic matching analysis, the model is fully verified. If the requirements are not met, each link is backtracked to optimize, ensuring that the final simulation model is highly adapted to the actual hydraulic system and that the model has excellent accuracy, generalization and robustness.

[0018] 4. This invention can be extended to fault condition simulation and emergency strategy training. By adding a fault injection module to collect fault response data and train emergency control strategies, the simulation model can not only be used for simulation analysis of normal working conditions, but also provide technical support for emergency handling of faults in the hydraulic system of gate hoists, thereby improving the safety and reliability of equipment operation. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0022] Example 1: See Figure 1 A method for constructing a hydraulic simulation model of a gate hoisting mechanism is proposed, which establishes a closed-loop technical route of "theoretical modeling - experimental verification - data-driven - iterative optimization", specifically including the following steps: S1. Load characteristic analysis and proportional relief valve control strategy design: First, a comprehensive and in-depth analysis of the load on the gate hoist is conducted. Through theoretical calculations and on-site investigations of the actual working conditions of the gate hoist, curves showing the load change over time under different working conditions are accurately plotted to fully capture the dynamic characteristics of the load. Simultaneously, considering the structural composition and working principle of the gate hoist, the inherent characteristics of the hydraulic system are analyzed. These inherent characteristics include the output characteristics of the hydraulic pump, the response characteristics of the hydraulic cylinder, and the hydraulic resistance characteristics of the pipeline. Based on the dynamic characteristics of the load and the inherent characteristics of the hydraulic system, a control strategy for the proportional relief valve is designed. This allows the proportional relief valve to precisely adjust its opening according to real-time changes in the load and the operating status of the system, achieving precise control of the hydraulic system pressure and ensuring a high degree of matching between the pressure output and load requirements.

[0023] S2. Construction and debugging of the hydraulic system test platform for gate hoist: A test platform for the hydraulic system of the gate hoist was constructed. During the platform construction process, high-precision detection elements were selected, including pressure sensors, flow sensors, displacement sensors, and temperature sensors. Standard metrology equipment was used to rigorously calibrate the output signals of each detection element to ensure that the measurement accuracy of the detection elements meets the preset requirements throughout the entire test measurement range. The hydraulic circuit of the test platform was meticulously debugged, and the installation sealing of each hydraulic component was checked. The starting performance of the hydraulic pump and the smoothness of the hydraulic cylinder movement were tested to ensure that the hydraulic circuit of the test platform operates stably and reliably, providing a high-precision and stable testing environment for subsequent tests.

[0024] S3. Multi-dimensional experimental data acquisition: A test plan was designed for the gate hoist under normal operating conditions, and the test operations were carried out in strict accordance with the test plan. During the test, the conventional parameters and characteristic parameters of the hydraulic system were collected synchronously using the constructed test platform. The conventional parameters included pressure, flow rate, and hydraulic oil temperature, while the characteristic parameters included piston pump vibration acceleration and high-frequency pressure at the piston pump outlet. This enabled multi-dimensional and multi-parameter data collection of the hydraulic system, providing a rich and accurate data source for subsequent simulation model training.

[0025] S4. Experimental Data Processing and Dataset Division: A comprehensive quality assessment is conducted on the collected multi-dimensional experimental data, including verification of data integrity and accuracy. An outlier detection algorithm based on statistical learning is used to identify and process outliers in the data, eliminating invalid outliers to ensure data accuracy. The processed data is then standardized. The outlier-processed and standardized experimental data is scientifically divided into training and testing sets to provide a standardized and accurate data foundation for simulation model training and testing.

[0026] S5. Parameter optimization and training of hydraulic simulation model based on training set: Based on the predefined training set data, the parameters of the hydraulic simulation model are adjusted and optimized in a targeted manner. The training of the hydraulic simulation model is completed through a multi-strategy parameter optimization method, so that the model initially has the ability to simulate the operating state of the hydraulic system of the gate hoist.

[0027] S6. Multi-index testing and closed-loop iterative optimization of the model: The trained hydraulic simulation model was tested for accuracy, generalization, and robustness. Accuracy testing used error evaluation indicators such as root mean square error and mean absolute error to comprehensively evaluate the degree of agreement between the model output and the actual experimental data. Generalization testing introduced experimental data that were not used in model training to verify the model's adaptability to unfamiliar data. Robustness testing added random noise or interference signals of a certain intensity to the test data and observed the changes in model performance to verify the model's stability in the presence of interference.

[0028] After completing multi-index testing, the output results of the simulation model are matched and analyzed with the test data of the actual hydraulic system from multiple dimensions, including static and dynamic characteristics, to verify the simulation model's ability to accurately simulate the characteristics of the gate hoist hydraulic system. If the simulation model does not meet the predetermined performance requirements, the following backtracking optimization process is followed: First, analyze whether there are problems such as incomplete coverage of test conditions or missing data acquisition parameters during the data acquisition stage. If so, supplement the corresponding tests and re-acquire data. Next, check whether there are problems such as improper handling of outliers or non-standardized processing during the data processing. If so, optimize the data processing process. Then, for the model training process, analyze whether the parameter adjustment strategy is appropriate and whether the division of the training set and test set is scientific, and adjust the model training strategy accordingly. After completing the above optimization, the model training, multi-index testing, and characteristic matching analysis steps are executed again to form a closed-loop iterative optimization process until a high-precision hydraulic simulation model that can accurately adapt to the gate hoist hydraulic system is obtained.

[0029] Example 2: This embodiment constructs a hydraulic simulation model for the start-up and braking conditions of a gate hoist. These conditions involve significant transient impact loads, requiring accurate capture of load abrupt changes. The specific steps are as follows: S1. Load characteristic analysis and proportional relief valve control strategy design: For gate start-up and braking conditions (gate start-up, gate start-up, gate stop-up, and gate stop-up), the magnitude and variation law of transient impact load are calculated using dynamic theory, and the sudden change curve of load over time (load rises / falls suddenly within 0.2s) is plotted in combination with field measured data; the transient response characteristics of the hydraulic system are analyzed, including the start-up and braking output characteristics of the hydraulic pump and the impact response characteristics of the hydraulic cylinder; a fuzzy control strategy for the proportional relief valve is designed, which adjusts the valve opening rapidly in stages according to the amplitude and rate of load change, suppresses transient pressure impact, and achieves precise matching between pressure output and sudden load change.

[0030] S2. Construction and debugging of the hydraulic system test platform for gate hoist: Based on the test platform of Example 1, a high-frequency dynamic pressure sensor (sampling frequency 2kHz) was added. All sensors were recalibrated using standard metrology equipment to ensure that the dynamic parameter acquisition accuracy met the preset requirements. The buffer device of the hydraulic circuit was debugged, the buffer performance of the hydraulic cylinder during start-up and braking was tested, the sealing of each hydraulic component was checked, and the start-up and braking response performance of the hydraulic pump was tested to avoid the impact of the test platform itself on data acquisition and to ensure the stable and reliable operation of the hydraulic circuit of the test platform.

[0031] S3. Multi-dimensional experimental data acquisition: A test scheme for start-up and braking conditions was designed, with an acceleration of 0.1 m / s² and a deceleration of 0.08 m / s² for door opening, and an acceleration of 0.06 m / s² and a deceleration of 0.05 m / s² for door closing. Five repeated tests were conducted. During the test, conventional and characteristic parameters of the hydraulic system were collected simultaneously. The sampling interval for conventional parameters was 0.01 s, and the sampling interval for characteristic parameters (vibration acceleration of the plunger pump and high-frequency pressure at the plunger pump outlet) was 0.0005 s (2 kHz). This accurately captured transient vibration and pressure impact characteristics, providing complete dynamic operating condition data for model training.

[0032] S4. Experimental Data Processing and Dataset Division: A comprehensive quality assessment was conducted on the five sets of experimental data to verify the data's completeness and accuracy. The Isolation Forest algorithm (a statistical learning outlier detection algorithm) was used to identify outliers in the data caused by instantaneous sensor jitter, and invalid outliers were removed. The processed data was standardized, and to improve the model's adaptability to sudden load characteristics, the data was scientifically divided into training and test sets in an 8:2 ratio, increasing the sample size of the training set to cover more transient operating conditions.

[0033] S5. Parameter optimization and training of hydraulic simulation model based on training set: The hydraulic simulation model described in this invention is a mechanism model built based on AMESim. Based on the pre-divided training set data, the dynamic parameters of the model are optimized in a targeted manner, adjusting key physical parameters such as the elastic modulus of the pipeline, the bulk elastic modulus of the hydraulic oil, and the response time of the buffer valve. A genetic algorithm is used to complete the parameter identification and training of the model. Through iterative optimization, the parameter combination that matches the actual working conditions is determined, so that the model initially has the ability to simulate the transient impact conditions of starting and braking.

[0034] S6. Multi-index testing and closed-loop iterative optimization of the model: The trained AMESim mechanism model was tested for accuracy, generalization, and robustness. Specifically, the accuracy test used normalized root mean square error (NRMSE) and mean absolute percentage error (MAPE) as evaluation metrics, with NRMSE ≤ 5% and MAPE ≤ 4% for the model output compared to the actual experimental data. The generalization test incorporated one set of start-stop test data that was not used in the training, and the model's fit to the unfamiliar data was ≥ 97%. The robustness test added 8% random noise to the test data, and the model's performance degradation was ≤ 3%.

[0035] After completing multi-index testing, the simulation model output results were matched and analyzed with the actual hydraulic system test data from multiple dimensions, including static and dynamic characteristics. It was found that there was a slight deviation between the simulated pressure value and the actual value at the moment of braking. Subsequently, the model training stage was traced back, and the number of iterations of the genetic algorithm was increased from 100 to 200 to optimize the parameter search range. The model was re-identified and trained. After completing multi-index testing and characteristic matching analysis again, the model deviation was eliminated, and the predetermined performance requirements were met. A high-precision simulation model adapted to the hydraulic system of the gate hoist under starting and braking conditions was obtained.

[0036] Example 3: This embodiment constructs a hydraulic simulation model for the variable load fluctuation condition of the gate hoist. In this condition, the load fluctuates periodically due to changes in the impact force of water flow on the gate. It is necessary to ensure the model's ability to follow the periodic load. The specific steps are as follows: S1. Load characteristic analysis and proportional relief valve control strategy design: For gate operating conditions with variable load fluctuations (water flow velocity 0.5-2 m / s, load fluctuation period 5s, fluctuation amplitude ±10%), the impact force of water flow at different flow velocities is calculated using fluid dynamics theory, and the periodic fluctuation curve of load over time is plotted in conjunction with the field conditions; the steady-state response characteristics of the hydraulic system under variable loads are analyzed; a composite control strategy (PID + feedforward control) for the proportional relief valve is designed to adjust the valve opening in advance based on the predicted value of load fluctuations, improve the system pressure's ability to follow the periodic load, and ensure that the pressure output matches the fluctuating load in real time.

[0037] S2. Construction and debugging of the hydraulic system test platform for gate hoist: Based on the test platform of Example 1, a load simulation device was added to simulate the periodic load fluctuations caused by the impact force of water flow; all sensors were dynamically calibrated using standard measuring equipment to ensure the accuracy of parameter acquisition during load fluctuations; the flow regulation performance of the hydraulic pump was adjusted, the response capability of the hydraulic pump to adjust the output flow in real time according to load fluctuations was tested, and the sealing performance of each pipe joint and component in the hydraulic circuit was checked to ensure the stable operation of the test platform.

[0038] S3. Multi-dimensional experimental data acquisition: Design a test scheme for variable load fluctuation conditions, setting three different load fluctuation amplitudes (±5%, ±10%, ±15%), and conducting four repeated tests for each condition. During the test, conventional parameters and characteristic parameters of the hydraulic system are collected simultaneously. The sampling interval for conventional parameters is 0.01s, and the sampling interval for characteristic parameters (piston pump vibration acceleration and piston pump outlet high-frequency pressure) is 0.001s (1kHz). The focus is on collecting the fluctuation data of piston pump outlet pressure and the periodic change data of vibration acceleration to fully capture the system operation characteristics under variable load conditions.

[0039] S4. Experimental Data Processing and Dataset Division: A comprehensive quality assessment was conducted on 12 sets of experimental data to verify data completeness, accuracy, and absence of missing values. The Local Outlier Factor (LOF) algorithm was used to detect outliers in the data, and three invalid data points were removed. The processed data was standardized and scientifically divided into training and test sets in a 7:3 ratio. The test set contained experimental data with three different fluctuation amplitudes to ensure that the test set could fully verify the model's adaptability to different load fluctuation conditions.

[0040] S5. Parameter optimization and training of hydraulic simulation model based on training set: The hydraulic simulation model described in this invention is a mechanism model built based on AMESim. Based on the pre-divided training set data, the key parameters of the model are optimized, and physical parameters such as the response coefficient of the proportional relief valve, the gain coefficient of the hydraulic circuit, and the volumetric efficiency of the hydraulic pump are adjusted. The particle swarm optimization algorithm is used to complete the parameter identification and training of the model. Taking advantage of the fast convergence speed of the algorithm, the parameter combination that matches the variable load fluctuation condition is quickly optimized, so that the model initially has the ability to simulate periodic load fluctuation conditions.

[0041] S6. Multi-index testing and closed-loop iterative optimization of the model: The trained AMESim mechanism model was tested for accuracy, generalization, and robustness. For accuracy, the normalized root mean square error (NRMSE) and mean absolute percentage error (MAPE) were used as evaluation metrics. The NRMSE of the model output was ≤4% and the MAPE was ≤2% compared with the actual experimental data. In the generalization test, the model's fit for three different load fluctuation conditions was ≥98.5%. In the robustness test, 10% random noise was added to the test data, and the model performance degradation was ≤1.5%.

[0042] After completing multi-index tests, the simulation model output results were matched and analyzed with the actual hydraulic system test data from multiple dimensions of static and dynamic characteristics. The verification results showed that the model can accurately simulate the system pressure and flow change patterns under load fluctuations, and all performance indicators meet the predetermined requirements, thus obtaining a high-precision simulation model of the gate hoist hydraulic system adapted to variable load fluctuation conditions.

[0043] Example 4: This embodiment takes sensor failure as an example to construct a simulation model of the hydraulic system of a gate hoist under fault conditions, and completes emergency control strategy training, so that the simulation model has the ability to simulate fault conditions and provide emergency handling technical support. The specific steps are as follows: S1. Basic Construction of Normal Operating Condition Simulation Model: First, following the method in Example 1, a mechanism simulation model based on AMESim was constructed for the gate hoist under normal operating conditions. Through multi-index testing and characteristic matching analysis, the normalized root mean square error (NRMSE) of the model under normal operating conditions was ensured to be ≤5%, the fit was ≥98%, and the simulation accuracy met the preset performance requirements, which served as the basis for constructing the fault condition simulation model.

[0044] S2. The test plan fault injection module has been added as follows: Based on the normal operating condition test plan, a fault injection module is added to simulate the failure of the pressure sensor through electrical control. Specifically, it includes two typical failure states: interruption of sensor output signal and fixed value of sensor output. The timing and duration of fault triggering parameters are defined, and a standardized fault operating condition test plan is formulated.

[0045] S3. Multi-dimensional data collection of fault conditions: Design a test scheme for sensor failure failure conditions. Trigger the pressure sensor failure during the normal opening and closing of the gate and conduct 6 repeated tests. During the test, system response data under the failure state are collected synchronously, including conventional parameters such as sudden change in flow, no pressure feedback, and abnormal movement speed of hydraulic cylinder, as well as abnormal change characteristic parameters such as piston pump vibration acceleration and high-frequency pressure. The sampling interval for conventional parameters is 0.01s, and the sampling interval for characteristic parameters is 0.001s (1kHz), so as to fully capture the changes in the operating state of the hydraulic system after sensor failure.

[0046] S4. Construction of Fault Condition Simulation Model and Training of Emergency Control Strategies: Based on the collected fault response data, secondary parameter identification and optimization are performed on the constructed AMESim mechanism model under normal operating conditions. The parameters related to sensor signal feedback in the model are adjusted to construct a hydraulic simulation model adapted to sensor failure fault conditions. At the same time, the emergency control strategy of the hydraulic system is trained based on the fault response data. Specifically, when the model detects abnormal states such as interruption of pressure sensor output signal or fixed output value, the proportional relief valve is temporarily controlled based on the model prediction values ​​of other effective parameters such as flow rate and displacement to maintain the basic operation of the hydraulic system and ensure the safety of gate opening and closing.

[0047] S5. Fault Condition Model Testing and Closed-Loop Iterative Optimization: Multi-index tests were conducted on the sensor failure fault condition simulation model to verify the model's simulation accuracy of the fault condition and the effectiveness of the emergency control strategy. The tests revealed that the model's predicted pressure value deviated significantly from the actual value at the initial stage of the fault. Therefore, the data acquisition stage was revisited, and three additional special tests were conducted at the moment of fault triggering. Data was re-acquired, and the fault response data was processed for outliers and standardized. The fault condition model was then re-identified and optimized for secondary parameters, and the emergency control strategy was iteratively trained.

[0048] After the test was completed again, the fault condition simulation model can accurately simulate the system response under sensor failure. The normalized root mean square error (NRMSE) of the model for fault conditions is ≤6%. Moreover, the emergency control strategy can be effectively triggered, enabling the gate to safely complete the opening and closing action at a low speed of 0.2m / min, which meets the technical requirements for emergency handling of the hydraulic system failure of the gate hoist.

[0049] Although the preferred embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention 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 specific modifications under the guidance of the present invention without departing from the spirit of the invention and the scope of protection of the claims, and these modifications all fall within the scope of protection of the present invention.

Claims

1. A method for constructing a hydraulic simulation model of a gate hoist, characterized in that, Includes the following steps: S1. Through theoretical calculations and actual working condition surveys, plot the curves of load change with time under different working conditions of the gate hoist, analyze the inherent characteristics of the hydraulic system, and design the control strategy of the proportional relief valve so that the proportional relief valve can adjust the opening degree according to the real-time changes in load and the system operating status. S2. Build a test platform for the hydraulic system of the gate hoist, select and calibrate high-precision detection components, debug the hydraulic circuit, and ensure the stable operation of the test platform; S3. Design a test plan for normal operating conditions and conduct the test, while simultaneously collecting the conventional and characteristic parameters of the hydraulic system; S4. Evaluate the quality of the collected experimental data, use a statistical learning outlier detection algorithm to process outliers, and divide the processed data into training and test sets. S5. Optimize and train the parameters of the hydraulic simulation model based on the training set; S6. Perform multi-index tests on the trained model for accuracy, generalization and robustness. Perform static and dynamic characteristic matching analysis between the model output results and the actual hydraulic system test data. If the model does not meet the predetermined performance requirements, backtrack and optimize the data acquisition, data processing and model training links in sequence, re-execute the corresponding steps and retrain and test the model to form a closed loop iterative optimization until a hydraulic simulation model suitable for the gate opening and closing mechanism hydraulic system is obtained.

2. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S1, the inherent characteristics of the hydraulic system include the output characteristics of the hydraulic pump, the response characteristics of the hydraulic cylinder, and the hydraulic resistance characteristics of the pipeline.

3. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S2, the detection element includes a pressure sensor, a flow sensor, a displacement sensor, and a temperature sensor. The output signal of the detection element is calibrated using standard metrology equipment to ensure that the measurement accuracy of the detection element within the test measurement range meets the preset requirements. The hydraulic circuit debugging includes checking the installation and sealing of hydraulic components, testing the starting performance of the hydraulic pump, and the smoothness of the hydraulic cylinder movement.

4. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S3, the conventional parameters include pressure, flow rate, and hydraulic oil temperature, and the characteristic parameters include piston pump vibration acceleration and piston pump outlet high-frequency pressure.

5. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S4, the data quality assessment includes data integrity and data accuracy verification, the outlier handling includes identifying and removing invalid outlier data, and the data processing also includes standardizing the data.

6. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S6, the accuracy test uses one or more error evaluation indicators, such as root mean square error and mean absolute error, to evaluate the degree of agreement between the model output and the actual experimental data; the generalization test introduces experimental data that was not used in model training to verify the model's ability to adapt to unfamiliar data; the robustness test observes the changes in model performance by adding random noise or interference signals to the test data.

7. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 1, characterized in that, In step S6, when the model fails to meet the predetermined performance requirements, the specific process of backtracking optimization is as follows: First, analyze whether there are problems such as incomplete coverage of test conditions or omission of data acquisition parameters during the data acquisition stage. If so, supplement the test and re-acquire the data. Next, check whether there are any problems with improper handling of outliers in the data processing process. If so, optimize the data processing flow. Then, analyze the parameter adjustment strategy and the scientific nature of the division of training and test sets during model training, and adjust the model training strategy accordingly.

8. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to any one of claims 1-7, characterized in that, It also includes fault condition simulation and emergency strategy training steps: a fault injection module is added to the test scheme to simulate common sudden faults in the hydraulic system, collect system response data under fault conditions, and train the emergency control strategy of the hydraulic system based on the response data, so that the simulation model has the ability to simulate fault conditions and provide technical support for emergency handling.

9. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 8, characterized in that, The common sudden failures include one or more of the following: distribution plate wear, pump cavitation, proportional relief valve jamming, and sensor failure. The system response data includes parameter data corresponding to sudden pressure drop and flow interruption.

10. The method for constructing a hydraulic simulation model of a gate hoisting mechanism according to claim 8, characterized in that, The emergency control strategy includes a strategy for temporarily controlling the hydraulic system based on simulation model predictions when sensors fail.

Images