High-pressure detection equipment for ESC systems based on the gas-liquid equivalent detection principle

Abstract

This application provides a high-pressure testing device for an ESC system based on the gas-liquid equivalent detection principle, relating to the field of automotive braking system testing technology. The device includes: a frame; a positioning, clamping, and sealing module; a high-pressure air circuit module; and a controller. The controller stores a gas-liquid equivalent mapping model, which is constructed by training a mapping transformation function using synchronous experimental data of the gas pressure and hydraulic pressure of the same ESC product, and using the coupling relationship between gas compressibility and cavity compliance derived from computational fluid dynamics simulation as constraints. The controller is configured to: control the positioning, clamping, and sealing module to fix and seal the ESC under test; control the high-pressure air circuit module to input high-pressure gas into the ESC under test, and control the solenoid valves and motors inside the ESC to operate according to a preset timing sequence; acquire real-time gas pressure data collected by multiple pressure sensors; input the gas pressure data into the gas-liquid equivalent mapping model, convert it into equivalent hydraulic performance parameters, compare it with a preset threshold, and output the test results.

Description

Technical Field

[0001] This invention belongs to the field of automotive braking system testing technology, specifically relating to a high-pressure testing device for ESC systems based on the gas-liquid equivalent testing principle. Background Technology

[0002] Brake fluid is typically used as the test medium for factory performance testing of Electronic Stability Control (ESC) systems. However, hydraulic testing suffers from problems such as medium contamination, high maintenance costs, and slow testing cycles. While using high-pressure gas instead of brake fluid can avoid these drawbacks, gas is compressible, and the response characteristics of the internal flow channels and cavities of the ESC to gas pressure differ significantly from those of a liquid.

[0003] Existing pneumatic testing solutions often directly apply the judgment logic of hydraulic testing, or simply convert pneumatic pressure data linearly and compare it with hydraulic thresholds. This ignores the nonlinear influence of the coupling relationship between gas compressibility and cavity compliance on pressure transmission, making it difficult for pneumatic testing results to truly and equivalently reflect the core indicators of ESC under actual hydraulic conditions, such as sealing performance, valve response, and pressure build-up capacity. The test confidence is low, and it cannot replace hydraulic testing for mass production. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, a high-pressure detection device for an ESC system based on the gas-liquid equivalent detection principle is provided, comprising:

[0005] frame;

[0006] Positioning and clamping sealing module, used to fix and seal the ESC to be tested;

[0007] The high-pressure gas circuit module is used to input high-pressure gas into the ESC under test; the high-pressure gas circuit module includes multiple pressure sensors for real-time acquisition of gas pressure data at each monitoring point during the detection process;

[0008] The controller stores a gas-liquid equivalent mapping model, which is trained using synchronous experimental data of gas pressure and hydraulic pressure of the same ESC product to train the mapping transformation function, and constructed using the coupling relationship between gas compressibility and cavity compliance analyzed by computational fluid dynamics simulation as a constraint condition.

[0009] The controller is configured to perform the following steps:

[0010] The positioning, clamping, and sealing module is controlled to fix and seal the ESC to be tested;

[0011] The high-pressure gas circuit module is controlled to input high-pressure gas into the ESC under test, and the solenoid valves and motors inside the ESC under test are controlled to operate according to a preset timing sequence.

[0012] Acquire air pressure data collected in real time by multiple pressure sensors;

[0013] The air pressure data is input into the gas-liquid equivalent mapping model and converted into equivalent hydraulic performance parameters.

[0014] The equivalent hydraulic performance parameters are compared with preset thresholds, and the detection results are output.

[0015] According to the technical solution provided in this application, the controller uses a minimum excitation set strategy to execute the internal solenoid valve of the controlled ESC according to a preset timing sequence.

[0016] The minimum incentive set strategy includes:

[0017] The various detection functions of the ESC under test are decoupled, and the optimal combination of pressure excitation and the optimal combination of flow excitation are selected.

[0018] The process involves three phases: rapid screening, targeted verification, and gray zone retesting.

[0019] During the rapid screening phase, the optimal pressure excitation combination and the optimal flow excitation combination are applied, air pressure data at each monitoring point are collected and compared with the rapid screening threshold, and the detection items are divided into three categories: qualified, questionable, and boundary.

[0020] During the targeted verification phase, only testing items deemed questionable are subject to targeted incentives for confirmatory testing.

[0021] During the gray zone retesting phase, the detection items determined to be boundaries are repeatedly stimulated and sampled multiple times. The statistical confidence level is calculated based on the sampling results. When the confidence level is lower than the preset threshold, the number of retests is extended until the threshold is reached or the maximum cycle time constraint is exceeded.

[0022] According to the technical solution provided in this application, the plurality of pressure sensors include a first pressure sensor disposed at the outlet end of the air pressure regulating valve, a second pressure sensor disposed in each independent detection air path, and a third pressure sensor disposed on the side of the ESC wheel cylinder to be tested.

[0023] When the pressure holding condition is steady or the inflation / deflation rate is lower than a preset gradual change threshold, the controller is also configured to:

[0024] Based on the reading difference and reading delay between the first and second pressure sensors, the compressibility compensation coefficient of the gas in the high-pressure gas path is calculated.

[0025] Based on the reading difference and reading delay between the second and third pressure sensors, the pressure decay and response lag time of the gas in the internal flow channel of the ESC are calculated.

[0026] The equivalent hydraulic performance parameters output by the gas-liquid equivalent mapping model are superimposed with the compressibility compensation coefficient and pressure attenuation to obtain the corrected equivalent hydraulic performance parameters.

[0027] If the response delay time exceeds the preset duration, it is determined that there is a blockage in the flow channel or that the valve is slow to operate.

[0028] According to the technical solution provided in this application, the controller is further configured to perform a pump capacity test, including:

[0029] The motor inside the ESC under test is started, driving the plunger pump to work;

[0030] During motor startup, the motor's drive current is collected, and pressure change data on the wheel cylinder side is collected through the third pressure sensor.

[0031] The actual pressure build-up rate is calculated based on the pressure change data, and the actual pressure build-up rate is compared with the preset standard pressure build-up rate.

[0032] If the driving current deviates from the preset standard current curve and the actual pressure build-up rate is lower than the standard pressure build-up rate, then the pump body is determined to have a mechanical fault.

[0033] If the driving current is within the preset standard current curve range but the actual pressure build-up rate is lower than the standard pressure build-up rate, it is determined that there is a leak in the high-pressure gas circuit module or the internal flow channel of the ESC under test.

[0034] According to the technical solution provided in this application, the controller is further configured to perform a high-pressure gas path thermodynamic transient correction step, which includes:

[0035] During the process of the high-pressure gas circuit module filling the internal cavity of the ESC under test with high-pressure gas or venting high-pressure gas from the internal cavity of the ESC under test, the transient pressure change curve of the pressure sensor from the start of filling or venting to the pressure stabilization time is recorded.

[0036] Based on the preset gas adiabatic process equation, combined with the fixed volume of the high-pressure gas circuit module and the equivalent volume of the internal cavity of the ESC under test, the instantaneous temperature change of the gas during the charging and discharging process is calculated in reverse from the pressure overshoot and pressure recovery time constant in the transient pressure change curve.

[0037] If the calculated instantaneous temperature change does not exceed the preset thermodynamic safety threshold, then the pressure data collected in real time by the pressure sensor is thermodynamically compensated and corrected according to the instantaneous temperature change, and the equivalent isothermal pressure value is calculated. This equivalent isothermal pressure value is then used to replace the original collected pressure data and is sent into the gas-liquid equivalent mapping model.

[0038] If the calculated instantaneous temperature change exceeds the thermodynamic safety threshold, it is determined that there is an abnormal thermal effect in the current test, the inflation or deflation rate is reduced, and the current test procedure is repeated.

[0039] According to the technical solution provided in this application, the controller is further configured as follows:

[0040] Obtain the operating parameters of the current detection process, including the inflation / deflation rate, the upstream / downstream pressure ratio, and the gas flow rate of each pressure sensor in the detection stage.

[0041] The installation positions of each pressure sensor are obtained, including a first position at the outlet of the pressure regulating valve, a second position in each independent detection air path, and a third position on the side of the ESC wheel cylinder to be tested.

[0042] Based on the operating parameters and installation location, each pressure sensor is divided into:

[0043] Type 1: No calibration required. Suitable for all pressure sensors in steady-state pressure holding or slowly changing conditions, or pressure sensors in the first position in non-steady-state conditions, or pressure sensors in the first and second positions in high-speed throttling conditions.

[0044] Type 2: Requires thermodynamic transient correction and is suitable for pressure sensors at the second and third positions in rapid charge and discharge transient conditions.

[0045] Specifically, for pressure sensor readings classified as the first type, the controller directly inputs them into the gas-liquid equivalent mapping model; for pressure sensor readings classified as the second type, the high-pressure gas path thermodynamic transient correction step is invoked for processing, and the equivalent isothermal pressure value obtained after correction is input into the gas-liquid equivalent mapping model.

[0046] According to the technical solution provided in this application, the controller is further configured as follows:

[0047] After the high-pressure gas circuit module fills the internal cavity of the ESC under test with high-pressure gas and reaches a stable pressure, the shut-off valve located upstream of the ESC under test is controlled to close instantly, and the pressure decay curve of the pressure sensor within a preset time window after the shut-off valve closes is recorded.

[0048] Based on the initial slope of the pressure decay curve, the equivalent volume of the internal cavity of the ESC under test is deduced by looking up a table. The mapping relationship used for the table lookup is pre-established and stored in the controller through calibration experiments.

[0049] According to the technical solution provided in this application, the gas-liquid equivalent mapping model is pre-built and stored in the controller through the following steps:

[0050] Based on the 3D CAD model of the ESC hydraulic actuator to be tested, a computational fluid dynamics simulation model is constructed; the physical property parameters of the pneumatic medium and the hydraulic medium are set respectively, and transient simulation is performed under the same boundary conditions to obtain the pressure response curves at the corresponding detection points under pneumatic and hydraulic conditions.

[0051] The deviation between the gas pressure simulation results and the hydraulic simulation results is analyzed, and a gas compressibility compensation function and a cavity compliance compensation function are established as physical constraint priors for the gas-liquid equivalent mapping model.

[0052] Synchronous experimental data of multiple ESC products of the same model on pneumatic and hydraulic testing equipment were collected to form a pneumatic-hydraulic feature dataset.

[0053] Using the physical constraint prior as a regularization term and the gas-hydraulic feature dataset as training samples, a mapping transformation function with physical constraints is trained; the input of the mapping transformation function is the gas pressure detection feature parameters, and the output is the equivalent hydraulic performance parameters.

[0054] The trained mapping transformation function is stored in the controller as the gas-liquid equivalent mapping model.

[0055] According to the technical solution provided in this application, it also includes a whisker-type micro switch and a safety light curtain;

[0056] The controller is also configured to:

[0057] When the trigger of the touch-type micro switch is detected, the positioning and sealing module is controlled to start operating, and the safety light curtain is activated simultaneously to form a safety protection light curtain.

[0058] When an obstruction signal is received from the safety light curtain due to obstruction, an emergency stop is triggered.

[0059] According to the technical solution provided in this application, the positioning and pressing sealing module includes an X-axis hydraulic cylinder and a Z-axis hydraulic cylinder; the X-axis hydraulic cylinder is used to extend and retract along the X-axis direction to achieve X-axis positioning of the ESC under test; the Z-axis hydraulic cylinder is used to extend and retract along the Z-axis direction to achieve Z-axis pressing of the ESC under test, and the bottom of the Z-axis hydraulic cylinder is provided with two air outlet ports, which are respectively connected to the two main cylinder inlets of the ESC under test and connected to the high-pressure air circuit module.

[0060] Compared with existing technologies, the advantages of this application are as follows: The gas-liquid equivalent mapping model stored in the controller of this device uses the coupling relationship between gas compressibility and cavity compliance derived from computational fluid dynamics simulation as physical constraints, and trains the mapping transformation function using synchronous experimental data of gas pressure and hydraulic pressure from the same ESC product, achieving high-fidelity conversion from gas pressure detection data to equivalent hydraulic performance parameters. This model takes into account both the unique physical laws of gas medium, such as compressibility and cavity compliance, and the statistical characteristics of measured data, enabling the gas pressure detection results to truly reflect the performance parameters of the ESC under actual hydraulic conditions. It solves the problem of detection distortion caused by differences in the physical properties of gas and liquid media, providing a high-confidence, media-free alternative for high-pressure performance testing of ESCs. Attached Figure Description

[0061] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0062] Figure 1 A front structural schematic diagram of the high-pressure detection equipment of the ESC system based on the gas-liquid equivalent detection principle provided in this application;

[0063] Figure 2 A schematic diagram of the back structure of the high-pressure detection equipment of the ESC system based on the gas-liquid equivalent detection principle provided in this application;

[0064] The text labels in the image represent:

[0065] 1. Frame; 2. Touch screen; 3. Microswitch; 4. X-axis hydraulic cylinder; 5. Power distribution cabinet; 6. Switch protective cover; 7. Emergency stop button; 8. Test piece positioning fixture; 9. Defective product tray; 10. Good product tray; 11. High-pressure air supply circuit; 12. Z-axis hydraulic cylinder; 13. Alarm indicator light; 14. Silencer air outlet; 15. Air distribution block; 16. Shut-off valve; 17. Air pressure regulating valve; 18. High-pressure air inlet block; 19. Air control unit; 20. Digital pressure gauge. Detailed Implementation

[0066] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0067] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0068] Example 1

[0069] As mentioned in the background section, this application proposes a high-pressure detection device for an ESC system based on the gas-liquid equivalent detection principle, such as... Figure 1-2 As shown, it includes:

[0070] Rack 1;

[0071] Positioning and clamping sealing module, used to fix and seal the ESC to be tested;

[0072] The high-pressure gas circuit module is used to input high-pressure gas into the ESC under test; the high-pressure gas circuit module includes multiple pressure sensors for real-time acquisition of gas pressure data at each monitoring point during the detection process;

[0073] The controller stores a gas-liquid equivalent mapping model, which is trained using synchronous experimental data of gas pressure and hydraulic pressure of the same ESC product to train the mapping transformation function, and constructed using the coupling relationship between gas compressibility and cavity compliance analyzed by computational fluid dynamics simulation as a constraint condition.

[0074] Furthermore, it also includes a whisker-type micro switch 3 and a safety light curtain;

[0075] The controller is also configured to:

[0076] When the triggering of the touch-type micro switch 3 is detected, the positioning and clamping sealing module is controlled to start operating, and the safety light curtain is activated simultaneously to form a safety protection light curtain.

[0077] When an obstruction signal is received from the safety light curtain due to obstruction, an emergency stop is triggered.

[0078] Furthermore, the positioning, pressing, and sealing module includes an X-axis hydraulic cylinder 4 and a Z-axis hydraulic cylinder 12; the X-axis hydraulic cylinder 4 is used to extend and retract along the X-axis direction to achieve X-axis positioning of the ESC under test; the Z-axis hydraulic cylinder 12 is used to extend and retract along the Z-axis direction to achieve Z-axis pressing of the ESC under test, and the bottom of the Z-axis hydraulic cylinder 12 is provided with two air outlet ports, which are respectively connected to the two main cylinder inlets of the ESC under test and connected to the high-pressure air circuit module.

[0079] Specifically, frame 1 is an aluminum profile frame structure, serving as the main load-bearing structure for the entire equipment. The bottom of frame 1 is equipped with a support structure featuring casters and leveling feet, facilitating equipment movement and workstation leveling. A power distribution cabinet 5 is located at the bottom of frame 1, integrating core electrical components such as a motion control module, data acquisition card, and communication module, providing power and control support for the entire machine. A switch protective cover 6 is installed on the outside of the operating buttons to prevent accidental button presses during operation, which could lead to equipment malfunction. An emergency stop button 7 is installed in a prominent position on the equipment's operating panel for stopping the equipment in emergency situations. An alarm indicator light 13 is installed on the top of frame 1 to indicate equipment malfunctions and test results (e.g., green for qualified, red for unqualified, yellow for fault). A microswitch 3 and a safety light curtain (not specifically labeled in the diagram) constitute the safety protection subsystem. The microswitch 3 is installed on the side of the operating station to trigger the positioning and clamping procedure. The safety light curtain is arranged around the equipment's operating area, forming a safety light curtain.

[0080] Specifically, the positioning and clamping sealing module includes a test piece positioning fixture 8, an X-axis hydraulic cylinder 4, and a Z-axis hydraulic cylinder 12. The test piece positioning fixture 8 is fixed to the equipment operating table, and the ESC under test is precisely positioned and fixed in the Y-axis direction through positioning pins and limiting structures. The X-axis hydraulic cylinder 4 is arranged horizontally along the X-axis direction, and its piston rod end is connected to the positioning clamp. When the X-axis hydraulic cylinder 4 extends, it pushes the clamp against the ESC under test, completing the positioning and fixing in the X-axis direction; when it retracts, the positioning is released. The Z-axis hydraulic cylinder 12 is arranged vertically along the Z-axis direction, located above the ESC under test. The bottom of the Z-axis hydraulic cylinder 12 has two air outlets, which are respectively connected to the inlets of the two main cylinder chambers (MC1 chamber and MC2 chamber) of the ESC under test, and are connected to the high-pressure air circuit module. When the Z-axis hydraulic cylinder 12 extends downward, its bottom sealing joint presses against the inlets of the MC1 and MC2 chambers of the ESC under test, completing the clamping and air circuit sealing in the Z-axis direction. High-Pressure Gas Circuit Module: This module is used to input high-pressure gas into the ESC under test and collect pressure data in real time. It mainly includes the following components: High-Pressure Inlet Block 18: Installed on the back or side of the frame 1, it is used to centrally connect to an external high-pressure gas source (suitable for ≥15MPa testing conditions), serving as the power input interface for the entire gas circuit system. Gas Pressure Regulating Valve 17: Connected in series with the high-pressure inlet block 18, it is used to stably regulate the gas source pressure to the standard working pressure required for ESC testing (such as 5MPa, 15MPa, etc.) according to preset testing process parameters. Gas Distribution Block 15: Connected to the outlet of the gas pressure regulating valve 17, it is used to distribute the regulated high-pressure gas into multiple independent testing gas circuits. Shut-off Valve 16: Connected in series with each independent testing gas circuit output from the gas distribution block 15, it realizes independent on / off control of a single loop, facilitating phased testing and equipment debugging. Gas Control Unit 19: Internally integrates multiple solenoid valves and reversing valves; each independent testing gas circuit is connected to the gas control unit 19. The pneumatic control unit 19 precisely controls the on / off, pressure holding, and pressure release sequence of each air path according to the controller's instructions. Pressure sensors include multiple sensors, specifically: a first pressure sensor located at the outlet of the pressure regulating valve 17 to monitor the pressure of the air source after pressure regulation; a second pressure sensor located in each independent detection air path (i.e., after the air distribution block 15 and before or after the shut-off valve 16) to monitor the real-time pressure of each detection branch; and a third pressure sensor located on the side of the ESC wheel cylinder under test (i.e., in the air path between the bottom air outlet of the Z-axis hydraulic cylinder 12 and the inlet of the main cylinder cavity of the ESC under test) to monitor the pressure entering the internal cavity of the ESC. A silencer outlet 14 connects to the exhaust port of the pneumatic control unit 19 and has a built-in noise reduction structure (such as porous sound-absorbing cotton or an expansion silencer) to reduce noise when releasing high-pressure gas after testing. A digital pressure gauge 20 is optionally installed on the equipment panel for on-site operators to visually observe the current air path pressure. High-pressure air supply circuit 11: It consists of several high-pressure pneumatic hard pipes and connectors, which connect the above-mentioned pneumatic components to form a complete air circuit.Electrical Control and Human-Machine Interaction Module: Human-Machine Interaction Touch Screen 2: Installed on the equipment operation panel, used for setting detection parameters, process monitoring, and displaying detection reports. Controller (Motion control module, data acquisition card, communication module, etc. integrated in distribution cabinet 5): The controller stores a gas-liquid equivalent mapping model and executes all control steps. Data Traceability and Sorting Module: Good Product Pallet 10 and Defective Product Pallet 9: Located on the equipment's discharge side, used to sort ESC products to the corresponding pallets using a robotic arm or manually based on the detection results. Communication Module (Integrated in distribution cabinet 5): Used to upload detection data to the MES system in real time, enabling full-process data traceability during the detection process.

[0081] Hardware Connection and Workflow Summary: An external high-pressure air source is connected to the device via a high-pressure air inlet block 18, flowing sequentially through a pressure regulating valve 17, an air distribution block 15, a shut-off valve 16, and a pneumatic control unit 19. Finally, it enters the MC1 and MC2 chambers of the ESC under test through the air outlet at the bottom of the Z-axis hydraulic cylinder 12. Multiple pressure sensors are installed at the outlet of the pressure regulating valve 17, in each independent detection air path, and on the wheel cylinder side of the ESC under test, respectively, to collect air pressure data in real time and send it to the controller. The operator or robot places the ESC under test on the positioning fixture 8, and after toggling the microswitch 3, the controller controls the X-axis hydraulic cylinder 4 and the Z-axis hydraulic cylinder 12 to move sequentially to complete the positioning and clamping, while simultaneously activating the safety light curtain to form a safety protection. During the test, the controller controls the pneumatic control unit 19 and the solenoid valves inside the ESC under test according to a preset timing sequence, collects pressure sensor data, and calls the gas-liquid equivalent mapping model for conversion and judgment. After the test is completed, the controller controls the air circuit to depressurize, and the gas is discharged through the silencer outlet 14. The hydraulic cylinder is reset, and the products are sorted to the good product tray 10 or the defective product tray 9 according to the test results.

[0082] Example 2

[0083] Based on Example 1, this example proposes that the controller be configured to perform the following steps:

[0084] The positioning, clamping, and sealing module is controlled to fix and seal the ESC to be tested;

[0085] The high-pressure gas circuit module is controlled to input high-pressure gas into the ESC under test, and the solenoid valves and motors inside the ESC under test are controlled to operate according to a preset timing sequence.

[0086] Specifically, the controller retrieves the corresponding timing parameters from the testing timing table based on the model of the ESC under test and the current testing procedure. Taking eight testing items as an example, the testing timing table stores the following parameters:

[0087] High pressure leakage test: The air pressure regulating valve 17 outputs 15.5MPa, and the pressure is maintained after 3 seconds of inflation. After 6 seconds of pressure maintenance, the sample is taken. The valve excitation state is to excite the pressure reducing valves corresponding to the right rear wheel and the left front wheel, while the other valves remain at their default settings.

[0088] Pressurization rate test: The pressure regulating valve 17 outputs 5MPa, the inflation time is 2 seconds, the sampling time is 0.7 seconds, and the valve status is that the pressure relief valve and the pressurizing valve are open, and the pressure reducing valve and the suction valve are closed.

[0089] Pressure relief rate test: After inflation to 5MPa in 2 seconds, the pressure relief valve is opened for 0.7 seconds. The sampling time is 0.7 seconds after the pressure relief valve is opened.

[0090] Pressure relief valve test: After inflating to 5MPa for 2 seconds, the pressure relief valve is activated to disconnect, and sampling is performed 10 seconds later.

[0091] Pressure booster valve closure test: Excite the pressure booster valve to close, and sample after 5 seconds of inflation.

[0092] Pressure booster valve opening test: Excite the pressure booster valve to close, then open it for 0.4 seconds before sampling.

[0093] Pressure reducing valve test: The pressure reducing valve is activated and sampled after 200 milliseconds.

[0094] Pump capacity test: Inflate to 5MPa in 2 seconds, excite the right rear wheel pressure relief valve to open and start the motor, and sample after 2 seconds.

[0095] The controller executes the eight processes sequentially according to the timing schedule. The pressure data collected in each process is subjected to thermodynamic correction (if applicable), then input into the gas-liquid equivalent mapping model to convert it into equivalent hydraulic performance parameters, and finally compared with the corresponding threshold. After all processes are completed, the controller outputs the comprehensive test results.

[0096] Furthermore, since the test conditions for the right rear wheel and the left front wheel are the same as those for the right front wheel and the left rear wheel, the following test conditions will be explained using the right rear wheel and the left front wheel as examples.

[0097] The high-pressure leakage test process is as follows: The pressure-reducing valves corresponding to the right rear wheel and left front wheel are activated, while the remaining pressure-limiting valves remain open by default, the intake valve remains closed by default, and the booster valve remains open by default. High-pressure gas of 15–16 MPa is introduced through the high-pressure intake block 18 for 3 seconds. The high-pressure gas flows sequentially through the pressure-limiting valve, booster valve, right and left front wheel cylinders, and pressure-reducing valve via port MC1. Since both sides of the intake valve are supplied with high-pressure gas, it is not necessary to open the intake valve. After the system maintains pressure for 6 seconds, the pressure of the right and left front wheel cylinders is collected. Once the pressure is deemed acceptable (optional: a pressure drop of less than 5 bar is considered acceptable), the equipment automatically proceeds to the next testing step.

[0098] The pressurization rate test process is as follows: the pressure relief valve and the pressurization valve remain open in their default states, while the pressure reducing valve and the intake valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced through the high-pressure intake block 18. After being stabilized to 5 MPa by the pressure regulating valve 17, the gas flows sequentially through the pressure relief valve and the pressurization valve from port MC1 and then merges into the right front wheel and left front wheel cylinders. The system collects the pressure of the right front wheel and left front wheel cylinders at 0.7s. After the pressure is deemed qualified (optional: pressure not lower than 5 MPa is considered qualified), the equipment automatically proceeds to the next testing step.

[0099] The pressure relief rate test process is as follows: the pressure relief valve and booster valve remain open in their default states, while the pressure reducing valve and intake valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18. After being stabilized to 5 MPa by the pressure regulating valve 17, the gas is inflated for 2 seconds. The gas flows sequentially from port MC1 through the pressure relief valve and booster valve, then merges into the right front wheel and left front wheel cylinders. At this time, the pressure relief valve is activated to open, releasing the pressure in the air path from the pressure relief valve to the ESC section. The gas is then discharged from the system through port MC1 of the ESC via the muffler outlet 14. Subsequently, the activation is canceled, and the pressure relief valve returns to its default open state. The pressure portion locked in the left front wheel and right rear wheel cylinders is released into the air path from the pressure relief valve to the ESC section. Afterward, the activation is canceled, and the pressure relief valve remains open in its default state. The pressure portion locked in the left front wheel and right rear wheel cylinders is released into the air path from the pressure relief valve to the ESC section. When the pressure relief valve opens for 0.7 seconds, the system collects the cylinder pressure of the left front wheel and the right rear wheel. After the pressure is deemed qualified (optional: pressure below 15 bar is considered qualified), it automatically proceeds to the next testing procedure.

[0100] The pressure relief valve test process is as follows: the pressure relief valve and booster valve remain open by default, while the pressure reducing valve and intake valve remain closed by default. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18. After being stabilized to 5 MPa by the pressure regulating valve 17, the gas is inflated for 2 seconds. The gas flows sequentially through the pressure relief valve and booster valve from port MC1, then merges into the right front wheel and left front wheel cylinders. At this point, the pressure relief valve is activated to open, releasing the pressure in the ESC section of the air path. After 10 seconds, the system collects the pressure of the left front wheel and right rear wheel cylinders. Once the pressure is deemed acceptable (optional: pressure above 40 bar is considered acceptable), the system automatically proceeds to the next testing step.

[0101] The booster valve closure test process is as follows: the booster valve is activated to close, the pressure relief valve remains open in its default state, and the intake valve and pressure reducing valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18, and after being stabilized to 5 MPa by the pressure regulating valve 17, the gas is inflated for 5 seconds. The gas flows sequentially through the pressure relief valve and the booster valve from port MC1 and then merges into the right front wheel and left front wheel cylinders. After 5 seconds, the system collects the cylinder pressure of the left front wheel and right rear wheel. After the pressure is deemed qualified (optional: pressure not exceeding 25 bar is considered qualified), the system automatically proceeds to the next testing procedure.

[0102] The booster valve opening test process is as follows: the booster valve is closed, the pressure relief valve remains open in its default state, and the intake valve and pressure reducing valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18. The gas flows sequentially through the pressure relief valve and the booster valve from port MC1, then merges into the right front wheel and left front wheel cylinders. The booster valve is opened for 0.4 seconds. After 0.4 seconds, the system collects the pressure of the left front wheel and right rear wheel cylinders. Once the pressure is deemed acceptable (optional: pressure not exceeding 70 bar is considered acceptable), the system automatically proceeds to the next testing step.

[0103] The pressure reducing valve test process is as follows: the pressure limiting valve and the pressure boosting valve remain open in their default states, while the pressure reducing valve and the suction valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18. The gas flows sequentially through the pressure limiting valve and the pressure boosting valve from port MC1, then converges into the right rear wheel and left front wheel cylinders. This activates the pressure reducing valve to open. The system collects the pressure of the left front wheel and right rear wheel cylinders every 200 ms. After passing the test (optionally, pressure within the range of 10 MPa–13 MPa is considered passing), the system automatically proceeds to the next testing step.

[0104] The pump capacity test process is as follows: the pressure relief valve and booster valve remain open in their default states, while the pressure reducing valve and suction valve remain closed in their default states. High-pressure gas of 15–16 MPa is introduced into the high-pressure intake block 18. After being stabilized to 5 MPa by the pressure regulating valve 17, the gas is inflated for 2 seconds. The gas flows sequentially from port MC1 through the pressure relief valve and booster valve before converging into the right front wheel and left front wheel cylinders. The right rear wheel pressure reducing valve is activated to open, while the left front wheel pressure reducing valve remains closed. Simultaneously, the motor is activated and its current is monitored. The plunger pump starts operating, and the gas sequentially passes through the right rear wheel pressure reducing valve, the plunger pump, and the left front wheel booster valve before entering the left front wheel cylinder. The system collects the cylinder pressure of the left front wheel and right rear wheel at 2 seconds. Once the pressure is deemed acceptable (optionally, a pressure increase of 2.1 bar or more is considered acceptable), the gas test ends. After the test, the equipment sorts the ESC products to the good product tray 10 or the defective product tray 9 using a robotic arm or manually, based on the test results.

[0105] Acquire air pressure data collected in real time by multiple pressure sensors;

[0106] The air pressure data is input into the gas-liquid equivalent mapping model and converted into equivalent hydraulic performance parameters.

[0107] The equivalent hydraulic performance parameters are compared with preset thresholds, and the detection results are output.

[0108] Specifically, Step 1: Fixing and sealing the ESC under test: The operator or robot places the ESC under test on the positioning fixture 8, aligning the positioning hole at the bottom of the ESC with the positioning pin on the fixture to achieve Y-axis positioning. The operator flicks the micro switch 3 with their finger, which outputs a rising edge signal to the controller. After receiving the signal, the controller first sends an extension command to the solenoid valve of the X-axis hydraulic cylinder 4 through the digital output module. After the solenoid valve is energized, hydraulic oil enters the rodless chamber of the X-axis hydraulic cylinder 4, and the piston rod extends at a speed of 10 mm / s, pushing the positioning fixture to move to the side of the ESC. When the magnetostrictive displacement sensor installed on the X-axis hydraulic cylinder 4 detects that the piston rod has moved a distance that reaches the preset positioning stroke (e.g., 30 mm), the controller checks whether the pressure relay has reached a clamping force of 500 N. If the standard is met, the X-axis positioning is determined to be complete, the holding current of the X-axis solenoid valve is cut off, and the pressure is maintained only by the hydraulic lock. Subsequently, the controller sends an extension command to the solenoid directional valve of the Z-axis hydraulic cylinder 12. The piston rod of the Z-axis hydraulic cylinder 12 extends downward at a speed of 5 mm / s, with its two exhaust ports at the bottom aligned with the inlets of chambers MC1 and MC2 of the ESC, respectively. When the displacement sensor of the Z-axis hydraulic cylinder 12 detects that the movement distance has reached 20 mm, the sealing joint contacts the inlet of the main cylinder chamber and compresses the O-ring. When the contact pressure reaches 1 MPa, the controller determines that the clamping is complete and cuts off the current to the Z-axis solenoid valve. At the same time, the controller sends a start signal to the safety light curtain, which begins to emit infrared beams, forming a protective light curtain in the operating area.

[0109] Step 2: Input high-pressure gas and control the solenoid valve and motor according to the preset timing sequence: The controller calls the corresponding parameters from the internally stored detection timing table according to the current detection item. Taking the high-pressure leakage test as an example, the detection timing sequence is as follows: The controller sends a 12 mA current signal (corresponding to 15.5 MPa output) to the air pressure regulating valve 17 through the analog output module. The air pressure regulating valve 17 adjusts its opening according to the internal PID controller to stabilize the outlet pressure at 15.5 ± 0.1 MPa. The controller sends a high-level signal to the main air intake solenoid valve in the air control unit 19 through the digital output module. The main valve opens, and the high-pressure gas enters the MC1 and MC2 chambers of the ESC from the high-pressure air intake block 18 through the air pressure regulating valve 17, the air distribution block 15, the shut-off valve 16, the air control unit 19, and the bottom interface of the Z-axis hydraulic cylinder 12.

[0110] Simultaneously, the controller sends excitation signals to the solenoid valves inside the ESC. The valve status for the high-pressure leakage test is as follows: the pressure reducing valves corresponding to the right rear wheel and left front wheel are activated (sending 24V DC power); the remaining pressure limiting valves remain open by default (not energized); the intake valve remains closed by default (not energized); and the booster valve remains open by default (not energized). The controller outputs these signals through a dedicated valve drive board, and each drive channel has overcurrent protection and short-circuit detection functions. After inflation lasts for 3 seconds, the controller closes the main intake solenoid valve and enters the pressure holding phase.

[0111] For pump capacity testing, after inflation is complete, the controller sends a pulse width modulation signal with a frequency of 10 kHz to the motor drive circuit inside the ESC. The duty cycle gradually increases from 0 to 100%, with an acceleration time of 200 milliseconds, causing the plunger pump to reach its rated speed. Simultaneously, the controller activates the right rear wheel depressurization valve to open, while the left front wheel depressurization valve remains closed.

[0112] Step 3: Acquire real-time air pressure data from multiple pressure sensors: The controller activates the data acquisition card for synchronous sampling. The outputs of the first, second, and third pressure sensors are all 4 to 20 mA current, which is converted to 1 to 5 volts by a 250-ohm precision resistor, and then converted to digital values ​​by a 16-bit analog-to-digital converter at a sampling rate of 1000 Hz. The controller reads the values ​​of each channel every 1 millisecond and converts them into pressure values. The conversion formula is: Pressure value (MPa) = (Voltage value - 1) / 4 × 20. The converted data is stored in a circular buffer, with 4096 storage units allocated to each sensor. The controller performs median filtering on the raw data: five consecutive sampling points are taken, sorted by value, and the median value is taken as the effective pressure to eliminate impulse noise. For high-pressure leakage testing, the controller records the reading of the third pressure sensor at the 6th second after the start of pressure holding, denoted as P_end, and compares it with the pressure at the end of inflation, P_start, to calculate the pressure drop ΔP = P_start - P_end.

[0113] Step 4: Input the air pressure data into the gas-liquid equivalent mapping model to convert it into equivalent hydraulic performance parameters: The controller assembles the extracted air pressure feature parameters into an input vector. For the high-pressure leakage test, the input features include: pressure difference ΔP before and after pressure holding, pressure holding time of 6 seconds, ambient temperature (read by the temperature sensor inside the chassis), and gas type identifier (air). The controller calls the application programming interface of the gas-liquid equivalent mapping model to pass the input vector to the model. The internal structure of the model is a three-layer feedforward neural network with 6 nodes in the input layer (corresponding to: pressure decay rate, pressure rise slope, pressure fall time constant, pressure change amplitude, pressure fluctuation frequency, and ambient temperature, respectively). Unused features are set to zero for different detection items. For example, the high-pressure leakage test only uses the pressure decay rate and ambient temperature; the pressurization rate test only uses the pressure rise slope, etc.), 12 nodes in the hidden layer, and 4 nodes in the output layer. The four nodes of the output layer correspond to: equivalent hydraulic leakage Q_leak (unit: ml / h), equivalent hydraulic pressure hold rate R_hold (dimensionless), equivalent hydraulic volume V_eq (unit: cubic centimeters), and confidence score. The model performs forward propagation calculations: the input vector is multiplied by the input weight matrix W1 (12×6) to obtain the net input to the hidden layer, and after adding a bias b1, it is activated by the hyperbolic tangent function to obtain the hidden layer output; the hidden layer output is multiplied by the output weight matrix W2 (4×12) and added to the bias b2, and the output layer uses a linear activation function. In the calculation results, the equivalent hydraulic leakage is the core output of this detection item. For the boost rate test, the controller input is the pressure value collected at 0.7 seconds, and the model outputs the equivalent hydraulic boost rate.

[0114] Step 5: Compare the equivalent hydraulic performance parameters with the preset threshold and output the test results: The controller reads the judgment threshold corresponding to the item from the parameter storage area. The pass threshold for the high-pressure leakage test is that the equivalent hydraulic leakage is less than 5 ml / hour. If the model output Q_leak<5, the item is judged as qualified; otherwise, it is unqualified. After all test items are completed, the controller summarizes the judgment results of each item into a one-byte status code, with each bit representing whether an item is qualified or not. The controller displays the result table on the touch screen 2 and marks it with different colors (green for qualified, red for unqualified). At the same time, the controller drives the alarm indicator 13 through digital output: the green light is always on when all are qualified, and the red light flashes and the buzzer sounds briefly when any item is unqualified. The controller packages the test results, pressure curve data, product QR code and other information into JSON format through the Ethernet module and sends it to the MES system. Finally, the controller controls the exhaust solenoid valve in the pneumatic control unit 19 to open, and the high-pressure gas is discharged through the silencer outlet 14 for 2 seconds. The controller then sends a retraction command to the X-axis hydraulic cylinder 4 and the Z-axis hydraulic cylinder 12, and the hydraulic cylinders are reset. Based on the overall results, the controller sends a signal to the sorting robot: if the product is qualified, the robot will pick up the ESC and place it on the good product pallet 10; if the product is unqualified, it will pick it up and place it on the defective product pallet 9.

[0115] In a preferred embodiment, the controller employs a minimum excitation set strategy to execute the internal solenoid valve of the controlled ESC according to a preset timing sequence.

[0116] The minimum incentive set strategy includes:

[0117] The various detection functions of the ESC under test are decoupled, and the optimal combination of pressure excitation and the optimal combination of flow excitation are selected.

[0118] The process involves three phases: rapid screening, targeted verification, and gray zone retesting.

[0119] During the rapid screening phase, the optimal pressure excitation combination and the optimal flow excitation combination are applied, air pressure data at each monitoring point are collected and compared with the rapid screening threshold, and the detection items are divided into three categories: qualified, questionable, and boundary.

[0120] During the targeted verification phase, only testing items deemed questionable are subject to targeted incentives for confirmatory testing.

[0121] During the gray zone retesting phase, the detection items determined to be boundaries are repeatedly stimulated and sampled multiple times. The statistical confidence level is calculated based on the sampling results. When the confidence level is lower than the preset threshold, the number of retests is extended until the threshold is reached or the maximum cycle time constraint is exceeded.

[0122] Specifically, functional decoupling and excitation optimization: The controller pre-decouples the eight detection functions of the ESC through offline experiments. The experiments were conducted on a prototype during the R&D phase, using 10 qualified ESC products from the same batch. For each detection function, the controller independently applies excitation and measures the response, while simultaneously recording the sensor outputs corresponding to other functions. Taking the pressure boosting rate test and pressure relief valve test as examples, both involve pressure relief valves and pressure boosting valves. The controller conducted two sets of experiments: the first set only performed the pressure boosting rate test (pressure relief valve remained open, pressure boosting valve opened, pressure relief valve not energized), and the second set only performed the pressure relief valve test (pressure relief valve was energized to open after inflation). The pressure curves of the third pressure sensor were recorded under both sets of experiments, and the cross-sensitivity coefficient was calculated. The cross-sensitivity coefficient is defined as: the change in the characteristic parameter corresponding to function B when function A is executed, divided by the excitation amplitude of function A. Calculations show that the influence coefficient of the pressure boosting rate test on the pressure relief valve characteristics is 0.12, and the influence coefficient of the pressure relief valve test on the pressure boosting rate characteristics is 0.08, both less than 0.15. Therefore, it is considered that the two can be decoupled, meaning that they can be tested sequentially within the same testing cycle without interfering with each other.

[0123] The controller employs a genetic algorithm to select the optimal combination of pressure and flow excitation. The objective function is to maximize the defect coverage Cov, minimize the total detection time T, and simultaneously ensure a false positive rate of less than 0.5%. Decision variables include: output pressure P_set of pressure regulating valve 17 (range 0 to 20 MPa), inflation time t_fill (range 0.1 to 5 seconds), sampling delay t_sample (range 0.1 to 3 seconds), and the excitation timing of each solenoid valve (binary encoding). Genetic algorithm parameters: population size 100, crossover probability 0.8, mutation probability 0.05, generation number 200. The fitness function is F = w1·Cov + w2·(1 / T) + w3·(1-false positive rate), where w1 = 0.5, w2 = 0.3, and w3 = 0.2. After 200 generations of evolution, the optimal solution is: P_set=5.2 MPa, t_fill=2 seconds, t_sample=0.7 seconds, with the pressure boosting valve open, pressure reducing valve closed, pressure limiting valve open, and suction valve closed. This combination can simultaneously expose three common defects: pressure boosting rate, pressure relief rate, and pressure limiting valve.

[0124] Three - level detection process: The first level: Rapid screening stage. The controller performs a rapid detection according to the above - mentioned optimal excitation combination. Taking the right rear wheel and the left front wheel as an example, the controller controls the pneumatic pressure regulating valve 17 to output 5.2 MPa, opens the main intake solenoid valve for 2 seconds, and collects the pressure value of the third pressure sensor at the 0.7 - second moment, denoted as P_raw. Input P_raw into the gas - liquid equivalent mapping model to obtain the equivalent hydraulic pressure P_eq. The rapid screening threshold is set to 1.2 times or 0.8 times the qualified threshold (depending on the index direction: specifically, for lower - limit type indexes (such as pressure not less than a certain value), the rapid screening threshold is 0.8 times the qualified threshold; for upper - limit type indexes (such as leakage amount not higher than a certain value), the rapid screening threshold is 1.2 times the qualified threshold). In this embodiment, the qualified threshold of the pressurization rate is P_eq ≥ 5.2 MPa, so the lower limit of rapid screening is set to 4 MPa. If P_eq ≥ 5.2 MPa, it is determined as qualified; if 4 ≤ P_eq < 5.2 MPa, it is determined as doubtful; if P_eq < 4 MPa, it is determined as marginal. For other items such as high - pressure leakage testing, the qualified threshold is leakage amount ≤ 5 ml / h, the rapid screening threshold is set to ≤ 6 ml / h, leakage amount > 6 ml / h is marginal, and between 5 and 6 is doubtful.

[0125] The second level: Directional verification stage. The controller collects all the items marked as doubtful, sorts them according to the priority (the priority is set according to historical statistical data, and the items with high missed - detection risk are prioritized). For each doubtful item, the controller calls the dedicated parameters from the directional detection parameter library. For example, when the pressurization rate test is doubtful, the directional parameters are: the output pressure of the pneumatic pressure regulating valve 17 is accurate to 5.00 ± 0.02 MPa, the inflation time is extended to 5 seconds, the sampling delay is still 0.7 seconds, and samples are continuously collected 3 times and averaged. The controller pauses other unfinished items in the rapid screening stage and separately executes the directional detection of this item. After the detection is completed, it is judged using the standard qualified threshold (5 MPa), and the result is used as the final conclusion without re - testing in the gray area.

[0126] The third level: Gray - area re - testing stage. For the items marked as marginal, the controller starts the statistical re - testing process. Set the initial re - testing times n = 5, the maximum allowed re - testing times Nmax = 20, and the maximum beat constraint Tmax = 45 seconds. The controller records the current used time t_used. If t_used+(n×single - detection time)>Tmax, then dynamically adjust n = floor((Tmax - t_used) / single - detection time). For each marginal item, the controller repeatedly applies the same excitation and collects the equivalent hydraulic parameters to obtain a set of samples x1, x2,..., xn. Calculate the sample mean μ=(1 / n)∑xi, and the sample standard deviation s = sqrt(∑(xi - μ) 2 / (n-1)). The confidence level uses a 95% confidence interval: μ ± t_(0.025, n-1)·s / Where t_(0.025, n-1) is the critical value of the t-distribution. If the confidence interval is completely within the acceptable threshold, it is considered acceptable; if it is completely outside the acceptable threshold, it is considered unacceptable; otherwise, the number of retests is increased (n = n+5), and the calculation is repeated until the confidence condition is met or Nmax is reached. If Nmax is reached but the condition is still not met, the current mean μ is compared with the threshold, and it is marked as requiring manual review. This process ensures statistical reliability.

[0127] In a preferred embodiment, the plurality of pressure sensors include a first pressure sensor located at the outlet of the pressure regulating valve 17, a second pressure sensor located in each independent detection air path, and a third pressure sensor located on the side of the ESC wheel cylinder to be tested.

[0128] When the pressure holding condition is steady or the inflation / deflation rate is lower than a preset gradual change threshold, the controller is also configured to:

[0129] Based on the reading difference and reading delay between the first and second pressure sensors, the compressibility compensation coefficient of the gas in the high-pressure gas path is calculated.

[0130] Based on the reading difference and reading delay between the second and third pressure sensors, the pressure decay and response lag time of the gas in the internal flow channel of the ESC are calculated.

[0131] The equivalent hydraulic performance parameters output by the gas-liquid equivalent mapping model are superimposed with the compressibility compensation coefficient and pressure attenuation to obtain the corrected equivalent hydraulic performance parameters.

[0132] If the response delay time exceeds the preset duration, it is determined that there is a blockage in the flow channel or that the valve is slow to operate.

[0133] This embodiment further defines the compensation calculation performed by the controller under steady-state pressure holding conditions or when the inflation / deflation rate is lower than a preset gradual change threshold. The steady-state pressure holding condition is defined as an absolute value of the pressure change rate being less than 0.01 MPa per second and a duration exceeding 2 seconds; the inflation / deflation rate being lower than the gradual change threshold is defined as a mass flow rate being less than 0.5 g per second, which the controller determines by real-time monitoring of the differential value of the first pressure sensor.

[0134] The controller simultaneously reads pressure sensor data from three locations. The first pressure sensor is installed at the outlet of the air pressure regulating valve 17 to monitor the air source pressure; the second pressure sensor is installed in each independent detection air path (after the air distribution block 15 and before the shut-off valve 16) to monitor the branch pressure; and the third pressure sensor is installed on the side of the ESC wheel cylinder under test (between the bottom air outlet of the Z-axis hydraulic cylinder 12 and the inlet of the ESC main cylinder chamber) to monitor the pressure entering the ESC chamber.

[0135] Under steady-state pressure maintenance or gradually changing operating conditions, the controller first calculates the compressibility compensation coefficient of the gas in the high-pressure gas path. Specifically, it continuously collects readings from the first and second pressure sensors, calculates the average pressure difference between them, and simultaneously measures the time delay of the pressure wave propagating from the first sensor to the second sensor using the step response method (e.g., suddenly closing the main intake valve and recording the time difference between the pressure drop at the first sensor and the pressure drop at the second sensor). The compressibility compensation coefficient is calculated based on the pressure difference and time delay, combined with pipeline geometric parameters (6 mm inner diameter, 500 mm length) and empirical formulas, and is used to correct the deviation between upstream and downstream pressures caused by the compressibility of the gas. The specific method for the controller to calculate the compressibility compensation coefficient is as follows: First, the controller continuously collects the readings P1 from the first pressure sensor and P2 from the second pressure sensor, takes the average value over 10 seconds, and calculates the pressure difference ΔP12 = P1_avg - P2_avg. Meanwhile, the time delay Δt12 is measured by the step response method: the controller quickly closes the main intake solenoid valve, and records the time t1 when the pressure of the first pressure sensor begins to decrease and the time t2 when the pressure of the second pressure sensor begins to decrease. Then Δt12 = t2 - t1. Repeat the measurement 3 times and take the average value.

[0136] The compressibility compensation coefficient α is calculated using the following formula: α = (ΔP12×D) / (2×ρ×L×v) 2 ), where D is the inner diameter of the pipe (6 mm), L is the length of the pipe between the first and second sensors (500 mm), ρ is the gas density (calculated from the ideal gas law based on P1 and temperature T), and v is the gas velocity (measured by a mass flow meter or estimated based on valve opening). This formula is derived based on the principle in fluid mechanics that friction loss is proportional to the square of the velocity. To avoid the complexity of online calculations, a mapping table of pressure difference and compressibility compensation coefficient can be established experimentally during the equipment calibration stage and stored in the controller. During actual testing, the controller directly looks up the table or uses linear interpolation based on the measured ΔP12 to obtain α. The calibration method is as follows: Before the equipment leaves the factory, multiple sets of compressed air at different pressures (such as 1 MPa, 2 MPa, 3 MPa...20 MPa) are introduced into the pipe, ΔP12 is measured at each pressure, and the corresponding α value is derived to form a standard mapping table.

[0137] Secondly, the controller calculates the pressure decay and response lag time of the gas in the internal flow channel of the ESC. Specifically: the difference in steady-state readings between the second and third pressure sensors is used as the pressure decay; the time difference between the pressure rise of the second sensor and the pressure rise of the third sensor is measured using a step response (e.g., rapidly opening the main intake solenoid valve), and this time difference is used as the response lag time. This pressure decay reflects the flow resistance loss from the outlet of the shut-off valve 16 through the pneumatic control unit 19, the high-pressure air supply circuit 11, the Z-axis hydraulic cylinder 12 interface, and finally to the ESC main cylinder chamber; the response lag time reflects the inertia of gas propagation and cavity filling in the flow channel.

[0138] Then, the controller superimposes the equivalent hydraulic performance parameters output from the gas-liquid equivalent mapping model with the aforementioned compressibility compensation coefficient and pressure attenuation to obtain the corrected equivalent hydraulic performance parameters. Specifically, for pressure parameters (such as the pressure value corresponding to the equivalent hydraulic leakage), the original equivalent hydraulic pressure is first multiplied by the compressibility compensation coefficient (which is greater than 1), and then the pressure attenuation is subtracted to obtain a pressure closer to the actual pressure acting inside the ESC; for rate parameters (such as the boost rate), a time axis shift correction is performed based on the response lag time. The corrected parameters are then compared with the acceptable threshold.

[0139] Finally, if the response lag time exceeds the preset duration (e.g., 50 milliseconds), the controller determines that there is a blockage in the flow path or that the valve action is sluggish. At this time, the controller performs self-diagnosis: first, it closes the shut-off valve 16 to isolate the gas path from the ESC, and then tests the response lag time of the unloaded pipeline separately. If the response lag time still exceeds the tolerance under no-load conditions, it determines that there is a blockage inside the high-pressure gas path module (e.g., filter blockage, pipeline bend), and outputs an alarm code; if the no-load condition is normal, it determines that there is a problem with the flow path or valve inside the ESC under test (e.g., pressure relief valve not open, pressure booster valve stuck), and outputs a corresponding fault message. This determination result, along with the test data, is uploaded to the MES system.

[0140] The technical principle of this embodiment is to utilize the differential and time delay information from multiple pressure sensors along the flow path to identify the influence of gas compressibility and flow resistance on pressure measurement online and perform physical compensation. Simultaneously, it uses response lag time as a diagnostic indicator of flow path health. The technical effect is to eliminate the influence of the gas path system itself on the detection results, improve the input accuracy of gas-liquid equivalent mapping, and achieve online diagnosis of flow path blockage and sluggish valve action.

[0141] In a preferred embodiment, the controller is further configured to perform a pump capacity test, including:

[0142] The motor inside the ESC under test is started, driving the plunger pump to work;

[0143] During motor startup, the motor's drive current is collected, and pressure change data on the wheel cylinder side is collected through the third pressure sensor.

[0144] The actual pressure build-up rate is calculated based on the pressure change data, and the actual pressure build-up rate is compared with the preset standard pressure build-up rate.

[0145] If the driving current deviates from the preset standard current curve and the actual pressure build-up rate is lower than the standard pressure build-up rate, then the pump body is determined to have a mechanical fault.

[0146] If the driving current is within the preset standard current curve range but the actual pressure build-up rate is lower than the standard pressure build-up rate, it is determined that there is a leak in the high-pressure gas circuit module or the internal flow channel of the ESC under test.

[0147] Specifically, this embodiment further defines the specific steps of the pump capability test. The pump capability test is the last of the eight test items and is used to evaluate the pressure build-up capability of the ESC internal plunger pump and distinguish the type of failure.

[0148] At the start of the test, the controller first controls the air circuit to enter the pump capacity test state according to the timing schedule: the pressure relief valve and the booster valve remain open by default, while the pressure reducing valve and the intake valve remain closed by default. The controller controls the air pressure regulating valve 17 to output a pressure of 5 MPa, opens the main intake solenoid valve to inflate for 2 seconds, so that the pressure in the right front wheel and left front wheel cylinders reaches 5 MPa. After inflation is completed, the controller closes the main intake solenoid valve.

[0149] Subsequently, the controller sends a pulse-width modulation (PWM) signal to the motor drive circuit inside the ESC to start the motor. This signal has a frequency of 10 kHz, a duty cycle that gradually increases from 0 to 100%, and an acceleration time of 200 milliseconds, allowing the plunger pump to smoothly reach its rated speed. Simultaneously, the controller activates the right rear wheel pressure relief valve to open, while the left front wheel pressure relief valve remains closed, resulting in the following gas flow path: right rear wheel pressure relief valve → plunger pump → left front wheel booster valve → left front wheel cylinder.

[0150] During motor startup, the controller collects the motor's drive current through a Hall effect current sensor at a sampling frequency of 2 kHz, recording all current data from motor startup to stable operation, forming a curve of current change over time. The preset standard current curve is the average value obtained by testing 10 qualified ESC products, with the upper and lower envelopes being ±15% of the standard value.

[0151] The controller simultaneously collects pressure change data on the wheel cylinder side via a third pressure sensor, with a sampling frequency of 200 Hz and a duration of 3 seconds. It focuses on collecting the pressure value of the left front wheel cylinder 2 seconds after motor startup and comparing it with the initial pressure before startup (5 MPa) to calculate the actual pressure build-up rate, which is the pressure increase value divided by 2 seconds.

[0152] Based on the drive current curve and the actual voltage build-up rate, the controller performs a joint diagnostic:

[0153] If the drive current deviates from the standard current curve (i.e., the current exceeds the standard value by ±15% at a certain moment), and the actual pressure build-up rate is lower than 90% of the standard pressure build-up rate (the standard pressure build-up rate is 0.105 MPa per second, corresponding to an increase of 0.21 MPa in 2 seconds), then it is determined that there is a mechanical fault in the pump body, which may be due to wear of the plunger pump, jamming of the motor bearing, or damage to the internal seal.

[0154] If the drive current is within the standard current curve range (not exceeding ±15% throughout), but the actual pressure build-up rate is lower than 90% of the standard pressure build-up rate, then a leak is determined to exist in the high-pressure air circuit module or the internal flow channel of the ESC under test. The leak may occur in the pipeline between the right rear wheel pressure reducing valve and the plunger pump, the pipeline between the plunger pump and the left front wheel booster valve, or due to poor valve sealing.

[0155] If the drive current deviates but the voltage build-up rate is normal, it is determined that the current sensor is abnormal or the motor characteristics are abnormal, and a warning is output.

[0156] If both are normal, the pump capacity test is deemed qualified.

[0157] The controller stores the diagnostic results in a test report and displays the specific fault codes via touchscreen 2, while simultaneously uploading them to the MES system. The technical principle of this embodiment is to distinguish between pump mechanical faults and system leakage faults through joint analysis of motor current and pressure build-up rate. The technical effect is to achieve accurate assessment of pump capacity and fault location, avoiding the blindness of judging based on a single parameter.

[0158] In a preferred embodiment, the controller is further configured to perform a high-pressure gas path thermodynamic transient correction step, which includes:

[0159] During the process of the high-pressure gas circuit module filling the internal cavity of the ESC under test with high-pressure gas or venting high-pressure gas from the internal cavity of the ESC under test, the transient pressure change curve of the pressure sensor from the start of filling or venting to the pressure stabilization time is recorded.

[0160] Based on the preset gas adiabatic process equation, combined with the fixed volume of the high-pressure gas circuit module and the equivalent volume of the internal cavity of the ESC under test, the instantaneous temperature change of the gas during the charging and discharging process is calculated in reverse from the pressure overshoot and pressure recovery time constant in the transient pressure change curve.

[0161] If the calculated instantaneous temperature change does not exceed the preset thermodynamic safety threshold, then the pressure data collected in real time by the pressure sensor is thermodynamically compensated and corrected according to the instantaneous temperature change, and the equivalent isothermal pressure value is calculated. This equivalent isothermal pressure value is then used to replace the original collected pressure data and is sent into the gas-liquid equivalent mapping model.

[0162] If the calculated instantaneous temperature change exceeds the thermodynamic safety threshold, it is determined that there is an abnormal thermal effect in the current test, the inflation or deflation rate is reduced, and the current test procedure is repeated.

[0163] Specifically, this embodiment further defines a high-pressure gas path thermodynamic transient correction step. This correction step is performed during rapid gas charging and discharging, specifically triggered when the charging and discharging rate exceeds 0.5 MPa per second (determined by the rate of change of the first pressure sensor). This correction is used to eliminate the influence of temperature changes caused by adiabatic compression or expansion of the gas on pressure measurement.

[0164] When the controller determines that thermodynamic transient correction is required, during the inflation or deflation process, the transient pressure change curve of the pressure sensor to be corrected is recorded at a sampling rate of 1000 Hz from the start of inflation or deflation to the pressure stabilization point. The start of inflation is the moment when the main intake solenoid valve opens, and the start of deflation is the moment when the exhaust valve opens; the pressure stabilization point is defined as the moment when the pressure change at 10 consecutive sampling points is less than 0.01 MPa.

[0165] The controller, based on a preset gas adiabatic process equation and considering the fixed volume of the high-pressure gas path module and the equivalent volume of the internal cavity of the ESC under test, inversely calculates the instantaneous temperature change of the gas during the charging and discharging process using the pressure overshoot and pressure recovery time constant from the transient pressure change curve. Specifically, it extracts the initial peak value and the stable pressure value from the pressure curve, calculating the difference as the pressure overshoot; it measures the time required for the pressure to drop from the peak value to the peak value minus 0.632 times the overshoot, using this as the pressure recovery time constant (where 0.632 is derived from the definition of the time constant corresponding to 63.2% of the change from the initial value to the steady-state value in the step response of a first-order system, and this time constant characterizes the speed of pressure recovery). The pressure overshoot is mainly caused by the temperature rise due to adiabatic compression of the gas and is proportional to the instantaneous temperature change. The controller calculates the instantaneous temperature rise during the charging process or the instantaneous temperature drop during the discharging process based on the ambient temperature (measured by a temperature sensor inside the chassis) and the stable pressure, according to thermodynamic relationships.

[0166] The controller has a preset thermodynamic safety threshold, such as 30 Kelvin. If the calculated instantaneous temperature change does not exceed this threshold, compensation correction is performed: based on the instantaneous temperature change and the pressure recovery time constant, assuming that the temperature change decays exponentially over time, the raw air pressure data at each sampling point is multiplied by a temperature correction factor. This factor is equal to the ambient temperature divided by the ambient temperature plus the instantaneous temperature change at the current moment, to calculate the equivalent isothermal pressure value. This equivalent isothermal pressure value replaces the original collected air pressure data and is then fed into the gas-liquid equivalent mapping model for further processing.

[0167] If the calculated instantaneous temperature change exceeds the thermodynamic safety threshold, an abnormal thermal effect is detected. This could be due to excessively rapid inflation or deflation rates, potentially causing seal damage or a risk of gas explosion. In this case, the controller automatically reduces the inflation or deflation rate (e.g., extending the pressurization time of the pressure regulating valve 17 from 0.5 seconds to 2 seconds) and re-executes the current detection procedure. During re-execution, the controller monitors the temperature change again until it meets the safety threshold or the cumulative retry count reaches the upper limit, at which point an error is reported.

[0168] The technical principle of this embodiment is to use the pressure overshoot of the adiabatic process to infer the temperature change and perform isothermal correction to eliminate the interference of thermodynamic transient effects on pressure measurement. This correction step is applicable to rapid charging and discharging conditions such as pressurization rate testing, depressurization rate testing, pressurization valve opening and closing testing, and depressurization valve testing in the eight test items. It has been mentioned in the previous section "Thermodynamic Correction (if applicable)", and a complete implementation is given here.

[0169] In a preferred embodiment, the controller is further configured to:

[0170] Obtain the operating parameters of the current detection process, including the inflation / deflation rate, the upstream / downstream pressure ratio, and the gas flow rate of each pressure sensor in the detection stage.

[0171] The installation positions of each pressure sensor are obtained, including a first position at the outlet of the pressure regulating valve 17, a second position in each independent detection air path, and a third position on the side of the ESC wheel cylinder to be tested.

[0172] Based on the operating parameters and installation location, each pressure sensor is divided into:

[0173] Type 1: No calibration required. Suitable for all pressure sensors in steady-state pressure holding or slowly changing conditions, or pressure sensors in the first position in non-steady-state conditions, or pressure sensors in the first and second positions in high-speed throttling conditions.

[0174] Type 2: Requires thermodynamic transient correction and is suitable for pressure sensors at the second and third positions in rapid charge and discharge transient conditions.

[0175] Specifically, for pressure sensor readings classified as the first type, the controller directly inputs them into the gas-liquid equivalent mapping model; for pressure sensor readings classified as the second type, the high-pressure gas path thermodynamic transient correction step is invoked for processing, and the equivalent isothermal pressure value obtained after correction is input into the gas-liquid equivalent mapping model.

[0176] Specifically, the controller first acquires the operating parameters of the current detection process. These parameters include the charging / discharging rate, the upstream / downstream pressure ratio, and the gas flow rate at each pressure sensor's detection stage. The charging / discharging rate is obtained by differentiating the reading of the first pressure sensor; a rate greater than 0.5 MPa per second is considered a rapid charging / discharging condition. The upstream / downstream pressure ratio refers to the ratio of the upstream pressure (taken from the second pressure sensor reading) to the downstream pressure (taken from the third pressure sensor reading) for the third pressure sensor. The gas flow rate is directly measured by a mass flow meter installed at the outlet of the pressure regulating valve 17.

[0177] The controller simultaneously acquires the installation position of each pressure sensor. Each sensor channel has been calibrated at the factory, and the channel number is associated with its physical location and stored in the controller's non-volatile memory. The first position is the outlet of the pressure regulating valve 17; the second position is in each independent detection air path (after the air distribution block 15 and before the shut-off valve 16); and the third position is on the side of the ESC wheel cylinder under test (between the bottom air outlet of the Z-axis hydraulic cylinder 12 and the inlet of the ESC main cylinder chamber).

[0178] Based on operating parameters and installation location, the controller classifies the readings of each pressure sensor into two types:

[0179] Type 1: No calibration required. Applicable to any of the following situations: The current process is a steady-state pressure holding or slowly changing condition (gas charging / discharging rate is less than 0.01 MPa / s and the duration exceeds 2 seconds), in which case all position sensors do not require calibration; or the current process is a non-steady-state condition but the sensor is in the first position (because the first position is after the pressure regulating valve 17, and the airflow has stabilized); or the current process is a high-speed throttling condition (upstream and downstream pressure ratio is less than 0.528) but the sensor is in the first or second position (because the upstream pressure at these positions is less affected by throttling).

[0180] Type 2: Requires thermodynamic transient correction. Suitable for rapid charge / discharge transient conditions (charge / discharge rate higher than 0.5 MPa / s and upstream / downstream pressure ratio not less than 0.528), with the sensor located in the second or third position. Because sensors in these positions directly sense rapid pressure changes, temperature changes caused by adiabatic compression or expansion of the gas will significantly affect the pressure reading, requiring correction.

[0181] The controller performs this classification in real time after each sampling. For pressure sensor readings classified as type 1, the controller directly feeds them into the gas-liquid equivalent mapping model without any correction. For pressure sensor readings classified as type 2, the controller invokes the previously described high-pressure gas path thermodynamic transient correction step to convert the reading into an equivalent isothermal pressure value before feeding it into the gas-liquid equivalent mapping model.

[0182] The technical principle of this embodiment is to automatically determine whether thermodynamic compensation is needed based on the sensor location and the physical characteristics of the operating conditions. The technical effect is to improve the targeting and efficiency of data processing, avoiding the computational overhead or insufficient compensation caused by blindly correcting all data.

[0183] In a preferred embodiment, the controller is further configured to:

[0184] After the high-pressure gas circuit module fills the internal cavity of the ESC under test with high-pressure gas and reaches a stable pressure, the shut-off valve 16 located upstream of the ESC under test is controlled to close instantly, and the pressure decay curve of the pressure sensor within a preset time window after the shut-off valve 16 is closed is recorded.

[0185] Based on the initial slope of the pressure decay curve, the equivalent volume of the internal cavity of the ESC under test is deduced by looking up a table. The mapping relationship used for the table lookup is pre-established and stored in the controller through calibration experiments.

[0186] Specifically, before performing thermodynamic transient correction, the controller first executes the identification step of this embodiment. The specific operation is as follows:

[0187] First, the controller controls the air pressure regulating valve 17 to output a pressure of 5 MPa, opening the main intake solenoid valve to fill the internal cavities of the ESC under test (including MC1, MC2, and the connected wheel cylinder cavities) with air until the reading of the third pressure sensor stabilizes at 5 MPa, with fluctuations less than 0.01 MPa per second. At this point, the internal cavities of the ESC under test are filled with high-pressure gas, and the controller records the stable pressure P_stable and the ambient temperature T (measured by the temperature sensor inside the chassis).

[0188] Then, the controller controls the shut-off valve 16 located upstream of the ESC under test to close instantaneously. This shut-off valve 16 is a pneumatic angle seat valve with a closing time of less than 20 milliseconds. After the shut-off valve 16 closes, the internal cavity of the ESC under test is isolated from the upstream gas source, forming a closed volume. At this time, due to minor leaks in the internal cavity of the ESC under test (such as valve sealing surfaces, piston clearance, etc.), or due to slight diffusion of gas to the pipe wall, the reading of the third pressure sensor will slowly decrease over time. The controller records the pressure decay curve P(t) within 20 seconds after the shut-off valve 16 closes at a sampling rate of 100 Hz.

[0189] The controller estimates the equivalent volume of the internal cavity of the ESC under test based on the initial slope of the pressure decay curve. The initial slope is defined as the rate of pressure drop within the first 5 seconds after the shut-off valve 16 closes, calculated as: Slope S = (P_stable - P(5 seconds)) / 5, in megapascals per second. This slope reflects the rate of pressure decay: under the same leakage conditions, the larger the volume, the slower the pressure drop (the smaller the absolute value of the slope); the smaller the volume, the faster the pressure drop (the larger the absolute value of the slope).

[0190] To convert the slope to volume, the equipment needs to be calibrated before leaving the factory. The calibration method is as follows: Prepare a set of standard tanks with known volumes, such as 50 cubic centimeters, 100 cubic centimeters, 150 cubic centimeters, and 200 cubic centimeters. Connect each standard tank to the equipment sequentially (replacing the position of the ESC to be measured), and follow the same operating steps (inflate to 5 MPa, close shut-off valve 16, record the pressure decay curve), measuring the pressure drop slope for the first 5 seconds for each tank. Create a mapping table between the volume and the corresponding slope value and store it in the controller. During calibration, corrections should also be made for different ambient temperatures, such as calibrating at 20 degrees Celsius, 30 degrees Celsius, and 40 degrees Celsius respectively, forming a three-dimensional mapping table of temperature-volume-slope.

[0191] In actual testing, the controller, based on the current ambient temperature, searches the mapping table for the two calibration points closest to the measured slope S, and calculates the equivalent volume V_esc of the internal cavity of the ESC under test through linear interpolation. The linear interpolation formula is: V_esc = V_low + (V_high - V_low) × (S - S_low) / (S_high - S_low), where S_low and S_high are the two calibration slopes adjacent to the measured slope in the mapping table, and V_low and V_high are the corresponding volumes.

[0192] The controller replaces the previously used equivalent volume of the internal cavity of the ESC under test with the derived equivalent volume V_esc, which is used for temperature estimation and compensation correction in the subsequent thermodynamic transient correction step. At the same time, the controller compares V_esc with the preset nominal equivalent volume (from the design drawings, for example, 150 cubic centimeters). If the deviation exceeds ±20%, it is determined that the current ESC under test has an abnormal cavity volume, such as due to manufacturing errors causing deviations in flow channel dimensions or residual foreign objects inside. An alarm code is output and the detection is terminated.

[0193] In a preferred embodiment, the gas-liquid equivalent mapping model is pre-built and stored in the controller through the following steps:

[0194] Based on the 3D CAD model of the ESC hydraulic actuator to be tested, a computational fluid dynamics simulation model is constructed; the physical property parameters of the pneumatic medium and the hydraulic medium are set respectively, and transient simulation is performed under the same boundary conditions to obtain the pressure response curves at the corresponding detection points under pneumatic and hydraulic conditions.

[0195] The deviation between the gas pressure simulation results and the hydraulic simulation results is analyzed, and a gas compressibility compensation function and a cavity compliance compensation function are established as physical constraint priors for the gas-liquid equivalent mapping model.

[0196] Synchronous experimental data of multiple ESC products of the same model on pneumatic and hydraulic testing equipment were collected to form a pneumatic-hydraulic feature dataset.

[0197] Using the physical constraint prior as a regularization term and the gas-hydraulic feature dataset as training samples, a mapping transformation function with physical constraints is trained; the input of the mapping transformation function is the gas pressure detection feature parameters, and the output is the equivalent hydraulic performance parameters.

[0198] The trained mapping transformation function is stored in the controller as the gas-liquid equivalent mapping model.

[0199] Specifically, this embodiment describes in detail the pre-construction steps of the gas-liquid equivalent mapping model, which is stored in the controller and used to convert the gas pressure detection characteristic parameters into equivalent hydraulic performance parameters.

[0200] The first step involved constructing a computational fluid dynamics (CFD) simulation model based on the 3D CAD model of the ESC hydraulic actuator to be tested. Using ANSYS Fluent or similar software, the CAD model was meshed using tetrahedral meshes, with localized refinement at valve orifices, abrupt changes in flow channels, and sealing gaps. The total mesh size was approximately 5 million. Two media were used: a pneumatic medium employing an ideal gas model, with density following the ideal gas law and a viscosity of 1.79 × 10⁻⁶. -5 Pascal-second; the hydraulic medium is based on a brake fluid model, with a density of 850 kg / m³ and a viscosity of 1.5 × 10⁻⁶. - ³ Pascal-second. Set the same boundary conditions: inlet pressure 5 MPa, outlet pressure atmospheric, initial temperature 20°C. Perform transient simulation with a time step of 0.001 seconds and a total duration of 5 seconds. Extract pressure response curves for corresponding detection points (e.g., the wheel cylinder side) under both pneumatic and hydraulic conditions to obtain multiple sets (e.g., 100 sets) of pressure-time curve pairs under different valve conditions.

[0201] The second step is to analyze the deviation between the gas pressure simulation results and the hydraulic simulation results. By comparing the gas pressure value P_gas(t) and the hydraulic pressure value P_oil(t) at each moment, the deviation ΔP(t) = P_gas(t) - P_oil(t) is calculated. Analysis reveals that the deviation is mainly related to gas compressibility and cavity compliance. The gas compressibility compensation function is defined as f_comp(ΔP, V) = (ΔP / P_ref) × (V_ref / V), where P_ref is the reference pressure (taken as 5 MPa), V_ref is the reference volume (taken as 100 cubic centimeters), and V is the current cavity volume. The cavity compliance compensation function is defined as f_compliance(ΔP, A) = ΔP × A / K, where A is the cavity wall area, and K is the equivalent stiffness (calibrated through simulation or experiment). These two functions are used as physical constraint priors for regularization in subsequent model training.

[0202] The third step involves collecting synchronous experimental data from multiple ESC products of the same model on both pneumatic and hydraulic testing equipment. One hundred qualified ESC products from the same batch were selected. First, a full range of tests were performed on a traditional hydraulic testing bench, recording the hydraulic performance parameters for each product, including leakage (mL / h), pressurization rate (MPa / s), depressurization rate (MPa / s), and valve opening pressure (MPa). Then, the same products were tested on the pneumatic testing equipment of this invention, recording pneumatic characteristic parameters, including pressure decay rate during pressure holding, pressure rise slope during pressurization, time constant during depressurization, and pressure surge amplitude during valve opening. The two sets of data were then paired to form a pneumatic-hydraulic feature dataset, with each sample containing six pneumatic features and four hydraulic labels.

[0203] The fourth step involves training a physically constrained mapping transformation function using physical constraint priors as the regularization term and the gas-hydraulic feature dataset as the training samples. A three-layer feedforward neural network is employed, with 6 nodes in the input layer, 12 nodes in the hidden layer, and 4 nodes in the output layer. The loss function consists of two parts: mean squared error and a regularization term. The mean squared error measures the difference between the model output and the true hydraulic label; the regularization term is the penalty for violating physical constraints, i.e., the sum of the absolute values ​​of the deviations between the equivalent hydraulic parameters output by the model and the theoretical values ​​after substituting them into the compressibility compensation function and the cavity compliance compensation function. The regularization coefficient is set to 0.1. TensorFlow or PyTorch frameworks are used, with Adam as the optimizer, a learning rate of 0.001, a batch size of 32, and 500 training epochs. During training, the model learns both the statistical regularities in the data and is guided by physical constraints, ensuring that the output conforms to the basic physical relationship between gas compressibility and cavity compliance.

[0204] The fifth step is to permanently store the trained mapping transformation function in the controller. After training, the weight matrix and bias vector of the neural network are exported as C language header files or binary files and burned into the controller's flash memory. When the controller is powered on, these parameters are loaded into memory. During inference, the air pressure feature vector is input, forward propagation calculations (matrix multiplication and activation functions) are performed, and the equivalent hydraulic performance parameters are output.

[0205] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A high-pressure detection device for an ESC system based on the gas-liquid equivalent detection principle, characterized in that, include: Rack (1); Positioning and clamping sealing module, used to fix and seal the ESC to be tested; The high-pressure gas circuit module is used to input high-pressure gas into the ESC under test; the high-pressure gas circuit module includes multiple pressure sensors for real-time acquisition of gas pressure data at each monitoring point during the detection process; The controller stores a gas-liquid equivalent mapping model, which is trained using synchronous experimental data of gas pressure and hydraulic pressure of the same ESC product to train the mapping transformation function, and constructed using the coupling relationship between gas compressibility and cavity compliance analyzed by computational fluid dynamics simulation as a constraint condition. The controller is configured to perform the following steps: The positioning, clamping, and sealing module is controlled to fix and seal the ESC to be tested; The high-pressure gas circuit module is controlled to input high-pressure gas into the ESC under test, and the solenoid valves and motors inside the ESC under test are controlled to operate according to a preset timing sequence. Acquire air pressure data collected in real time by multiple pressure sensors; The air pressure data is input into the gas-liquid equivalent mapping model and converted into equivalent hydraulic performance parameters. The equivalent hydraulic performance parameters are compared with preset thresholds, and the detection results are output. The gas-liquid equivalent mapping model is pre-built and stored in the controller through the following steps: Based on the 3D CAD model of the ESC hydraulic actuator to be tested, a computational fluid dynamics simulation model is constructed; the physical property parameters of the pneumatic medium and the hydraulic medium are set respectively, and transient simulation is performed under the same boundary conditions to obtain the pressure response curves at the corresponding detection points under pneumatic and hydraulic conditions. The deviation between the gas pressure simulation results and the hydraulic simulation results is analyzed, and a gas compressibility compensation function and a cavity compliance compensation function are established as physical constraint priors for the gas-liquid equivalent mapping model. Synchronous experimental data of multiple ESC products of the same model on pneumatic and hydraulic testing equipment were collected to form a pneumatic-hydraulic feature dataset. Using the physical constraint prior as a regularization term and the gas-hydraulic feature dataset as training samples, a mapping transformation function with physical constraints is trained; the input of the mapping transformation function is the gas pressure detection feature parameters, and the output is the equivalent hydraulic performance parameters. The trained mapping transformation function is stored in the controller as the gas-liquid equivalent mapping model.

2. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 1, characterized in that, The controller employs a minimum excitation set strategy to execute the control of the internal solenoid valve of the ESC under test according to a preset timing sequence. The minimum incentive set strategy includes: The various detection functions of the ESC under test are decoupled, and the optimal combination of pressure excitation and the optimal combination of flow excitation are selected. The process involves three phases: rapid screening, targeted verification, and gray zone retesting. During the rapid screening phase, the optimal pressure excitation combination and the optimal flow excitation combination are applied, air pressure data at each monitoring point are collected and compared with the rapid screening threshold, and the detection items are divided into three categories: qualified, questionable, and boundary. During the targeted verification phase, only testing items deemed questionable are subject to targeted incentives for confirmatory testing. During the gray zone retesting phase, the detection items determined to be boundaries are repeatedly stimulated and sampled multiple times. The statistical confidence level is calculated based on the sampling results. When the confidence level is lower than the preset threshold, the number of retests is extended until the threshold is reached or the maximum cycle time constraint is exceeded.

3. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 1, characterized in that, The plurality of pressure sensors include a first pressure sensor located at the outlet of the pressure regulating valve, a second pressure sensor located in each independent detection air path, and a third pressure sensor located on the side of the ESC wheel cylinder to be tested. When the pressure holding condition is steady or the inflation / deflation rate is lower than a preset gradual change threshold, the controller is also configured to: Based on the reading difference and reading delay between the first and second pressure sensors, the compressibility compensation coefficient of the gas in the high-pressure gas path is calculated. Based on the reading difference and reading delay between the second and third pressure sensors, the pressure decay and response lag time of the gas in the internal flow channel of the ESC are calculated. The equivalent hydraulic performance parameters output by the gas-liquid equivalent mapping model are superimposed with the compressibility compensation coefficient and pressure attenuation to obtain the corrected equivalent hydraulic performance parameters. If the response delay time exceeds the preset duration, it is determined that there is a blockage in the flow channel or that the valve is slow to operate.

4. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 3, characterized in that, The controller is also configured to perform a pump capacity test, including: The motor inside the ESC under test is started, driving the plunger pump to work; During motor startup, the motor's drive current is collected, and pressure change data on the wheel cylinder side is collected through the third pressure sensor. The actual pressure build-up rate is calculated based on the pressure change data, and the actual pressure build-up rate is compared with the preset standard pressure build-up rate. If the driving current deviates from the preset standard current curve and the actual pressure build-up rate is lower than the standard pressure build-up rate, then the pump body is determined to have a mechanical fault. If the driving current is within the preset standard current curve range but the actual pressure build-up rate is lower than the standard pressure build-up rate, it is determined that there is a leak in the high-pressure gas circuit module or the internal flow channel of the ESC under test.

5. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 1, characterized in that, The controller is also configured to perform a high-pressure gas path thermodynamic transient correction step, which includes: During the process of the high-pressure gas circuit module filling the internal cavity of the ESC under test with high-pressure gas or venting high-pressure gas from the internal cavity of the ESC under test, the transient pressure change curve of the pressure sensor from the start of filling or venting to the pressure stabilization time is recorded. Based on the preset gas adiabatic process equation, combined with the fixed volume of the high-pressure gas circuit module and the equivalent volume of the internal cavity of the ESC under test, the instantaneous temperature change of the gas during the charging and discharging process is calculated in reverse from the pressure overshoot and pressure recovery time constant in the transient pressure change curve. If the calculated instantaneous temperature change does not exceed the preset thermodynamic safety threshold, then the pressure data collected in real time by the pressure sensor is thermodynamically compensated and corrected according to the instantaneous temperature change, and the equivalent isothermal pressure value is calculated. This equivalent isothermal pressure value is then used to replace the original collected pressure data and is sent into the gas-liquid equivalent mapping model. If the calculated instantaneous temperature change exceeds the thermodynamic safety threshold, it is determined that there is an abnormal thermal effect in the current test, the inflation or deflation rate is reduced, and the current test procedure is repeated.

6. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 5, characterized in that, The controller is also configured to: Obtain the operating parameters of the current detection process, including the inflation / deflation rate, the upstream / downstream pressure ratio, and the gas flow rate of each pressure sensor in the detection stage. The installation positions of each pressure sensor are obtained, including a first position at the outlet of the pressure regulating valve, a second position in each independent detection air path, and a third position on the side of the ESC wheel cylinder to be tested. Based on the operating parameters and installation location, each pressure sensor is divided into: Type 1: No calibration required. Suitable for all pressure sensors in steady-state pressure holding or slowly changing conditions, or pressure sensors in the first position in non-steady-state conditions, or pressure sensors in the first and second positions in high-speed throttling conditions. Type 2: Requires thermodynamic transient correction and is suitable for pressure sensors at the second and third positions in rapid charge and discharge transient conditions. Specifically, for pressure sensor readings classified as the first type, the controller directly inputs them into the gas-liquid equivalent mapping model; for pressure sensor readings classified as the second type, the high-pressure gas path thermodynamic transient correction step is invoked for processing, and the equivalent isothermal pressure value obtained after correction is input into the gas-liquid equivalent mapping model.

7. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 5, characterized in that, The controller is also configured to: After the high-pressure gas circuit module fills the internal cavity of the ESC under test with high-pressure gas and reaches a stable pressure, the shut-off valve located upstream of the ESC under test is controlled to close instantly, and the pressure decay curve of the pressure sensor within a preset time window after the shut-off valve closes is recorded. Based on the initial slope of the pressure decay curve, the equivalent volume of the internal cavity of the ESC under test is deduced by looking up a table. The mapping relationship used for the table lookup is pre-established and stored in the controller through calibration experiments.

8. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 1, characterized in that, It also includes a whisker-type micro switch (3) and a safety light curtain; The controller is also configured to: When the triggering of the touch-type micro switch (3) is detected, the positioning and pressing sealing module is controlled to start operating, and the safety light curtain is activated simultaneously to form a safety protection light curtain; When an obstruction signal is received from the safety light curtain due to obstruction, an emergency stop is triggered.

9. The high-pressure detection equipment for the ESC system based on the gas-liquid equivalent detection principle according to claim 1, characterized in that, The positioning, pressing, and sealing module includes an X-axis hydraulic cylinder (4) and a Z-axis hydraulic cylinder (12). The X-axis hydraulic cylinder (4) is used to extend and retract along the X-axis direction to achieve X-axis positioning of the ESC under test. The Z-axis hydraulic cylinder (12) is used to extend and retract along the Z-axis direction to achieve Z-axis pressing of the ESC under test. The bottom of the Z-axis hydraulic cylinder (12) is provided with two air outlets. The two air outlets are respectively connected to the two main cylinder inlets of the ESC under test and are connected to the high-pressure air circuit module.

Images