Method and system for testing durability of energy accumulator

Abstract

The invention relates to an energy accumulator durability test method and system, and the system comprises an extreme working condition simulation module which can achieve high-voltage, wide-temperature-range and high-frequency dynamic loading; the multi-sensor monitoring network integrates a fiber bragg grating strain sensor, a three-axis acceleration sensor and an acoustic emission sensor to realize collaborative monitoring of strain, vibration and microcracks; the intelligent control and energy recovery module is combined with a fuzzy PID and reinforcement learning algorithm, a magnetorheological fluid generator and a super capacitor bank, so that the control precision and the energy utilization rate are improved; and the data analysis and life evaluation platform accurately predicts the residual life by using an improved Bayesian network model and a real-time health index calculation module. According to the test method, a loading mode is dynamically adjusted through a real-time health index, an energy recovery strategy is optimized, and life early warning is triggered. The invention breaks through the limitation of the traditional test and provides an integrated solution.

Description

Technical Field

[0001] The present invention relates to the technical field of hydraulic component durability testing, and particularly relates to a durability testing method and system for an accumulator. Background Art

[0002] An accumulator is a core component in a hydraulic system for storing and releasing hydraulic energy. Through the energy conversion between compressed gas and hydraulic oil, functions such as system pressure stability, energy recovery, and shock buffering are achieved, and it is widely used in fields with extremely high reliability requirements such as aerospace, deep-sea equipment, and high-end construction machinery. Its durability (i.e., the life under cyclic loading) directly determines the operation safety and maintenance cost of the hydraulic system. If the accumulator fails prematurely under extreme working conditions, it may lead to the shutdown of the entire system and even cause safety accidents. Therefore, accurately simulating extreme working conditions and evaluating its durability are key technical requirements in the research, production, and maintenance of accumulators.

[0003] Currently, the durability testing of accumulators mainly relies on the following technical solutions:

[0004] (1) Simulating pressure, temperature, or dynamic load respectively through an independent high-pressure pump station, temperature control box, or mechanical actuator. For example, an electro-hydraulic servo system is used to apply static pressure, or temperature control is achieved through a resistance heater. Although such technologies can partially simulate a single environment, they cannot achieve the coupled loading of multiple physical fields;

[0005] (2) Using strain gauges or uniaxial acceleration sensors to monitor local deformation or vibration, or detecting cracks through regular disassembly. Such methods have problems of single monitoring dimension and poor data real-time performance, and it is difficult to capture microscopic material damage;

[0006] (3) In the testing system, hydraulic energy is mostly dissipated in the form of heat. Some technologies attempt to recover energy through a traditional generator, but due to the low conversion efficiency and the dynamic response ability of the energy storage unit, efficient closed-loop utilization of energy cannot be achieved;

[0007] (4) Estimating the life based on empirical formulas or single parameters, ignoring the coupled effect of multiple parameters, resulting in a large prediction error.

[0008] The above existing technologies have the following key problems: The traditional testing system cannot synchronously reproduce the composite extreme environment of high pressure, wide temperature range, and high-frequency dynamic loading, resulting in a serious disconnection between the test conditions and the actual working conditions; The discrete sensor network is difficult to comprehensively capture the multi-dimensional state information of the accumulator, affecting the accuracy of early damage identification; The traditional generator has a low energy recovery rate and lacks a dynamic matching mechanism with the system power supply demand, and cannot support the energy consumption self-sustainability of the testing system; The life assessment method based on single parameters or static weights does not consider the dynamic correlation of multi-source monitoring data, resulting in a remaining life prediction error exceeding the actual engineering tolerance. Summary of the Invention

[0009] Objective of the Invention: In order to overcome the deficiencies of the prior art, the present invention provides a durability test method and system for an accumulator. By constructing a composite working condition simulation module for high-pressure-wide temperature range-high-frequency dynamic loading, integrating a multi-dimensional sensor network to achieve collaborative monitoring of strain, vibration and micro-cracks, combining a magnetorheological fluid energy recovery device with fuzzy reinforcement learning hybrid control to improve energy utilization efficiency, and dynamically predicting the remaining life based on an improved Bayesian network model by fusing multi-source data, so as to solve the technical problems of distorted simulation of extreme working conditions, single monitoring dimension, low energy recovery efficiency and large life assessment error.

[0010] An accumulator durability test system, comprising: an extreme working condition simulation module, which is used to provide a high-pressure, wide temperature range and high-frequency dynamic loading environment. The pressure range of the high-pressure environment is 0-75 MPa, covering industrial-grade high-pressure working conditions. The temperature range of the wide temperature range environment is -55°C - 150°C, simulating high and low temperature extreme scenarios. The high-frequency dynamic loading frequency range is 0-20 Hz, reproducing high-frequency vibration or impact working conditions; a multi-sensor monitoring network, which includes a fiber Bragg grating strain sensor arranged along the axis of the accumulator to capture minute deformations, a three-axis acceleration sensor for capturing vibration signals, and an acoustic emission sensor for analyzing vibration modes and energy distribution and locating material micro-cracks to achieve early damage warning. The resolution of the fiber Bragg grating strain sensor is ≤±2 με, the sampling frequency of the three-axis acceleration sensor is ≥10 kHz, and the sensitivity of the acoustic emission sensor is ≥60 dB (reference standard: 1 pascal@1 kHz); an intelligent control and energy recovery module, which includes a hybrid controller based on fuzzy PID algorithm and reinforcement learning, and an energy recovery device composed of a magnetorheological fluid generator and a supercapacitor bank. The response time of the hybrid controller is ≤50 ms, and the energy recovery rate of the energy recovery device is ≥75%; a data analysis and life assessment platform, which is configured with an improved Bayesian network model and a real-time health index calculation module. The improved Bayesian network model fuses multi-dimensional monitoring data to predict the remaining life, and the prediction error is ≤5%. The real-time health index is obtained by dynamically adjusting the weights through the analytic hierarchy process; an edge computing node, which is used to perform real-time filtering, noise reduction and feature extraction on multi-sensor data. The preprocessing delay is ≤10 ms, and the preprocessed feature parameters include strain gradient, vibration main frequency, and acoustic emission event count rate.

[0011] An accumulator durability test system according to the present invention, wherein the extreme condition simulation module includes: a two-stage supercharging cylinder for achieving a high-pressure output of 0-75 MPa, covering the extreme pressure test requirements of the accumulator. The supercharging ratio of the two-stage supercharging cylinder under steady-state conditions is 10:1-20:1. Through a two-stage supercharging design (low-pressure stage + high-pressure stage), the input power requirement is significantly reduced while ensuring high-pressure stability. A double-piston structure is used to optimize the pressure conversion efficiency, and a servo motor closed-loop control is used to ensure the linearity and repeatability of the high-pressure output; a temperature-controlled oil source, including a high-temperature electric heater, a liquid nitrogen cooler, and a temperature sensor. The power of the high-temperature electric heater is ≥50 kW, and the cooling rate of the liquid nitrogen cooler is ≥10 °C / min; a composite loading device composed of a magnetorheological fluid damper and a piezoelectric ceramic actuator in series. The damping force adjustment range of the magnetorheological fluid damper is 0-50 kN, the displacement resolution of the piezoelectric ceramic actuator is ≤1 μm. The magnetorheological fluid damper adopts a multi-stage excitation coil design to improve the damping force response speed. The piezoelectric ceramic actuator is based on the inverse piezoelectric effect and is combined with a nanometer-level displacement sensor to achieve sub-micron-level positioning.

[0012] An accumulator durability test system according to the present invention, in the multi-sensor monitoring network: 4-8 groups of fiber Bragg grating strain sensors are evenly distributed along the circumferential direction of the accumulator, covering the circumferential direction of the accumulator (such as 0°, 90°, 180°, 270°) to capture strain differences in different orientations (such as uneven circumferential expansion under high pressure). Each group contains 2-4 axial measurement points arranged along the axis (such as the top, middle, and bottom) to monitor the axial stress gradient (such as stress concentration at the support connection); the triaxial acceleration sensor is installed at the connection between the accumulator support and the pipeline to monitor the X, Y, and Z three-axis vibration acceleration signals; the acoustic emission sensor is attached to the middle of the accumulator shell through a coupling agent, and the sampling frequency is ≥1 MHz.

[0013] An accumulator durability test system according to the present invention, in the intelligent control and energy recovery module: the fuzzy PID algorithm of the hybrid controller uses a Gaussian membership function, and the reinforcement learning algorithm is a deep Q network. The loading pressure fluctuation amplitude, temperature cycle rate, and loading frequency are dynamically adjusted through the real-time health index; the equivalent series resistance of the supercapacitor bank of the energy recovery device is ≤10 mΩ. The magnetorheological fluid generator and the hydraulic circuit are connected through a speed coupler to convert the mechanical energy generated by dynamic loading (such as the piston movement of the magnetorheological fluid damper and the vibration of the piezoelectric ceramic actuator) into electrical energy, and the power density is ≥500 W / L. The recovered energy is used to drive the temperature-controlled oil source and supply power to the sensors.

[0014] An accumulator durability test system according to the present invention. In the data analysis and life assessment platform: the improved Bayesian network model is based on Bayes' theorem. By considering the joint probability distribution of multiple monitoring parameters, the prediction of the remaining life state of the accumulator is realized, and its conditional probability calculation formula is:

[0015]

[0016] Where Ri (i = 1, 2,..., n) is the i-th monitoring parameter, n is the total number of monitoring parameters, including at least three of pressure, temperature, and strain; L is the remaining life state of the accumulator, and L′ represents all possible values of the remaining life state; P(L) is the prior probability, representing the estimation of the remaining life state before obtaining the monitoring data; P(R1|R2,..., Rn, L) is the likelihood function, describing the probability of observing the monitoring parameters R1, R2,..., Rn given the remaining life state L.

[0017] The real-time health index calculation formula is:

[0018]

[0019] Where the weight wi is dynamically adjusted according to the working condition type by the analytic hierarchy process, and Ri,min and Ri,max are the minimum and maximum values of the i-th monitoring parameter respectively. This formula quantifies the real-time health state of the accumulator by normalizing and weighted summing each monitoring parameter.

[0020] An accumulator durability test method based on the above system, including the following steps:

[0021] Initialization configuration: Set the target pressure, temperature range, and loading frequency, start the temperature control oil source, adjust the oil temperature to the target value through a high-temperature electric heater or a liquid nitrogen cooler, and use the temperature sensor for closed-loop feedback control; Dynamic loading and data acquisition: The hybrid controller adjusts the loading mode according to the real-time health index, and multiple sensors synchronously collect strain, vibration, and acoustic emission signals and transmit them to the edge computing node.

[0022] Energy recovery and optimization: In the unloading stage, the magnetorheological fluid energy recovery device converts hydraulic energy into electrical energy, and the supercapacitor bank stores the energy and feeds it back to the temperature control oil source.

[0023] Life assessment and decision-making: The improved Bayesian network model calculates the remaining life in real time, and triggers an alarm and adjusts the test strategy when the predicted life < 1000 cycles.

[0024] A test method according to the present invention, and the loading mode switching rule is:

[0025] When the real-time health index HI > 0.8, high-frequency impact loading is initiated, and the pressure fluctuation amplitude is ±10% - ±15% of the set target pressure; when the real-time health index HI < 0.3, it switches to static pressure holding loading, and the pressure deviation is controlled within ±1%. During the healthy period, high-frequency impact accelerates damage evolution and shortens the test cycle; during the degradation period, static pressure holding ensures that the failure process is controllable, facilitating the capture of critical state data for microcrack propagation.

[0026] In the energy recovery step of the testing method described in the present invention, the energy management algorithm dynamically adjusts the recovery power according to the state of charge (SOC) of the supercapacitor: when SOC < 30%, the magnetorheological fluid generator is used to increase the energy capture intensity to quickly replenish the capacitor charge and avoid energy recovery interruption due to insufficient power, and the recovery power is increased to 120% of the rated value; when SOC > 80%, the recovery power is reduced to 80% of the rated value to prevent overcharging of the supercapacitor.

[0027] From the above technical solutions, it can be seen that the present invention has the following beneficial effects:

[0028] The accumulator durability testing system described in the present invention constructs a multi-physical field coupling environment of high pressure, wide temperature range, and high-frequency dynamic loading through the extreme working condition simulation module. The multi-sensor monitoring network realizes high-precision synchronous acquisition of strain, vibration, and acoustic emission signals. The intelligent control and energy recovery module combines the fuzzy PID-reinforcement learning hybrid algorithm and the magnetorheological fluid power generation technology to improve the control accuracy and energy utilization rate. The data analysis platform realizes accurate prediction of the remaining life through the improved Bayesian network and the dynamic weight health index, breaking through the limitations of traditional testing in terms of environmental coverage, monitoring dimensions, intelligent level, and energy consumption control, and having the comprehensive advantages of test comprehensiveness, prediction accuracy, operation economy, and system safety, which can significantly improve the efficiency and depth of accumulator reliability verification.

[0029] The accumulator durability testing method described in the present invention realizes the adaptive optimization of the test process, the intelligent regulation of the energy flow, and the active prevention and control of failure risks through the dynamic switching rule of the loading mode based on the real-time health index, such as high-frequency impact in the healthy period to accelerate damage and static pressure holding in the degradation period to ensure safety, the closed-loop control of the energy management algorithm for the state of charge of the supercapacitor, and the life prediction and early warning mechanism driven by the Bayesian network model. And through multi-stage data acquisition, it ensures the integrity of the failure mechanism analysis, providing an integrated solution for accumulator durability testing. Description of the Drawings

[0030] Figure 1 It is an architecture diagram of an accumulator durability testing system of the present invention;

[0031] Figure 2 It is a flowchart of an accumulator durability testing method of the present invention. Detailed implementation mode

[0032] Example 1

[0033] I. System construction and parameter configuration

[0034] Refer to Figure 1 The accumulator durability test system is constructed according to the shown system architecture. The specific implementation of each module is as follows: 1. Extreme working condition simulation module

[0035] Double-stage supercharging cylinder: Its low-pressure stage supercharging ratio is 10:1, and the high-pressure stage supercharging ratio is 20:1. It realizes continuously adjustable high-pressure output of 0-75 MPa through servo motor drive, and the pressure fluctuation is ≤±1% under steady-state conditions.

[0036] Temperature-controlled oil source: Configure a high-temperature electric heater with a power of 60 kW and a liquid nitrogen cooler (cooling rate 15°C / min). The accuracy of the temperature sensor is ±0.5°C, realizing wide-temperature control from -55°C to 150°C, and maintaining the oil temperature stable through the PID closed-loop algorithm.

[0037] Composite loading device: Adopt a series structure of a magnetorheological fluid damper (continuously adjustable damping force of 0-50 kN) and a piezoelectric ceramic actuator (displacement resolution of 0.5 μm), and realize high-frequency dynamic loading of 0-20 Hz through a servo controller, with a displacement accuracy of ≤±1 μm.

[0038] 2. Multi-sensor monitoring network

[0039] Fiber Bragg grating strain sensor: Four groups are evenly distributed along the circumferential direction of the accumulator (at 0°, 90°, 180°, and 270° azimuths), and each group contains three axial measurement points (top, middle, and bottom). Fiber Bragg grating sensors with a resolution of ±1.5 με are selected, and distributed strain monitoring is realized through wavelength division multiplexing technology.

[0040] Three-axis acceleration sensor: Installed at the connection between the accumulator support and the pipeline, with a sampling frequency of 10 kHz, monitoring the vibration acceleration signals in the X, Y, and Z directions, and the sensitivity is ≥100 mV / g.

[0041] Acoustic emission sensor: The sensor is attached to the middle part of the accumulator shell through a coupling agent, with a sampling frequency of 1 MHz and a sensitivity of 65 dB, used to capture the acoustic emission signals generated by material microcracks.

[0042] 3. Intelligent control and energy recovery module

[0043] Hybrid Controller: Adopts a hybrid control algorithm of fuzzy PID and Deep Q-Network (DQN). The parameters of the Gaussian membership function of fuzzy PID are σ = 0.5 (control rule weight) and ε = 0.1 (defuzzification threshold); the DQN network contains 2 hidden layers (64 neurons in each layer), and dynamically adjusts the loading parameters (pressure fluctuation amplitude, temperature cycling rate, loading frequency) through the real-time Health Index (HI), with a response time ≤ 40 ms.

[0044] Energy Recovery Device: A magnetorheological fluid generator (power density 600 W / L) is connected to a supercapacitor bank (equivalent series resistance 8 mΩ, capacitance 500 F). The hydraulic circuit drives the generator through a speed coupler. The measured energy recovery rate is 78%, and the recovered energy is used for the temperature control oil source and sensor power supply.

[0045] 4. Data Analysis and Life Assessment Platform

[0046] Edge Computing Node: Adopts an embedded processor to perform real-time filtering (Butterworth filter, cut-off frequency 1 kHz), noise reduction (wavelet threshold method) and feature extraction on sensor data. The preprocessing delay is 8 ms, and parameters such as strain gradient, main vibration frequency (FFT analysis), and acoustic emission event count rate are output.

[0047] Improved Bayesian Network Model: The input parameters include 5 dimensions of pressure, temperature, strain, main vibration frequency, and acoustic emission count rate. The network is constructed through the PyMC3 library of Python, and the conditional probability table is trained based on historical test data. The remaining life prediction error is 4.2%.

[0048] Real-time Health Index Calculation: Determines the weights using the Analytic Hierarchy Process. When the working condition is "high-frequency impact", pressure (w1 = 0.3), strain (w2 = 0.25), and acoustic emission (w3 = 0.25) are the main weight factors. The calculation formula is:

[0049] II. Implementation Steps of the Test Method (refer to Figure 2 Process)

[0050] 1. Initialization Configuration

[0051] Set the target pressure to 50 MPa, the temperature range to -30°C to 120°C, and the loading frequency to 15 Hz.

[0052] Start the temperature control oil source. The high-temperature electric heater heats the oil to 120°C (takes 8 minutes), and then the liquid nitrogen cooler cools it to -30°C (cooling rate 12°C / min). The circulating temperature accuracy is maintained within ±2°C through closed-loop control by the temperature sensor.

[0053] 2. Dynamic Loading and Data Acquisition

[0054] The hybrid controller starts the high-frequency impact loading mode according to the initial health index (HI = 0.9, default value), with a pressure fluctuation amplitude of ±12% (i.e., 44 - 56 MPa), and the piezoelectric actuator outputs a sine wave vibration with an amplitude of 10 μm.

[0055] Multi-sensor synchronous data acquisition: The fiber Bragg grating strain sensor monitors the circumferential strain distribution (maximum strain value 1200 με), the triaxial acceleration sensor captures the main vibration frequency of 15 Hz (consistent with the loading frequency), and the acoustic emission sensor does not detect effective events (count rate = 0).

[0056] After being preprocessed by the edge computing node, the data is transmitted to the data analysis platform.

[0057] 3. Energy recovery and optimization

[0058] During the unloading stage (pressure drops from 50 MPa to 0 MPa), the magnetorheological fluid generator converts hydraulic energy into electrical energy, and the initial value of the supercapacitor bank SOC is 50%.

[0059] The energy management algorithm dynamically adjusts the recovery power according to the SOC: when the SOC rises to 70%, the recovery power maintains the rated value (10 kW); after 2 hours of continuous testing, the SOC stabilizes in the range of 65% - 75%, and the recovered energy meets 30% of the energy consumption requirements of the temperature-controlled oil source.

[0060] 4. Life assessment and decision-making

[0061] The improved Bayesian network model calculates the remaining life based on 500 cycles of data, and the prediction result is 1500 cycles (error 3.8%). The current HI = 0.78, and the high-frequency impact loading continues.

[0062] When the test reaches 1200 cycles, the HI drops to 0.28, triggering the static pressure holding loading mode, with the pressure controlled at 50 MPa ± 0.5 MPa. At the same time, the acoustic emission count rate rises to 50 times per minute, indicating the entry into the damage degradation period.

[0063] When the predicted life < 1000 cycles (measured as 950 cycles), the system issues a warning and automatically reduces the loading frequency to 5 Hz, extending the data acquisition period to capture the characteristics of microcrack propagation.

[0064] III. Implementation effects

[0065] Extreme working condition simulation: Achieve the synergistic effect of 75 MPa high pressure, a wide temperature range from -55 °C to 150 °C, and 20 Hz high-frequency dynamic loading, simulating the extreme working environment of aerospace hydraulic systems.

[0066] Multi-dimensional monitoring: The fiber Bragg grating strain sensor captures the stress concentration at the support (strain gradient of 200 με / mm), and the acoustic emission sensor detects the micro-crack initiation signal at the end of the service life (count rate suddenly increases to 200 times / minute), verifying the ability of early damage identification.

[0067] Energy recovery efficiency: The average energy recovery rate during the whole test is 76.5%. The supercapacitor bank realizes the closed-loop energy supply, and the system energy consumption is reduced by 40%.

[0068] Accuracy of remaining life prediction: The prediction error of the remaining life is controlled within 5%, which is more than 60% higher than the traditional single-parameter method, providing an accurate basis for the reliability verification of the accumulator.

[0069] The above embodiments are exemplary, aiming to illustrate the technical concept and features of the present invention, so that those skilled in this field can understand the content of the present invention and implement it accordingly. However, the protection scope of the present invention cannot be limited thereby. Any changes or modifications made according to the spirit and essence of the present invention should be covered within the protection scope of the present invention.

Claims

1. An accumulator durability test system, characterized in that Including: An extreme working condition simulation module, which is used to provide a high-pressure, wide temperature range and high-frequency dynamic loading environment. The pressure range of the high-pressure environment is 0 - 75 MPa, the temperature range of the wide temperature range environment is -55°C - 150°C, and the high-frequency dynamic loading frequency range is 0 - 20 Hz; A multi-sensor monitoring network, which includes fiber Bragg grating strain sensors arranged along the axis of the accumulator, triaxial acceleration sensors for capturing vibration signals, and acoustic emission sensors for locating material microcracks. The resolution of the fiber Bragg grating strain sensors is ≤ ±2 με, the sampling frequency of the triaxial acceleration sensors is ≥ 10 kHz, and the sensitivity of the acoustic emission sensors is ≥ 60 dB; An intelligent control and energy recovery module, which includes a hybrid controller based on a fuzzy PID algorithm and reinforcement learning, and an energy recovery device composed of a magnetorheological fluid generator and a supercapacitor bank. The response time of the hybrid controller is ≤ 50 ms, and the energy recovery rate of the energy recovery device is ≥ 75%; A data analysis and life assessment platform, which is configured with an improved Bayesian network model and a real-time health index calculation module. The improved Bayesian network model fuses multi-dimensional monitoring data for remaining life prediction, and the prediction error is ≤ 5%. The real-time health index is obtained by dynamically adjusting weights through the analytic hierarchy process; An edge computing node, which is used for real-time filtering, noise reduction and feature extraction of multi-sensor data. The preprocessing delay is ≤ 10 ms, and the preprocessed feature parameters include strain gradient, vibration main frequency, and acoustic emission event count rate.

2. The durability test system for an accumulator according to claim 1, wherein The extreme working condition simulation module includes: A two-stage booster cylinder, which is used to achieve a high-pressure output of 0 - 75 MPa. The pressure boost ratio of the two-stage booster cylinder under steady-state working conditions is 10:1 - 20:1; A temperature-controlled oil source, which includes a high-temperature electric heater, a liquid nitrogen cooler and a temperature sensor. The power of the high-temperature electric heater is ≥ 50 kW, and the cooling rate of the liquid nitrogen cooler is ≥ 10°C / min; A composite loading device, which is composed of a magnetorheological fluid damper and a piezoelectric ceramic actuator connected in series. The damping force adjustment range of the magnetorheological fluid damper is 0 - 50 kN, and the displacement resolution of the piezoelectric ceramic actuator is ≤ 1 μm.

3. The durability test system for an accumulator according to claim 1, wherein In the multi-sensor monitoring network: The fiber Bragg grating strain sensors are evenly distributed in 4 - 8 groups along the circumference of the accumulator, and each group contains 2 - 4 axial measuring points; The triaxial acceleration sensors are installed at the connection between the accumulator support and the pipeline to monitor the vibration acceleration signals in the X, Y, and Z directions; The acoustic emission sensors are attached to the middle of the accumulator shell through a coupling agent, and the sampling frequency is ≥ 1 MHz.

4. The durability test system for an accumulator according to claim 1, characterized in that, In the intelligent control and energy recovery module: The fuzzy PID algorithm of the hybrid controller adopts a Gaussian membership function, and the reinforcement learning algorithm is a deep Q network. The loading pressure fluctuation amplitude, temperature cycle rate, and loading frequency are dynamically adjusted through the real-time health index. For the energy recovery device, the equivalent series resistance of the supercapacitor bank is ≤10 mΩ, the magnetorheological fluid generator is connected to the hydraulic circuit through a speed coupler, and the recovered energy is used to drive the temperature control oil source and supply power to the sensors.

5. The durability test system for an accumulator according to claim 1, wherein In the data analysis and life assessment platform: The improved Bayesian network model is based on Bayes' theorem. By considering the joint probability distribution of multiple monitoring parameters, it realizes the prediction of the remaining life state of the accumulator. The conditional probability calculation formula is: where Ri (i = 1, 2,..., n) is the i-th monitoring parameter, n is the total number of monitoring parameters, including at least three of pressure, temperature, and strain; L is the remaining life state of the accumulator, and L′ represents all possible values of the remaining life state; P(L) is the prior probability, representing the estimation of the remaining life state before obtaining the monitoring data; P(R1|R2,..., Rn, L) is the likelihood function, describing the probability of observing the monitoring parameters R1, R2,..., Rn given the remaining life state L. The real-time health index calculation formula is: where the weight wi is dynamically adjusted according to the working condition type by the analytic hierarchy process, and Ri,min and Ri,max are the minimum and maximum values of the i-th monitoring parameter respectively.

6. A method for testing the durability of an accumulator based on the system according to any one of claims 1-5, characterized in that, It includes the following steps: Initial configuration: Set the target pressure, temperature range, and loading frequency, start the temperature control oil source, adjust the oil temperature to the target value through a high-temperature electric heater or a liquid nitrogen cooler, and use the temperature sensor for closed-loop feedback control. Dynamic loading and data acquisition: The hybrid controller adjusts the loading mode according to the real-time health index, and multiple sensors synchronously collect strain, vibration, and acoustic emission signals and transmit them to the edge computing node. Energy recovery and optimization: During the unloading stage, the magnetorheological fluid energy recovery device converts hydraulic energy into electrical energy, and the supercapacitor bank stores the energy and feeds it back to the temperature control oil source. Life assessment and decision-making: The improved Bayesian network model calculates the remaining life in real time. When the predicted life < 1000 cycles, an alarm is triggered and the test strategy is adjusted.

7. The test method according to claim 6, wherein The loading mode switching rule is: When the real-time health index HI > 0.8, start high-frequency impact loading, and the pressure fluctuation amplitude is ±10% - ±15% of the set target pressure. When the real-time health index HI < 0.3, switch to static pressure holding loading, and the pressure deviation is controlled within ±1%.

8. The test method according to claim 7, wherein In the energy recovery step, the energy management algorithm dynamically adjusts the recovery power according to the state of charge (SOC) of the supercapacitor: When SOC < 30%, the recovery power is increased to 120% of the rated value; when SOC > 80%, the recovery power is reduced to 80% of the rated value.

Images