A pressure reducing valve with reverse flow conduction and pressure stabilization switching functions

Abstract

The present application relates to a kind of pressure reducing valve with counterflow conduction and pressure stabilizing switching function, it relates to pressure reducing valve technical field, including: valve body, upstream inlet cavity, downstream outlet cavity, main pressure reducing flow channel and counterflow bypass flow channel are arranged in valve body, main pressure reducing flow channel is communicated with upstream inlet cavity and downstream outlet cavity, counterflow bypass flow channel is communicated with downstream outlet cavity and upstream inlet cavity;Realized when downstream pressure is abnormally higher than upstream, automatically open counterflow bypass flow channel and carry out controlled backflow pressure relief, and in the pressure relief process, pressure impact is gradually attenuated by gradient lattice structure, and downstream pressure is quickly converged to target value by continuously adjusting counterflow valve clack opening degree by sliding mode variable structure control, the problem that traditional pressure reducing valve cannot realize controlled counterflow conduction and active pressure stabilization under counterpressure condition is solved, with fast response speed, pressure pulsation suppression effect is good, control precision is high and the advantages of reliable operation.

Description

Technical Field

[0001] This disclosure relates to, but is not limited to, the field of pressure reducing valve technology, and in particular to a pressure reducing valve with reverse flow conduction and pressure stabilization switching functions. Background Technology

[0002] Pressure reducing valves are core regulating components in fluid pipeline systems used to reduce upstream high-pressure media to a set downstream pressure. They are widely used in industrial hydraulics, urban water supply, chemical transportation, and aerospace propulsion. Traditional pressure reducing valves typically employ spring-loaded or pilot-operated structures, relying on the force balance between the spring preload and the medium pressure to achieve unidirectional pressure reduction. Under normal operating conditions, the upstream pressure is always higher than the downstream pressure, and the medium flows in the forward direction, being throttled and depressurized by the valve core before being output to the downstream pipeline. However, in actual engineering projects, the downstream pipeline may experience a sudden and abnormal increase in downstream pressure, even exceeding the upstream pressure, due to factors such as sudden load unloading, accumulator release, pump station switching, or valve misoperation. Existing pressure reducing valves, facing such back pressure conditions, cannot provide a controlled reverse flow channel for the downstream high-pressure medium because their structure only provides unidirectional pressure reduction. This leads to downstream overpressure retention and water hammer impact in the pipeline, which can severely cause pipeline vibration, seal failure, or even equipment damage. Currently, a common engineering solution is to install an independent safety valve or relief valve in parallel downstream of the pressure reducing valve for passive pressure relief. However, this method has drawbacks such as slow response, uncontrollable pressure relief process, and complex system piping. Some improved solutions use a combination of an electromagnetic pilot valve and a pressure reducing valve to achieve reverse flow bypass, but the response bandwidth of the electromagnetic pilot valve is limited by the inductance time constant of the electromagnet, making it difficult to complete precise opening adjustment within a millisecond timescale. In addition, existing pressure reducing valves lack an effective damping and stabilizing mechanism during reverse flow pressure relief. When the high-pressure medium flows rapidly from downstream to upstream, it is prone to generating severe pressure pulsations and flow velocity oscillations, causing secondary impacts on the upstream pipeline. Therefore, how to integrate a reverse flow conduction mechanism with both fast response driving capability and gradient damping energy dissipation function into the pressure reducing valve body, and achieve active pressure stabilization and convergence of downstream pressure during reverse flow through a closed-loop control algorithm, is a problem that urgently needs to be solved in the field of pressure reducing valve technology. Summary of the Invention

[0003] This disclosure provides a pressure reducing valve with reverse flow conduction and pressure stabilization switching functions. It realizes the automatic opening of the reverse flow bypass channel for controlled backflow pressure relief when the downstream pressure is abnormally higher than the upstream pressure. During the pressure relief process, the pressure impact is gradually attenuated by the gradient lattice structure and the opening degree of the reverse flow valve is continuously adjusted by the sliding mode variable structure control to make the downstream pressure quickly converge to the target value. It solves the problem that traditional pressure reducing valves cannot achieve controlled reverse flow conduction and active pressure stabilization under back pressure conditions. It has the advantages of fast response speed, good pressure pulsation suppression effect, high control accuracy and reliable operation.

[0004] To solve the above problems, the technical solution of the present invention is implemented as follows: A pressure reducing valve with reverse flow conduction and pressure stabilization switching function includes: a valve body, wherein an upstream inlet cavity, a downstream outlet cavity, a main pressure reducing flow channel and a reverse flow bypass flow channel are provided in the valve body, the main pressure reducing flow channel connects the upstream inlet cavity and the downstream outlet cavity, and the reverse flow bypass flow channel connects the downstream outlet cavity and the upstream inlet cavity. The main valve core assembly is located in the main pressure reducing channel and is used to reduce the pressure of the medium flowing through the main pressure reducing channel in the forward direction. The magnetostrictive drive assembly, installed on the countercurrent bypass channel, includes an outer shell, a solenoid excitation coil, a permanent magnet biasing magnetic circuit, and a Terfenol-D magnetostrictive rod. The permanent magnet biasing magnetic circuit and the Terfenol-D magnetostrictive rod cooperate to form a biasing magnetic circuit closed along the axial direction of the Terfenol-D magnetostrictive rod. After receiving the driving current, the solenoid excitation coil generates a variable excitation magnetic field superimposed inside the Terfenol-D magnetostrictive rod. Under the action of the combined magnetic field of the biasing magnetic field and the variable excitation magnetic field, the Terfenol-D magnetostrictive rod elongates along the axial direction. The lifting counterflow valve disc is driven to the output end of the Terfenol-D super magnetostrictive rod and cooperates with the counterflow valve seat set in the counterflow bypass channel. Under the push of axial elongation displacement, it disengages from the counterflow valve seat to open the counterflow bypass channel. The damping pressure stabilizing cavity is set on the counterflow bypass channel and located between the downstream outlet and the upstream inlet. The interior of the damping pressure stabilizing cavity is set with a gradient density lattice structure with increasing porosity along the medium return direction. The pressure sensor group includes a first pressure sensor, a second pressure sensor, and a third pressure sensor respectively disposed in the upstream inlet cavity, the downstream outlet cavity, and the damping pressure stabilizing cavity; The controller receives pressure signals from the first, second, and third pressure sensors. It determines the reverse flow conduction condition based on the difference between the downstream and upstream pressure values ​​and uses a sliding mode variable structure control algorithm to generate a drive current command, which is output to the solenoid excitation coil to adjust the opening of the lift-type reverse flow valve, thereby achieving stable convergence of the downstream pressure to the preset downstream target pressure value.

[0005] Furthermore, the permanent magnet bias magnetic circuit consists of permanent magnet rings and magnetically conductive soft iron rings arranged alternately along the axial direction of the Terfenol-D magnetostrictive rod, as well as a cylindrical magnetically conductive outer yoke surrounding the outside. The permanent magnet rings, magnetically conductive soft iron rings, and cylindrical magnetically conductive outer yoke together form a closed bias magnetic circuit along the axial direction of the Terfenol-D magnetostrictive rod; the solenoid excitation coil is sleeved in the annular space between the outer side of the permanent magnet bias magnetic circuit and the inner side of the cylindrical magnetically conductive outer yoke.

[0006] Furthermore, the top of the Terfenol-D magnetostrictive rod maintains an axial pre-compression load through a disc spring pre-compression assembly to eliminate the magnetic hysteresis dead zone, and the output end of the disc spring pre-compression assembly maintains spherical contact with the valve stem end face of the lifting counterflow valve disc through a spherical force transmission pad.

[0007] Furthermore, the damping and stabilizing cavity is an integral component formed by selective laser melting process using 316L stainless steel powder.

[0008] Furthermore, the gradient density lattice structure, along the medium recirculation direction, sequentially includes a high-density lattice section at the inlet, a transitional lattice section, and a low-density lattice section at the outlet. All three sections utilize body-centered cubic lattice units. The high-density lattice section at the inlet has a lattice unit side length of 2 mm, a lattice rod diameter of 0.8 mm, and a porosity of 40%. The transitional lattice section has a lattice unit side length of 3 mm, a lattice rod diameter of 0.6 mm, and a porosity of 58%. The low-density lattice section at the outlet has a lattice unit side length of 4 mm, a lattice rod diameter of 0.4 mm, and a porosity of 75%. As the medium flows sequentially through the high-density lattice section at the inlet, the intermediate transition lattice section, and the low-density lattice section at the outlet, the dense bars in the high-density lattice section at the inlet force the return medium to repeatedly split and merge, thereby generating microscale vortices in the wake region of the bars. This converts the kinetic energy of the pressure impact into turbulent heat dissipation energy to attenuate the pressure peak. In the intermediate transition lattice section, the spacing between the bars gradually increases, causing the microscale vortices to gradually merge and attenuate, further reducing the pressure pulsation amplitude of the return medium. The low-density lattice section at the outlet provides a large pore flow area, allowing the return medium to flow out smoothly at a low velocity and merge into the upstream inlet cavity through the return channel.

[0009] Furthermore, the first, second, and third pressure sensors continuously collect upstream pressure values, downstream pressure values, and damping stabilizing chamber pressure values ​​at a sampling frequency of not less than 1 kHz and send them to the controller. Within each sampling cycle, the controller performs sliding average filtering on the downstream and upstream pressure values ​​for several consecutive sampling cycles. The filtered downstream pressure value is subtracted from the filtered upstream pressure value to obtain the reverse pressure difference value. The reverse pressure difference value of the current sampling cycle is subtracted from the reverse pressure difference value of the previous sampling cycle and then divided by the sampling cycle to obtain the reverse pressure difference change rate. When the reverse pressure difference value is greater than the preset reverse flow entry threshold and the reverse pressure difference change rate is greater than zero for more than the preset entry confirmation cycle number, the controller determines that the reverse flow conduction condition has been entered and generates a reverse flow start command.

[0010] Furthermore, the controller subtracts the filtered downstream pressure value from the preset downstream target pressure value to obtain the pressure tracking error. The pressure tracking error is multiplied by the control cycle and accumulated cycle by cycle to obtain the cumulative integral value of the pressure tracking error. The pressure tracking error is multiplied by the first sliding mode coefficient and then added to the cumulative integral value multiplied by the second sliding mode coefficient. The sum is used as the sliding mode switching variable.

[0011] Furthermore, the controller performs a superspiral second-order sliding mode operation on the sliding mode switching variable: setting a first superspiral state register and a second superspiral state register; within each control cycle, taking the absolute value of the sliding mode switching variable and calculating its arithmetic square root, multiplying the arithmetic square root by the first superspiral gain and then by the sign value of the sliding mode switching variable to obtain the first superspiral component, adding the current stored value of the first superspiral state register to the first superspiral component to obtain the basic control quantity and writing it into the first superspiral state register; multiplying the sign value of the sliding mode switching variable by the second superspiral gain to obtain the second superspiral component, adding the current stored value of the second superspiral state register to the product obtained by multiplying the second superspiral component by the control cycle, and writing it into the second superspiral state register.

[0012] Furthermore, the controller synchronously operates a nonlinear disturbance observer, which takes the damping stabilizing chamber pressure value and the current drive current value as input. In each control cycle, the current drive current value is converted into the nominal output force according to the drive current-output force lookup table obtained in advance through static calibration tests and stored in the controller. The nominal net force is obtained by subtracting the damping stabilizing chamber pressure value from the nominal output force and multiplying it by the equivalent force area parameter obtained in advance through valve disc geometry calibration. The difference between the disturbance estimate value currently stored in the disturbance estimate register and the nominal net force is multiplied by the observer gain and then by the control cycle and accumulated in the disturbance estimate register to update the disturbance estimate value. The controller adds the disturbance estimate value to the basic control quantity to obtain the compensated total control quantity, which is then converted into a drive current command output to the solenoid excitation coil after amplitude limiting.

[0013] Furthermore, when the absolute value of the pressure tracking error remains within the preset steady-state error band for more than the preset number of exit confirmation cycles, the controller will reduce the drive current command to zero cycle by cycle according to the preset decreasing slope. Under the restoring force of the transmission connection structure, the lifting counterflow valve disc will reseat on the conical sealing surface of the counterflow valve seat, and the counterflow bypass channel will be closed. After the counterflow bypass channel is closed, the controller enters a preset lock monitoring period. During the lock monitoring period, the controller will only continuously collect pressure signals and will not respond to the trigger judgment of the counterflow entry threshold. After the lock monitoring period ends, the controller will restore the counterflow conduction condition judgment function.

[0014] This invention discloses a pressure reducing valve with reverse flow conduction and pressure stabilization switching functions, which has the following beneficial effects: By integrating a super magnetostrictive drive component, a gradient lattice damping pressure stabilizing cavity, and a sliding mode variable structure control algorithm into the valve body, this invention achieves active switching between forward pressure reduction output and reverse flow conduction pressure relief, solving the technical deficiency of traditional pressure reducing valves that only have unidirectional pressure reduction function and cannot cope with abnormal downstream overpressure conditions. Using a Terfenol-D super magnetostrictive rod as the drive element of the reverse flow bypass valve core, it has a higher response bandwidth and a larger output force per unit volume compared to the traditional electromagnet drive method. It can complete the precise opening and continuous adjustment of the reverse flow valve disc in a very short time, significantly shortening the response delay of reverse flow conduction. The introduction of a permanent magnet bias circuit ensures that the Terfenol-D super magnetostrictive rod operates stably in the linear segment of the magnetostriction curve, guaranteeing the linear correspondence between the driving displacement and the excitation current, and reducing the implementation complexity of the control algorithm. The damping and stabilizing chamber employs a gradient density lattice structure integrally printed using selective laser melting technology. The porosity gradually increases along the medium recirculation direction, causing the recirculating medium to undergo strong diversion, convergence, and vortex dissipation at the inlet to attenuate pressure peaks. At the outlet, it flows out smoothly at a low velocity, effectively suppressing pressure pulsations and water hammer impacts during backflow depressurization, and avoiding secondary impacts on upstream pipelines. A second-order sliding mode controller based on a superspiral algorithm outputs a continuous and smooth drive current signal, eliminating the high-frequency chattering problem of traditional sliding mode control. Combined with a nonlinear disturbance observer for real-time estimation and feedforward compensation of fluid dynamic disturbances and frictional disturbances, it achieves rapid and stable convergence of downstream pressure to the target value during backflow conduction. The sliding mean filtering and multi-cycle confirmation mechanism introduced in the backflow entry judgment effectively avoid false triggering caused by pressure noise. The decreasing slope closing and lockout monitoring period set in the exit phase prevent frequent valve opening and closing oscillations, improving the overall reliability and service life of the system. Attached Figure Description

[0015] Figure 1 This is a cross-sectional structural schematic diagram of the supermagnetostrictive drive assembly provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the internal structure of the gradient lattice damping voltage regulator cavity provided in an embodiment of the present invention; Figure 3 The image shows the magnetostrictive characteristic curves and a schematic diagram of the biased operating point of the Terfenol-D super magnetostrictive rod provided in this embodiment of the invention. Detailed Implementation

[0016] A pressure reducing valve with reverse flow conduction and pressure stabilization switching function includes: a valve body, wherein an upstream inlet cavity, a downstream outlet cavity, a main pressure reducing flow channel and a reverse flow bypass flow channel are provided in the valve body, the main pressure reducing flow channel connects the upstream inlet cavity and the downstream outlet cavity, and the reverse flow bypass flow channel connects the downstream outlet cavity and the upstream inlet cavity. The main valve core assembly is located in the main pressure reducing channel and is used to reduce the pressure of the medium flowing through the main pressure reducing channel in the forward direction. The magnetostrictive drive assembly, installed on the countercurrent bypass channel, includes an outer shell, a solenoid excitation coil, a permanent magnet biasing magnetic circuit, and a Terfenol-D magnetostrictive rod. The permanent magnet biasing magnetic circuit and the Terfenol-D magnetostrictive rod cooperate to form a biasing magnetic circuit closed along the axial direction of the Terfenol-D magnetostrictive rod. After receiving the driving current, the solenoid excitation coil generates a variable excitation magnetic field superimposed inside the Terfenol-D magnetostrictive rod. Under the action of the combined magnetic field of the biasing magnetic field and the variable excitation magnetic field, the Terfenol-D magnetostrictive rod elongates along the axial direction. The lifting counterflow valve disc is driven to the output end of the Terfenol-D super magnetostrictive rod and cooperates with the counterflow valve seat set in the counterflow bypass channel. Under the push of axial elongation displacement, it disengages from the counterflow valve seat to open the counterflow bypass channel. The damping pressure stabilizing cavity is set on the counterflow bypass channel and located between the downstream outlet and the upstream inlet. The interior of the damping pressure stabilizing cavity is set with a gradient density lattice structure with increasing porosity along the medium return direction. The pressure sensor group includes a first pressure sensor, a second pressure sensor, and a third pressure sensor respectively disposed in the upstream inlet cavity, the downstream outlet cavity, and the damping pressure stabilizing cavity; The controller receives pressure signals from the first, second, and third pressure sensors. It determines the reverse flow conduction condition based on the difference between the downstream and upstream pressure values ​​and uses a sliding mode variable structure control algorithm to generate a drive current command, which is output to the solenoid excitation coil to adjust the opening of the lift-type reverse flow valve, thereby achieving stable convergence of the downstream pressure to the preset downstream target pressure value.

[0017] The magnetostrictive drive assembly has a cylindrical housing structure, mounted on the valve body along the axis of the counterflow bypass channel, with its output end facing the lift-type counterflow valve disc. The housing is made of austenitic stainless steel with a wall thickness between 3mm and 5mm. The housing serves a dual function of structural support and magnetic shielding, preventing the magnetic field inside the magnetostrictive drive assembly from leaking to the outside of the valve body and interfering with surrounding electronic components. A front cover and a rear cover are respectively located at both ends of the housing. The front cover has a through hole in its center for the force transmission rod to pass through, with a PTFE sliding seal ring embedded in the hole to prevent media from seeping into the housing cavity. The rear cover has a cable lead-out hole through which the lead of the solenoid excitation coil is led out and connected to the drive current output port of the controller.

[0018] The Terfenol-D magnetostrictive rod is located on the central axis of the outer casing. Terfenol-D is the trade name for a terbium-dysprosium-iron alloy, which exhibits significant dimensional elongation along the direction of an applied magnetic field, with a non-linear magnetostrictive characteristic curve relating elongation to the magnetic field strength. In one embodiment, the Terfenol-D magnetostrictive rod has a diameter of 10 mm and a length of 50 mm. At room temperature, a rod of this size can produce an axial free elongation of 0 to approximately 60 micrometers within a magnetic field strength range of 0 to 150 kA / m. In another embodiment, when a larger driving displacement is required, the length of the Terfenol-D magnetostrictive rod can be increased to 80 mm or even 100 mm, corresponding to a free elongation of approximately 100 to 120 micrometers. The diameter and length of the rod are selected based on the stroke and thrust required by the reverse bypass valve core; a larger diameter results in a larger output thrust, and a longer length results in a larger output displacement.

[0019] The magnetostrictive curve of Terfenol-D material exhibits a typical S-shaped nonlinear characteristic. Under zero bias magnetic field conditions, the rod is in the flat region at the bottom of the magnetostrictive curve, where applying a small excitation magnetic field can only induce an extremely weak elongation response, resulting in very low driving sensitivity. To address this issue, this embodiment introduces a permanent magnet bias circuit. Its function is to pre-apply a constant bias magnetic field to the Terfenol-D magnetostrictive rod, raising the operating point from the flat region at the bottom of the curve to an approximately linear segment in the middle. In the linear segment, the elongation of the Terfenol-D magnetostrictive rod is approximately proportional to the change in the superimposed excitation magnetic field. This allows the controller to precisely control the elongation of the rod by linearly adjusting the driving current, without requiring complex compensation calculations for nonlinear characteristics.

[0020] The specific structure of the permanent magnet biasing magnetic circuit is as follows. Along the axial direction of the Terfenol-D magnetostrictive rod, several permanent magnet rings and magnetically conductive soft iron rings are alternately sleeved on the outer circumferential surface of the rod. In a preferred embodiment, the permanent magnet rings are made of neodymium iron boron material, grade N35 or N42. A gap of 0.1 mm to 0.3 mm is left between the inner diameter of each permanent magnet ring and the outer diameter of the Terfenol-D magnetostrictive rod to facilitate assembly. The outer diameter of the permanent magnet ring is 20 mm, and the axial thickness is 5 mm. The magnetically conductive soft iron rings are made of industrial pure iron or electromagnetic pure iron, with a saturation magnetic induction intensity of not less than 1.8 T. The inner and outer diameters are consistent with those of the permanent magnet rings, and the axial thickness is 3 mm to 5 mm. Typically, 4 to 6 permanent magnet rings and a corresponding number of magnetically conductive soft iron rings are arranged along the entire length of the rod.

[0021] like Figure 1As shown, the overall magnetostrictive drive assembly is cylindrical and mounted on the valve body along the axial direction of the counter-flow bypass channel. The outermost layer is the outer shell, made of austenitic stainless steel, which serves both structural support and magnetic shielding functions. A front cover is located at the top of the outer shell, with a through hole in the center for the force transmission rod to pass through. A rear cover is located at the bottom of the outer shell, with a cable lead-out hole on the rear cover. The lead wire of the solenoid excitation coil is led out through the cable lead-out hole and connected to the controller. The Terfenol-D magnetostrictive rod is located on the central axis of the outer shell and extends axially. On the outer circumference of the Terfenol-D magnetostrictive rod, several permanent magnet rings and magnetically conductive soft iron rings are alternately nested along the axial direction. The magnetization polarities of adjacent permanent magnet rings are opposite along the axial direction; the direction of the north pole of each permanent magnet ring is indicated by an arrow in the figure. The magnetically conductive soft iron ring is placed between adjacent permanent magnet rings, providing a radial converging path for the magnetic flux. Outside the inner structure composed of a permanent magnet ring and a magnetically conductive soft iron ring, a cylindrical magnetically conductive outer yoke is set. The axial length of the cylindrical magnetically conductive outer yoke covers the entire arrangement range of the permanent magnet ring and the magnetically conductive soft iron ring. The closed path of the bias magnetic flux is indicated by a dashed arrow in the figure. The magnetic flux starts from one pole of the permanent magnet ring, passes axially through the Terfenol-D supermagnetostrictive rod, reaches the adjacent magnetically conductive soft iron ring, enters the cylindrical magnetically conductive outer yoke radially outward, then returns axially along the wall of the cylindrical magnetically conductive outer yoke and returns radially inward through the magnetically conductive soft iron ring on the other side to the other pole of the permanent magnet ring, forming a closed loop. The solenoid excitation coil is nested in the annular space between the outer side of the permanent magnet bias magnetic circuit and the inner side of the cylindrical magnetically conductive outer yoke. The multi-layer winding cross-section of the coil is represented by closely spaced short horizontal lines on both sides in the figure. The Terfenol-D magnetostrictive rod has a disc spring preload assembly at its top, which is composed of several stacked disc spring plates. The stacked cross-section of the spring plates is represented by a zigzag line in the figure. Above the disc spring preload assembly is a spherical force transmission pad. The upper surface of the spherical force transmission pad is spherical, forming a spherical point contact with the valve stem end face of the lift-type counterflow valve. The force transmission rod extends upward from the spherical force transmission pad through the through-hole of the front end cover and out of the outer shell, transmitting the axial elongation displacement of the Terfenol-D magnetostrictive rod to the lift-type counterflow valve.

[0022] The permanent magnet rings are magnetized axially, with adjacent rings having opposite magnetization polarities: the north pole of the first ring faces the front end cap, the north pole of the second ring faces the rear end cap, the third ring faces the front end cap again, and so on. This alternating polarity arrangement creates an axial magnetic flux loop between every two adjacent opposite-polarity rings, passing through the magnetically conductive soft iron ring and the Terfenol-D magnetostrictive rod. The magnetically conductive soft iron ring provides a low-resistivity path for the magnetic flux to converge from the radially outer side of the permanent magnet rings towards the center of the rod, while simultaneously guiding the magnetic flux axially through the Terfenol-D magnetostrictive rod.

[0023] Outside the inner magnetic circuit composed of a permanent magnet ring and a magnetically conductive soft iron ring, a cylindrical magnetically conductive outer yoke is installed. The cylindrical magnetically conductive outer yoke is made of the same electromagnetically pure iron material as the magnetically conductive soft iron ring, with a wall thickness of 3mm to 4mm, and its axial length covers the entire arrangement range of the permanent magnet ring and the magnetically conductive soft iron ring. The function of the cylindrical magnetically conductive outer yoke is to provide an external closed path for the magnetic flux. The magnetic flux originates from one pole of the permanent magnet ring, passes axially through the Terfenol-D magnetostrictive rod to the adjacent magnetically conductive soft iron ring, then radially outward through the magnetically conductive soft iron ring into the cylindrical magnetically conductive outer yoke, axially returns along the wall of the cylindrical magnetically conductive outer yoke to the magnetically conductive soft iron ring on the starting side, and finally radially inward back to the other pole of the permanent magnet ring. In this way, the permanent magnet ring, the magnetically conductive soft iron ring, the Terfenol-D super magnetostrictive rod, and the cylindrical magnetically conductive outer yoke together form a biased magnetic circuit that is closed along the axial direction of the Terfenol-D super magnetostrictive rod. The magnetic flux path is completely enclosed inside the above components, resulting in minimal magnetic leakage.

[0024] Through the above structure, a uniformly distributed constant bias magnetic field is established along the axial direction inside the Terfenol-D magnetostrictive rod via a permanent magnet bias circuit. In one embodiment, the bias magnetic field strength is set to 40 kA / m to 60 kA / m, which corresponds precisely to the central region of the linear operating segment of the magnetostrictive curve of the Terfenol-D material. If the bias magnetic field strength is below 30 kA / m, the operating point deviates towards the bottom of the curve, and the linearity deteriorates; if the bias magnetic field strength is above 80 kA / m, the operating point deviates towards the saturation region at the top of the curve, and the adjustment margin of the excitation magnetic field is compressed. In actual assembly, the bias magnetic field strength can be adjusted by replacing permanent magnet rings of different grades or thicknesses to adapt to the differences in material properties of different batches of Terfenol-D rods. In another alternative implementation, the permanent magnet ring uses samarium cobalt material instead of neodymium iron boron material. The curie temperature of samarium cobalt permanent magnets is higher than 700 degrees Celsius. Under high-temperature medium conditions, it can maintain a stable bias magnetic field without demagnetization and is suitable for scenarios with operating temperatures exceeding 150 degrees Celsius.

[0025] The solenoid excitation coil is housed in an annular space between the outer side of the permanent magnet bias circuit and the inner side of the cylindrical magnetic outer yoke. The excitation coil is wound with enameled copper wire, with a diameter of 0.5mm to 0.8mm and 800 to 1200 turns, in multiple layers of uniform winding to ensure the generated excitation magnetic field is distributed as evenly as possible along the rod's axial direction. When the controller applies a drive current to the solenoid excitation coil, the coil generates a variable excitation magnetic field along the axial direction within its internal space. This variable excitation magnetic field and the constant bias magnetic field provided by the permanent magnet bias circuit are vector-superimposed inside the Terfenol-D magnetostrictive rod, forming a composite magnetic field. When the drive current is positive, the variable excitation magnetic field and the bias magnetic field are in the same direction, increasing the strength of the composite magnetic field and increasing the rod's elongation. When the drive current is negative, the variable excitation magnetic field and the bias magnetic field are in opposite directions, decreasing the strength of the composite magnetic field and reducing the rod's elongation. In one embodiment, the driving current varies from -1A to +1A, and the corresponding variable excitation magnetic field varies from approximately -30kA / m to +30kA / m. After superimposing a bias magnetic field of 50kA / m, the resulting magnetic field varies between 20kA / m and 80kA / m. The axial elongation of the Terfenol-D magnetostrictive rod varies approximately linearly between 10 micrometers and 50 micrometers within this range.

[0026] A thin-walled magnetic separator can also be set between the solenoid excitation coil and the cylindrical magnetic outer yoke. The magnetic separator is made of silicon steel sheet and has a wall thickness of 0.5 mm to 1 mm. Its function is to further constrain the magnetic flux path of the excitation magnetic field, reduce the radial leakage of the excitation magnetic field, and improve the utilization rate of the excitation magnetic field inside the Terfenol-D super magnetostrictive rod.

[0027] A disc spring preload assembly is installed at the top of the Terfenol-D magnetostrictive rod, facing the lifting counterflow valve. The disc spring preload assembly consists of several stacked disc spring plates made of 60Si2Mn spring steel or an equivalent high-elasticity alloy. The plates are stacked in a mating manner, with the convex or concave surfaces of adjacent disc spring plates facing each other to achieve the required axial stiffness and preload. In one embodiment, the disc spring preload assembly consists of six disc spring plates stacked in three mating configurations, with the overall preload set to 200N to 400N. The disc spring preload assembly is fitted onto the outer circumference of the force transmission rod, with the lower end of the force transmission rod contacting the top surface of the Terfenol-D magnetostrictive rod, and the upper end of the force transmission rod connected to a spherical force transmission pad. The lower end face of the disc spring preload assembly abuts against the positioning step surface of the inner cavity of the outer shell, and the upper end face abuts against the annular flange of the force transmission rod, thereby applying an axial preload to the Terfenol-D magnetostrictive rod.

[0028] The reason for applying a pre-compression load is that Terfenol-D materials exhibit a significant hysteresis effect in the unstressed state. This means that the elongation corresponding to the same magnetic field strength differs between the increasing and decreasing magnetic field, resulting in a wide hysteresis loop in the elongation response curve. Applying an appropriate axial pre-compression load can place the magnetic domains within the material in a stress state more conducive to rotation, significantly narrowing the width of the hysteresis loop, thereby eliminating or reducing the hysteresis dead zone and improving the accuracy of the rod's elongation in response to magnetic field changes. In one embodiment, the pre-compression stress is set to 8 MPa to 12 MPa, which corresponds to the narrowest pre-stress range of hysteresis in the Terfenol-D rod. If the pre-compression stress is below 5 MPa, the hysteresis dead zone suppression effect is not significant; if the pre-compression stress is above 20 MPa, the maximum free elongation of the rod is suppressed and reduced, resulting in a decrease in usable stroke.

[0029] The spherical force transmission pad is a steel block with a spherical upper surface and a flat lower surface, with a hardness of not less than HRC58. The spherical surface of the spherical force transmission pad forms point contact or small-area contact with the valve stem end face of the lift-type counterflow valve disc. The reason for using spherical contact is that there will inevitably be assembly coaxiality deviation and thermal expansion deformation under working conditions between the magnetostrictive drive assembly and the lift-type counterflow valve disc. If a rigid force transmission method of plane to plane is used, any slight angular deviation will cause an off-center bending moment on the contact surface. This bending moment is transmitted to the Terfenol-D magnetostrictive rod through the force transmission rod, causing the rod to be subjected to transverse shear force. Terfenol-D is a brittle rare earth alloy material, and its bending and shear strengths are much lower than its compressive strength. Transverse shear force may cause the rod to break. The spherical force transmission pad ensures that the force transmission path always passes along the direction of the sphere's center. Even if there is a small angular wobble of the valve stem, the contact force is still transmitted axially in the form of pressure, avoiding the generation of transverse force. In one alternative embodiment, the spherical crown radius of the spherical force transmission pad is 8 mm to 15 mm, and the contact surface is finely ground and coated with a molybdenum disulfide solid lubricating film to reduce the friction of the contact surface.

[0030] As the drive current output by the controller gradually increases from zero, the variable excitation magnetic field generated by the solenoid excitation coil is superimposed on the bias magnetic field, increasing the combined magnetic field strength of the Terfenol-D magnetostrictive rod, causing the rod to elongate axially. This elongation is transmitted to the valve stem of the lift-type counterflow valve disc via the force transmission rod and spherical force transmission pad in the disc spring preload assembly, pushing the valve disc upwards axially. The sealing surface of the valve disc disengages from the conical sealing surface of the counterflow valve seat, opening the flow cross-section of the counterflow bypass channel. A larger drive current results in greater rod elongation, a higher lift height for the lift-type counterflow valve disc, a larger flow cross-sectional area in the counterflow bypass channel, and a larger allowable return medium flow rate. When the controller reduces the drive current, the combined magnetic field strength decreases, the rod retracts, and the elastic restoring force of the disc spring preload assembly pushes the lift-type counterflow valve disc back towards the closing direction, correspondingly reducing the flow cross-sectional area. When the drive current decreases to zero, only the constant bias magnetic field provided by the permanent magnet bias magnetic circuit acts on the bar. The bar retracts to the initial length corresponding to the pre-biased state. The disc spring preload assembly pushes the lifting counterflow valve disc completely back onto the counterflow valve seat. The conical sealing surface re-fits, and the counterflow bypass channel is closed.

[0031] In one optional embodiment, the magnetostrictive drive assembly further includes a water-cooled bushing disposed between the inner wall of the outer casing and the outer wall of the solenoid excitation coil. Cooling water is circulated through the water-cooled bushing to dissipate the Joule heat generated by the solenoid excitation coil during continuous energization, preventing the coil temperature from becoming too high and causing the insulation layer to age or the operating temperature of the Terfenol-D magnetostrictive rod to exceed the allowable range. The inlet and outlet water pipes of the water-cooled bushing are led out from the rear end cover and connected to an external circulating cooling device. Under intermittent operating conditions where the drive current does not exceed 0.5A and the duty cycle is less than 50%, the water-cooled bushing can be omitted, and the natural heat dissipation of the outer casing can be relied upon alone.

[0032] In another alternative implementation, strain gauges are attached to the outer circumferential surface of the force transmission rod. The resistance change of the strain gauges is proportional to the axial deformation of the force transmission rod. The resistance change of the strain gauges is converted into a voltage signal and output to the controller via a Wheatstone bridge circuit. The controller then obtains the indirect measurement value of the actual elongation displacement of the Terfenol-D magnetostrictive rod, thereby achieving closed-loop feedback of the drive displacement. This method is suitable for applications requiring high accuracy in valve disc positioning.

[0033] The lift-type counterflow valve disc has a rotating structure and is made of 17-4PH precipitation-hardening stainless steel. After solution treatment and age hardening, its surface hardness reaches HRC38 to HRC42, providing both corrosion resistance and sufficient contact hardness for the sealing surface. The lift-type counterflow valve disc is integrally machined from two parts: the valve disc head and the valve stem. The lower end face of the valve disc head is machined into a conical sealing surface with a cone angle of 60 degrees and a sealing surface width of 1.5mm to 2.5mm. After precision grinding, the surface roughness of the sealing surface is no greater than Ra0.4 micrometers. The valve stem is a cylindrical straight rod with a diameter of 6mm to 10mm, extending along the axis of the counterflow bypass channel. The upper end face of the valve stem, which is the end face that contacts the spherical force transmission pad, is surface ground to a surface roughness of no greater than Ra0.8 micrometers to ensure a stable spherical point contact between the spherical crown surface of the spherical force transmission pad and the valve stem end face.

[0034] The counterflow valve seat is embedded in the inner wall of the counterflow bypass channel and is made of stainless steel of the same material as the valve body or a cobalt-based alloy weld overlay with higher hardness. The upper end of the inner bore of the counterflow valve seat is machined with a tapered valve port that mates with the tapered sealing surface of the lift-type counterflow valve disc. The cone angle of the tapered valve port is the same as the cone angle of the tapered sealing surface of the valve disc head, both being 60 degrees. When the lift-type counterflow valve disc is in the closed position, the tapered sealing surface of the valve disc head is fully fitted with the tapered valve port of the counterflow valve seat, forming a hard metal seal. The sealing pressure is provided by the restoring force of the disc spring preload assembly. The reason for using a hard seal with a conical surface instead of a planar or elastic seal is that the counterflow bypass channel is closed most of the time, and the valve disc is subjected to lateral leakage pressure generated by the pressure difference in the main pressure-reducing channel for a long time. The conical seal has self-centering capability; during the valve disc's seating process, the conical surface naturally guides the valve disc back to the coaxial position. Even if there is a slight radial misalignment, the conical surface can restore the concentric seal. In contrast, planar seals are prone to uneven wear and leakage when there is radial misalignment. In one optional embodiment, the sealing surface of the conical valve port is machined with two concentric annular narrow ridges, each 0.3 mm wide, with a spacing of 1 mm between the ridges. This narrow ridge structure concentrates the contact stress on the sealing surface on an extremely narrow annular band, achieving a higher sealing specific pressure under the same spring preload, thus improving the sealing reliability under low pressure differential conditions.

[0035] The middle section of the valve stem passes through a guide sleeve set on the inner wall of the counterflow bypass channel. The guide sleeve is made of copper alloy or filled with polytetrafluoroethylene (PTFE), and the single-sided fit clearance between the inner hole and the outer surface of the valve stem is 0.02 mm to 0.05 mm. The function of the guide sleeve is to constrain the valve stem to move only axially, preventing the valve disc from radially swaying or tilting under the impact of the medium, and ensuring uniform contact of the conical sealing surface around the entire circumference. The length of the guide sleeve is 1.5 to 2 times the diameter of the valve stem to provide sufficient guiding accuracy. In an optional embodiment, several axially distributed lubrication grooves are formed on the inner surface of the guide sleeve. The grooves are 0.5 mm wide and 0.3 mm deep, and are filled with molybdenum disulfide-based solid grease to provide continuous dry lubrication for the valve stem under long-term maintenance-free conditions, reducing the friction of the valve stem movement.

[0036] The stroke of the lift-type reverse flow valve disc, i.e., the axial distance from the point where the conical sealing surface is fully engaged to the maximum open position, depends on the maximum output displacement of the magnetostrictive drive assembly. In the aforementioned embodiment, the usable elongation of the Terfenol-D magnetostrictive rod is 10 to 50 micrometers, which is relatively small for large-diameter valves. Therefore, in an optional embodiment, a flexible hinge displacement amplification mechanism is added between the disc spring preload assembly and the spherical force transmission pad. The flexible hinge displacement amplification mechanism adopts a diamond-shaped bridge structure, integrally machined from a single piece of spring steel via wire cutting. The input end receives the axial elongation displacement of the Terfenol-D magnetostrictive rod, and the output end amplifies the displacement and transmits it to the spherical force transmission pad, with an amplification factor of 5 to 10 times. The amplified effective output stroke can reach 50 to 500 micrometers, which is sufficient to meet the stroke requirements of reverse flow bypass valve cores with nominal diameters of DN15 to DN50. The wall thickness at each hinge of the flexible hinge displacement amplification mechanism is 0.3mm to 0.5mm, and the maximum stress at the hinge does not exceed 40% of the yield strength of the spring steel to ensure fatigue life.

[0037] The damping pressure-stabilizing chamber is located on the counter-flow bypass channel, between the downstream outlet and the upstream inlet. From a flow path perspective, the downstream outlet enters the inlet of the damping pressure-stabilizing chamber via a switchable channel formed by the counter-flow valve seat and the lift-type counter-flow valve disc. The outlet of the damping pressure-stabilizing chamber connects to the upstream inlet via a return channel. When the lift-type counter-flow valve disc disengages from the counter-flow valve seat, the high-pressure medium in the downstream outlet flows sequentially through the conical valve port of the counter-flow valve seat, the damping pressure-stabilizing chamber, and the return channel, ultimately converging into the upstream inlet, completing the counter-flow return path from downstream to upstream.

[0038] The damping and stabilizing cavity is a monolithic component formed by selective laser melting (SLM) using 316L stainless steel powder. SLM is a powder bed laser additive manufacturing technology that selectively melts and stacks metal powder layer by layer using a high-energy laser beam, enabling the creation of complex internal cavities and fine lattice structures that are difficult to achieve using traditional casting and machining. The 316L stainless steel powder has a particle size distribution of 15 to 45 micrometers, a printing layer thickness of 30 to 50 micrometers, a laser power of 200W to 400W, and a scanning speed of 800 mm / s to 1200 mm / s. After printing, the component undergoes stress-relief annealing at 650 degrees Celsius for 2 hours, followed by furnace cooling to room temperature. After annealing, the component has a tensile strength of not less than 500 MPa and a yield strength of not less than 200 MPa, meeting the mechanical performance requirements of the valve body's pressure-bearing components. After printing, the damping and stabilizing cavity is equipped with flange or threaded interfaces at both ends for sealing connection with the counterflow bypass channel of the valve body.

[0039] The damping and stabilizing cavity is internally divided into three sections along the medium return direction: a high-density lattice section at the inlet, a transitional lattice section, and a low-density lattice section at the outlet. All three sections utilize body-centered cubic (BCC) lattice units. Each BCC unit consists of eight corner nodes and one body node connected by connecting rods. Each lattice unit contains 12 edge rods and four body diagonal rods, totaling 16 rods. This lattice topology exhibits isotropic mechanical response and a high specific surface area, enabling sufficient flow around and separation of the fluid.

[0040] The high-density lattice section at the inlet is located at the upstream end of the medium recirculation direction, adjacent to the outlet of the counterflow valve seat. The lattice unit side length in this section is 2 mm, the lattice rod diameter is 0.8 mm, and the porosity is 40%. Porosity is calculated as the percentage of the void volume of the lattice unit to the total envelope volume of the lattice unit, i.e. ,in Porosity The total envelope volume of the lattice unit is the cube of the side length. This refers to the sum of the volumes of all the bars within the lattice unit. The high-density lattice section at the inlet end has densely packed bars and a narrow flow cross-section. When the high-pressure return medium enters this section at a high velocity, the dense array of bars forces the medium stream to repeatedly split and converge. A low-pressure separation zone forms at the tail of each bar. After bypassing the bars, the stream cannot maintain laminar adhesion and detaches in the wake region, creating numerous micro-scale vortices. These micro-scale vortices, through vortex-to-vortex interaction and viscous dissipation, gradually convert the pressure impact kinetic energy of the medium into heat energy, thus completing the first stage of pressure peak attenuation within the high-density lattice section at the inlet end. In one embodiment, the axial length of this section is set to 20 mm to 30 mm, accounting for 25% to 30% of the total length of the damping and stabilizing cavity.

[0041] like Figure 2 As shown, the damping pressure stabilizing chamber is a tubular structure with flange interfaces at both ends for sealing connection with the valve body's reverse bypass channel. The high-pressure return medium enters the damping pressure stabilizing chamber through the right flange interface, flows smoothly out through the left flange interface after passing through the internal lattice structure, and merges into the upstream inlet. The interior of the damping pressure stabilizing chamber is divided into three sections along the medium return direction, with adjacent sections marked by dashed lines. The first section on the right is the high-density lattice section at the inlet, filled with closely packed body-centered cubic lattice units. The lattice unit side length is 2mm, the lattice rod diameter is 0.8mm, and the porosity is 40%. This section has the densest arrangement of lattice rods and nodes, with the smallest spacing between rods. The second section in the middle is the intermediate transition lattice section, also using body-centered cubic lattice units, but with the lattice unit side length increased to 3mm, the lattice rod diameter decreased to 0.6mm, and the porosity increased to 58%. As shown in the figure, the spacing between the rods in this section is significantly larger than that in the high-density lattice section at the inlet, and the node distribution becomes sparser. The third section on the left is the low-density lattice section at the outlet, where the lattice unit side length further increases to 4 mm, the lattice rod diameter further decreases to 0.4 mm, and the porosity increases to 75%. The rod distribution in this section is the sparsest, and the flow space is the largest. The structural parameters of the three sections are marked below each section. From the high-density lattice section at the inlet to the low-density lattice section at the outlet, the lattice unit side length gradually increases, the rod diameter gradually decreases, and the porosity gradually increases, forming a gradient density lattice structure with increasing porosity along the medium recirculation direction. This gradient distribution causes the recirculating medium to experience the strongest splitting-converging-vortex dissipation effect at the inlet to attenuate the pressure peak, the vortices gradually merge and attenuate in the middle section to further reduce pressure pulsation, and the medium is smoothly output at the outlet with a large flow area and low velocity. The arrows in the diagram indicate the return flow direction of the medium, which enters from the high-pressure side on the right, undergoes energy dissipation in three stages, and then flows out slowly and steadily from the left.

[0042] The intermediate transition lattice section is located downstream of the high-density lattice section at the inlet. The lattice unit edge length in this section is 3 mm, the lattice rod diameter is 0.6 mm, and the porosity is 58%. Compared to the high-density lattice section at the inlet, the rod spacing is increased while the rods are thinner, resulting in a larger flow cross-sectional area. When the medium flows through the intermediate transition lattice section, the number of newly generated microscale vortices decreases due to the reduced rod density. Simultaneously, the vortices generated in the upstream high-density lattice section at the inlet gradually merge into larger-scale vortex structures within the larger rod spacing. The energy of these large-scale vortices is further dissipated through cascading under viscosity, resulting in a second-stage attenuation of the pressure pulsation amplitude within this section. In one embodiment, the axial length of this section is set to 25 mm to 35 mm, accounting for 30% to 35% of the total length of the damping stabilizing cavity.

[0043] The low-density lattice section at the outlet is located downstream of the intermediate transition lattice section and is the last section the medium passes through before leaving the damping stabilizing cavity. The lattice unit edge length in this section is 4 mm, the lattice rod diameter is 0.4 mm, and the porosity is 75%. The high porosity of 75% means that most of the space within this section is flow space, and the rods only serve to maintain structural self-support and final rectification. After the gradual attenuation of the first two sections, the velocity and pressure pulsations of the medium have been significantly reduced. The large pore flow area in the low-density lattice section at the outlet further decelerates the medium, allowing it to flow out of the damping stabilizing cavity at a low speed and with a stable flow pattern, merging into the upstream inlet cavity via the return channel. In one embodiment, the axial length of this section is set to 25 mm to 40 mm, accounting for 30% to 40% of the total length of the damping stabilizing cavity.

[0044] The porosity of the three sections increases from 40% to 58% and then to 75% along the direction of medium recirculation, forming a gradient density lattice structure with increasing porosity. The intention behind this gradient, rather than uniform, porosity design is to create a spatially non-uniform distribution of damping characteristics: applying maximum flow resistance and dissipation intensity at the inlet end where the pressure shock energy is most concentrated to quickly reduce the peak; gradually reducing the dissipation intensity in the middle section to avoid excessive pressure drop leading to insufficient recirculation flow; and providing the lowest flow resistance at the outlet end to ensure the medium leaves in a stable flow state. If a uniform high-density lattice is used throughout, although the peak-shaving effect is strong, the total pressure drop is too large, severely limiting the recirculation flow and preventing downstream pressure relief from being completed within a reasonable time. Conversely, if a uniform low-density lattice is used throughout, the inlet end lacks sufficient dissipation capacity, and the pressure shock wave passes through the damping and stabilizing cavity almost without attenuation, directly reaching the upstream inlet and triggering a secondary impact on the upstream pipeline. The gradient density lattice structure achieves a balance between these two extremes, balancing peak-shaving effect and smooth recirculation.

[0045] In one alternative embodiment, the porosity transition between the three sections is not abrupt but rather a linear gradient. Specifically, at the boundary between the high-density lattice section at the inlet and the intermediate transition lattice section, the edge length of the lattice unit and the diameter of the rods gradually change layer by layer within the transition zone of 3 to 5 lattice units to avoid localized eddies and pressure reflections at abrupt porosity changes. In another alternative embodiment, the topology of the lattice unit is not limited to a body-centered cubic structure; face-centered cubic lattice units or octagonal lattice units can also be used. In face-centered cubic lattice units, the rods are connected to the corner nodes and face-centered nodes via rods. In octagonal lattice units, multiple sets of rods are included along the body diagonal and face diagonal directions. The flow characteristics and specific surface area differ depending on the topology, and the appropriate topology can be selected based on the viscosity and Reynolds number range of the working medium.

[0046] During the selective laser melting (SLM) process of printing lattice rods, the rod surface naturally retains the micron-level roughness formed by the partial melting of powder, with a surface roughness of approximately Ra 10 to Ra 25 microns, far greater than that of machined surfaces. This micron-level roughness is beneficial to the fluid dissipation function of the damping pressure stabilizing cavity. Under low Reynolds number conditions, the rough rod surface promotes the boundary layer to transition to turbulence earlier, increasing the frictional resistance coefficient of the rod surface and further enhancing the dissipation effect of pressure pulsation. Therefore, the surface of the lattice rods in the damping pressure stabilizing cavity does not undergo additional polishing or electrolytic treatment after printing, but retains the original surface morphology of the selective laser melting process.

[0047] The pressure sensor group includes three pressure sensors: a first pressure sensor, a second pressure sensor, and a third pressure sensor, embedded in the upstream inlet wall, the downstream outlet wall, and the damping pressure regulating chamber wall, respectively. All three pressure sensors are identical piezoresistive pressure transmitters, covering the maximum operating pressure of the pressure reducing valve, with an accuracy class of no less than 0.25%FS. The pressure transmitter's diaphragm is in direct contact with the medium and is made of 316L stainless steel, compatible with the valve body and damping pressure regulating chamber materials to avoid electrochemical corrosion. Each pressure transmitter outputs a standard current signal of 4mA to 20mA or a voltage signal of 0V to 5V, connected to the controller's analog input port via a shielded cable.

[0048] The first pressure sensor is embedded in the upstream inlet wall at a distance of at least 5 times the pipe inner diameter upstream of the main pressure-reducing channel inlet. This distance ensures that the pressure sensed by the sensor is the fully developed static pressure within the upstream inlet, unaffected by localized flow acceleration at the main pressure-reducing channel inlet. The second pressure sensor is embedded in the downstream outlet wall at a distance of at least 5 times the pipe inner diameter downstream of the main pressure-reducing channel outlet, similarly ensuring that the measured pressure is the fully developed downstream outlet pressure. The third pressure sensor is embedded in the damping pressure-stabilizing cavity wall near the junction of the high-density lattice section and the intermediate transition lattice section at the inlet. At this location, the medium has undergone first-stage peak clipping in the high-density lattice section at the inlet but has not yet fully attenuated. The sensed pressure value from the damping pressure-stabilizing cavity reflects the actual pressure level of the current return medium passing through the damping pressure-stabilizing cavity, providing an effective disturbance estimation input for the controller's nonlinear disturbance observer.

[0049] The controller is implemented using a digital signal processor or an industrial programmable logic controller (PLC), and has at least three analog signal input channels and at least one analog signal output channel. The analog signal input channels are equipped with analog-to-digital converters (ADCs) with a resolution of at least 16 bits, and the analog signal output channel is equipped with digital-to-analog converters (DACs) with a resolution of at least 16 bits. The controller's sampling period and control period are set to the same value; in one implementation, this is set to 1 ms, corresponding to a sampling and control frequency of 1 kHz. In another implementation, when the system pressure pulsation frequency is high, the sampling and control frequency can be increased to 5 kHz or 10 kHz. The controller's internal calculations are performed in fixed-point or floating-point arithmetic mode. When using fixed-point arithmetic, all pressure values ​​and control quantities are scaled to a 32-bit integer range using a preset scaling factor to ensure calculation accuracy.

[0050] In each sampling period, the controller reads the upstream and downstream pressure values ​​from the first and second pressure sensors, respectively, and performs moving average filtering on both signals. In one embodiment, the window length of the moving average filter is set to 8 to 16 sampling periods. That is, the controller internally stores the upstream and downstream pressure values ​​for the most recent 8 to 16 sampling periods, and in each sampling period, the arithmetic mean of all stored values ​​within the window is used as the filtered upstream and downstream pressure values ​​for the current period. The moving average filter aims to suppress high-frequency measurement noise and electrical interference in the pressure signal. A larger window length results in a stronger filtering effect but also introduces greater signal delay. The selection of the window length requires a trade-off between noise suppression capability and response speed. In an optional embodiment, the moving average filter can be replaced by a first-order exponentially weighted filter. The recursive relationship of the first-order exponentially weighted filter is... ,in This is the filtered output value for the current sampling period. The original sampled values ​​for the current sampling period. This is the filtered output value from the previous sampling period. These are the filter coefficients, ranging from 0 to 1. Smaller values ​​indicate stronger filtering. In one implementation... Take a value between 0.1 and 0.3.

[0051] The controller subtracts the filtered upstream pressure value from the filtered downstream pressure value to obtain the reverse pressure difference value. Under normal forward pressure reduction operation, the upstream pressure value is always higher than the downstream pressure value, the reverse pressure difference is negative, and the pressure reducing valve is in the forward pressure reduction output state. When abnormal high pressure occurs in the downstream pipeline or the upstream supply pressure suddenly drops, the downstream pressure value may equal or even exceed the upstream pressure value, and the reverse pressure difference value changes from negative to zero or positive. The controller subtracts the reverse pressure difference value of the previous sampling period from the current sampling period's reverse pressure difference value and then divides by the sampling period to obtain the reverse pressure difference change rate, which characterizes the rate at which the reverse pressure difference value changes over time.

[0052] The controller internally maintains an entry confirmation counter, initially set to zero. When the reverse pressure differential exceeds a preset reverse flow entry threshold and the reverse pressure differential change rate is greater than zero, the entry confirmation counter increments by 1 in each sampling cycle. Once the reverse pressure differential is no greater than the preset reverse flow entry threshold or the reverse pressure differential change rate is no greater than zero, the entry confirmation counter is immediately reset to zero. When the count value of the entry confirmation counter exceeds a preset number of entry confirmation cycles, the controller determines the current reverse flow conduction condition and generates a reverse flow start command. In one embodiment, the preset reverse flow entry threshold is set to 5% to 10% of the upstream normal operating pressure, and in another embodiment, the number of entry confirmation cycles is set to 10 to 50 sampling cycles. The purpose of setting the number of entry confirmation cycles is to distinguish between genuine reverse flow conditions and brief disturbances in the pressure signal—reverse flow start is only triggered when the reverse pressure differential continuously exceeds the reverse flow entry threshold and continues to rise for a preset confirmation duration. Brief pressure pulses or momentary threshold exceedances caused by sensor noise will not trigger malfunctions.

[0053] After generating the reverse flow start command, the controller enters the reverse flow conduction mode. At this time, the controller activates the sliding mode variable structure control algorithm to perform closed-loop regulation of the drive current of the solenoid excitation coil. The controller first reads the downstream pressure value after sliding mean filtering in the current sampling period, and subtracts the filtered downstream pressure value from the preset downstream target pressure value to obtain the pressure tracking error. The preset downstream target pressure value is the nominal output pressure of the pressure reducing valve under normal forward pressure reducing operation, and in one embodiment, it is set as the rated outlet pressure value marked on the pressure reducing valve nameplate.

[0054] The controller has an internal integral register, which is cleared to zero at the start of the reverse-current conduction operation. In each subsequent control cycle, the controller multiplies the current pressure tracking error by the control cycle number and adds this multiplication to the stored value in the integral register, obtaining the cumulative integral value of the pressure tracking error. This cumulative integral value reflects the degree of accumulation of the pressure tracking error over time. The purpose of introducing the integral component is to include the ability to eliminate static deviations in the design of the sliding surface, ensuring that the pressure tracking error converges to zero in steady state rather than having an inherent steady-state residual.

[0055] The controller multiplies the pressure tracking error by a first sliding mode coefficient, then adds the accumulated integral value multiplied by a second sliding mode coefficient, and the sum is used as the sliding mode switching variable. The first and second sliding mode coefficients are preset positive real numbers, and their ratio determines the slope of the sliding surface in the phase plane. In one embodiment, the first sliding mode coefficient is set to 1.0, and the second sliding mode coefficient is set to 0.5 to 2.0. A larger first sliding mode coefficient results in a greater contribution of the pressure tracking error itself to the sliding mode switching variable, leading to a faster system response but a higher risk of overshoot. A larger second sliding mode coefficient results in a greater contribution of the integral component, strengthening the ability to eliminate steady-state deviations but potentially slowing down the response process. The specific values ​​of the two coefficients are determined through simulation or on-site debugging based on system parameters such as the pressure reducing valve's diameter, spring stiffness, and medium viscosity.

[0056] The controller performs a superspiral second-order sliding mode operation on the sliding mode switching variable. The superspiral algorithm is a continuous second-order sliding mode control law. Its core feature is that the output control quantity is a continuous signal rather than a discontinuous signal containing sign function jumps. Therefore, it can eliminate the high-frequency chattering caused by sign function switching in traditional first-order sliding mode control while ensuring convergence to the sliding surface in a finite time. The output signal of a traditional first-order sliding mode controller switches back and forth at extremely high frequencies on both sides of the sliding surface. Applying this high-frequency switching signal to the magnetostrictive drive component causes high-frequency micro-oscillations in the Terfenol-D magnetostrictive rod, leading to mechanical wear and acoustic noise. The superspiral algorithm, by integrating the sign function once, transforms the discontinuous terms in the control law into continuous terms, resulting in a smooth and continuous output drive current command, directly eliminating the aforementioned problems.

[0057] The second-order sliding mode operation of the superspiral is implemented internally by two state registers, designated as the first superspiral state register and the second superspiral state register, both of which are cleared to zero at the start of the reverse-current conduction condition. Within each control cycle, the controller performs the following operations: The absolute value of the sliding mode switching variable is taken and its square root is calculated; this square root is multiplied by the first superspiral gain and then by the sign value of the sliding mode switching variable to obtain the first superspiral component; the current stored value of the first superspiral state register is added to the first superspiral component to obtain the basic control quantity, which is then written to the first superspiral state register; the sign value of the sliding mode switching variable is multiplied by the second superspiral gain to obtain the second superspiral component; the current stored value of the second superspiral state register is added to the product of the second superspiral component and the control cycle, and then written to the second superspiral state register. The continuous-time domain superspiral control law corresponding to the above operations is: as well as ,in For sliding mode switching variables, The basic control values ​​stored in the first superspiral state register. The auxiliary integral quantity stored in the second superspiral state register. For the first superspiral gain, For the second superspiral gain, This is a sign function. Internally, the controller discretizes the above continuous-time differential equation using the Euler forward difference method, and the time step is the control period.

[0058] The values ​​of the first and second superspiral gains directly determine the convergence speed and robustness of the system. In one implementation, the first superspiral gain is set to 10 to 50, and the second superspiral gain is set to 5 to 30. The values ​​of the two gains must satisfy the sufficient convergence condition of the superspiral algorithm, that is, the second superspiral gain must be greater than the absolute value of the upper bound of the system disturbance, and the first superspiral gain must be greater than a certain proportional multiple of the sum of the second superspiral gain and the upper bound of the system disturbance. The upper bound of the system disturbance is determined based on the maximum estimated values ​​of the hydrodynamic disturbance and frictional disturbance of the pressure reducing valve under the worst operating conditions.

[0059] The controller executes a second-order sliding mode operation of the superspiral mechanism while simultaneously running a nonlinear disturbance observer. The nonlinear disturbance observer's function is to estimate the external disturbances acting on the lift-type counterflow valve disc in real time and feed the estimated disturbance values ​​forward to compensate the control output, thereby reducing the compensation burden on the sliding mode controller. The external disturbances experienced by the lift-type counterflow valve disc during operation mainly fall into two categories: one is the hydrodynamic force exerted by the return medium on the valve disc head, which varies with the flow cross-sectional area and medium velocity and cannot be directly measured by the controller; the other is the frictional force between the valve stem and the guide sleeve, which is related to the valve stem's movement speed and the wear state of the guide sleeve, and is also difficult to predict accurately. The nonlinear disturbance observer treats these two types of unmeasurable disturbances as a single lumped disturbance quantity for estimation.

[0060] The nonlinear disturbance observer takes the pressure value of the damped stabilizing chamber collected by the third pressure sensor and the current output drive current value of the controller as input. It internally maintains a disturbance estimation register, which is cleared to zero at the start of the reverse flow conduction condition. Within each control cycle, the nonlinear disturbance observer first converts the current drive current value into the nominal output force according to the Terfenol-D magnetostrictive rod drive current-output force lookup table, which was obtained in advance through static calibration tests and stored in the controller. The drive current-output force lookup table is obtained as follows: before the magnetostrictive drive assembly leaves the factory or is assembled into the valve body, the drive current is gradually increased from zero to its maximum value and then gradually decreased back to zero. Simultaneously, the output force at the tip of the Terfenol-D magnetostrictive rod is measured using a force sensor. The one-to-one correspondence between the drive current value and the corresponding output force value is recorded and stored in the controller's non-volatile memory in the form of a lookup table. In one embodiment, the current step size of the lookup table is set to 0.01A, covering the full range from 0 to the maximum drive current. During operation, the controller can obtain the nominal output force value by performing linear interpolation in a lookup table based on the current drive current value.

[0061] The nonlinear disturbance observer then subtracts the damping pressure chamber pressure from the nominal output force and multiplies it by the equivalent force area parameter obtained beforehand through valve disc geometry calibration, to obtain the nominal net force. The equivalent force area parameter is a scalar parameter, physically representing the proportionality between the axial force generated by the damping pressure chamber pressure acting on the effective area of ​​the lift-type counterflow valve disc and the damping pressure chamber pressure. The equivalent force area parameter is calibrated as follows: with the valve disc in a static closed state, a medium at a known pressure is injected into the damping pressure chamber, and the axial holding force required to maintain the valve disc's closure is measured. The holding force divided by the injection pressure is the equivalent force area parameter. In one embodiment, for a lift-type counterflow valve disc with a valve disc head diameter of 12 mm, the equivalent force area parameter is approximately 113 square millimeters. The nominal net force represents the net difference between the output force of the magnetostrictive drive component and the reaction force generated by the damping pressure chamber pressure under ideal conditions without external disturbance; this net difference drives the valve disc movement.

[0062] The nonlinear disturbance observer multiplies the difference between the disturbance estimate currently stored in the disturbance estimation register and the nominal net force by the observer gain, then multiplies by the control period, and finally adds this result back to the disturbance estimation register to update the disturbance estimate. The observer gain is a positive real number, and its physical meaning is the rate at which the observer corrects the disturbance estimate. The larger the gain, the faster the disturbance estimate tracks the actual disturbance, but the more sensitive it is to measurement noise. In one implementation, the observer gain is set to between 50 and 200. If the observer gain is too low, the disturbance estimate lags behind the actual disturbance, and the compensation effect decreases; if the observer gain is too high, measurement noise is amplified and injected into the control loop, causing drive current jitter.

[0063] The controller adds the disturbance estimate to the base control quantity to obtain the compensated total control quantity. This total control quantity, after limiting, is converted into a drive current command and output to the solenoid excitation coil. The limiting process restricts the total control quantity to a preset maximum and minimum drive current value, preventing it from exceeding the rated current range of the solenoid excitation coil or the safe operating magnetic field range of the Terfenol-D magnetostrictive rod. In one embodiment, the maximum drive current value is set to 1A, and the minimum drive current value is set to 0A, meaning negative current output is not allowed to avoid pulling the synthesized magnetic field below the bias magnetic field, causing the Terfenol-D magnetostrictive rod to exit the linear operating range.

[0064] During the continuous reverse flow operation, the aforementioned sliding mode variable structure control calculation is repeatedly executed in each control cycle. The solenoid excitation coil continuously adjusts the strength of the variable excitation magnetic field according to the drive current command, and the opening of the lift-type reverse flow valve changes accordingly. When the downstream pressure is higher than the downstream target pressure, the pressure tracking error is negative, and the sliding mode control algorithm increases the drive current to increase the valve opening, accelerating the backflow and pressure relief. As the downstream pressure gradually decreases and approaches the downstream target pressure, the pressure tracking error gradually decreases, the drive current decreases accordingly, the valve opening narrows, and the backflow rate decreases. This negative feedback closed loop causes the downstream pressure to continuously approach the preset downstream target pressure.

[0065] When the absolute value of the pressure tracking error remains continuously within a preset steady-state error band for more than a preset number of exit confirmation cycles, the controller determines that the downstream pressure has completed pressure stabilization and convergence. In one embodiment, the steady-state error band is set to ±2% of the downstream target pressure value, and in another embodiment, the number of exit confirmation cycles is set to 50 to 200 control cycles. The purpose of setting the number of exit confirmation cycles is the same as the purpose of setting the number of entry confirmation cycles: to prevent the pressure tracking error from accidentally entering the steady-state error band and then quickly leaving, which could be mistakenly judged as having completed convergence.

[0066] After determining that voltage stabilization and convergence are complete, the controller gradually decreases the drive current command to zero at a preset decreasing slope, rather than directly cutting off the drive current to zero in a step-like manner. The purpose of this gradual decreasing is to avoid sudden retraction of the Terfenol-D magnetostrictive rod due to abrupt changes in the drive current, which could cause the valve disc to violently impact the reverse flow valve seat, resulting in mechanical shock and water hammer effects. In one embodiment, the decreasing slope is set to 0.1 A / s to 0.5 A / s. If the current drive current is 0.5 A and the decreasing slope is 0.25 A / s, it takes 2 seconds for the drive current to decrease from 0.5 A to zero. During the decreasing process, the lifting reverse flow valve disc smoothly approaches the reverse flow valve seat under the restoring force of the disc spring preload assembly until the conical sealing surface re-fits, closing the reverse flow bypass channel.

[0067] like Figure 3As shown, the horizontal axis represents the magnetic field strength applied to the Terfenol-D magnetostrictive rod, in kA / m; the vertical axis represents the magnetostriction of the Terfenol-D magnetostrictive rod, i.e., the axial relative elongation, in ppm. The curve exhibits a typical S-shaped nonlinear characteristic. In the region where the magnetic field strength is close to zero, the curve is in a flat bottom region, and the magnetostriction changes extremely slowly with increasing magnetic field strength. This region is marked as the flat bottom region in the figure. Applying a small excitation magnetic field in this region can only induce a very weak elongation response, resulting in extremely low driving sensitivity, making it unsuitable as a working section. As the magnetic field strength increases, the curve enters a steep upward section in the middle, where the magnetostriction increases approximately linearly with the magnetic field strength. The shaded band in the figure indicates the range of this linear working section, corresponding to a magnetic field strength of approximately 25 kA / m to 85 kA / m. Within the linear working section, the magnetostriction... With magnetic field strength The changes in the quantities approximately satisfy a linear relationship. ,in The piezomagnetic coefficient represents the change in magnetostriction caused by a unit change in magnetic field strength under constant stress. When the magnetic field strength continues to increase beyond the linear operating range, the curve enters the top saturation region, where the magnetostriction approaches the material's saturation magnetostriction value. Further increases in magnetic field strength result in a significantly reduced elongation increment; this region is marked as the saturation region in the figure. A bias operating point is marked at the midpoint of the linear operating range. The magnetic field strength corresponding to this point is the constant bias magnetic field strength established within the Terfenol-D supermagnetostrictive rod by the permanent magnet bias circuit. The bias operating point locks the static operating state of the Terfenol-D supermagnetostrictive rod at the center of the linear range, ensuring that the combined magnetic field strength remains within the linear operating range when a variable excitation magnetic field (positive or negative) is superimposed on the solenoid excitation coil. The figure uses double-headed arrows to indicate the superposition range of the variable excitation magnetic field, which is symmetrically distributed about the bias operating point and corresponds to the effective controllable elongation range of the Terfenol-D supermagnetostrictive rod.

[0068] After the reverse flow bypass channel is closed, the controller enters a preset lockout monitoring period. During this period, the controller continuously collects pressure signals from the first, second, and third pressure sensors, performs sliding average filtering, and calculates the reverse pressure difference. However, it does not respond to the triggering of the reverse flow entry threshold; that is, even if the reverse pressure difference exceeds the threshold again during this period, the controller will not generate a reverse flow opening command. The purpose of setting the lockout monitoring period is that residual pressure oscillations may occur in the pipeline immediately after the reverse flow bypass channel is closed. These residual oscillations may cause the reverse pressure difference to briefly exceed the reverse flow entry threshold. Without the lockout monitoring period, the controller would immediately trigger the reverse flow opening again, resulting in repeated opening and closing of the reverse flow bypass channel within a short period. This oscillation behavior is detrimental to the sealing surface of the lift-type reverse flow valve and the lifespan of the magnetostrictive drive assembly. In one embodiment, the lockout monitoring period is set to 2 to 10 seconds, determined based on the inherent oscillation decay time of the pipeline. After the lockout monitoring period ends, the controller resumes the reverse flow conduction condition determination function, and the pressure reducing valve resumes its forward pressure reducing output state.

[0069] In one optional implementation, the reverse flow entry threshold and the reverse flow exit threshold are set to different values, with the reverse flow exit threshold being lower than the reverse flow entry threshold. The difference between the two constitutes a hysteresis interval. Within the hysteresis interval, if the system is currently in reverse flow conduction mode, it maintains the reverse flow conduction state; if the system is currently in forward decompression mode, it maintains the forward decompression state. This hysteresis design further enhances the stability of state switching, avoiding frequent state switching caused by small fluctuations in the reverse pressure difference near the reverse flow entry threshold.

[0070] The preferred embodiments of this disclosure have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of this disclosure. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of this disclosure shall be within the scope of the claims of this disclosure.

Claims

1. A pressure reducing valve with reverse flow conduction and pressure stabilization switching functions, characterized in that, include: The valve body is equipped with an upstream inlet chamber, a downstream outlet chamber, a main pressure reducing channel, and a counter-current bypass channel. The main pressure reducing channel connects the upstream inlet chamber and the downstream outlet chamber, and the counter-current bypass channel connects the downstream outlet chamber and the upstream inlet chamber. The main valve core assembly is located in the main pressure reducing channel and is used to reduce the pressure of the medium flowing through the main pressure reducing channel in the forward direction. The magnetostrictive drive assembly, installed on the countercurrent bypass channel, includes an outer shell, a solenoid excitation coil, a permanent magnet biasing magnetic circuit, and a Terfenol-D magnetostrictive rod. The permanent magnet biasing magnetic circuit and the Terfenol-D magnetostrictive rod cooperate to form a biasing magnetic circuit closed along the axial direction of the Terfenol-D magnetostrictive rod. After receiving the driving current, the solenoid excitation coil generates a variable excitation magnetic field superimposed inside the Terfenol-D magnetostrictive rod. Under the action of the combined magnetic field of the biasing magnetic field and the variable excitation magnetic field, the Terfenol-D magnetostrictive rod elongates along the axial direction. The lifting counterflow valve disc is driven to the output end of the Terfenol-D super magnetostrictive rod and cooperates with the counterflow valve seat set in the counterflow bypass channel. Under the push of axial elongation displacement, it disengages from the counterflow valve seat to open the counterflow bypass channel. The damping pressure stabilizing cavity is set on the counterflow bypass channel and located between the downstream outlet and the upstream inlet. The interior of the damping pressure stabilizing cavity is set with a gradient density lattice structure with increasing porosity along the medium return direction. The pressure sensor group includes a first pressure sensor, a second pressure sensor, and a third pressure sensor respectively disposed in the upstream inlet cavity, the downstream outlet cavity, and the damping pressure stabilizing cavity; The controller receives pressure signals from the first, second, and third pressure sensors. It determines the reverse flow conduction condition based on the difference between the downstream and upstream pressure values ​​and uses a sliding mode variable structure control algorithm to generate a drive current command, which is output to the solenoid excitation coil to adjust the opening of the lift-type reverse flow valve, thereby achieving stable convergence of the downstream pressure to the preset downstream target pressure value.

2. The pressure reducing valve according to claim 1, characterized in that, The permanent magnet bias magnetic circuit consists of permanent magnet rings and magnetically conductive soft iron rings arranged alternately along the axial direction of the Terfenol-D super magnetostrictive rod, as well as a cylindrical magnetically conductive outer yoke surrounding the outside. The permanent magnet rings, magnetically conductive soft iron rings, and cylindrical magnetically conductive outer yoke together form a closed bias magnetic circuit along the axial direction of the Terfenol-D super magnetostrictive rod. The solenoid excitation coil is nested in the annular space between the outer side of the permanent magnet bias magnetic circuit and the inner side of the cylindrical magnetically conductive outer yoke.

3. The pressure reducing valve according to claim 1, characterized in that, The top of the Terfenol-D magnetostrictive rod maintains an axial pre-compression load through a disc spring pre-compression assembly to eliminate the magnetic hysteresis dead zone. The output end of the disc spring pre-compression assembly maintains spherical contact with the valve stem end face of the lifting counterflow valve disc through a spherical force transmission pad.

4. The pressure reducing valve according to claim 1, characterized in that, The damping and stabilizing cavity is an integral component formed by selective laser melting process using 316L stainless steel powder.

5. The pressure reducing valve according to claim 1 or 4, characterized in that, The gradient density lattice structure, along the medium recirculation direction, sequentially comprises a high-density lattice section at the inlet, a transitional lattice section, and a low-density lattice section at the outlet. All three sections utilize body-centered cubic lattice units. The high-density lattice section at the inlet has lattice unit side lengths of 2 mm, lattice rod diameters of 0.8 mm, and a porosity of 40%. The transitional lattice section has lattice unit side lengths of 3 mm, lattice rod diameters of 0.6 mm, and a porosity of 58%. The low-density lattice section at the outlet has lattice unit side lengths of 4 mm and a lattice rod diameter of 0.6 mm. The rods have a diameter of 0.4 mm and a porosity of 75%. When the recirculating medium flows sequentially through the high-density lattice section at the inlet, the intermediate transition lattice section, and the low-density lattice section at the outlet, the dense rods in the high-density lattice section at the inlet force the recirculating medium to repeatedly split and merge, thereby generating microscale vortices in the rod wake region. This converts the pressure impact kinetic energy into turbulent heat dissipation energy to attenuate the pressure peak. As the rod spacing gradually increases in the intermediate transition lattice section, the microscale vortices gradually merge and attenuate, further reducing the pressure pulsation amplitude of the recirculating medium. The low-density lattice section at the outlet provides a large pore flow area, allowing the return medium to flow out smoothly at a low flow rate and merge into the upstream inlet cavity through the return channel.

6. The pressure reducing valve according to claim 1, characterized in that, The first, second, and third pressure sensors continuously collect upstream pressure, downstream pressure, and damping stabilizing chamber pressure values ​​at a sampling frequency of not less than 1 kHz and send them to the controller. In each sampling cycle, the controller performs sliding average filtering on the downstream and upstream pressure values ​​for several consecutive sampling cycles. The filtered downstream pressure value is subtracted from the filtered upstream pressure value to obtain the reverse pressure difference value. The reverse pressure difference value of the current sampling cycle is subtracted from the reverse pressure difference value of the previous sampling cycle and then divided by the sampling cycle to obtain the reverse pressure difference change rate. When the reverse pressure difference value is greater than the preset reverse flow entry threshold and the reverse pressure difference change rate is greater than zero for more than the preset entry confirmation cycle, the controller determines that the reverse flow conduction condition has been entered and generates a reverse flow start command.

7. The pressure reducing valve according to claim 6, characterized in that, The controller subtracts the filtered downstream pressure value from the preset downstream target pressure value to obtain the pressure tracking error. The pressure tracking error is multiplied by the control cycle and accumulated cycle by cycle to obtain the cumulative integral value of the pressure tracking error. The pressure tracking error is multiplied by the first sliding mode coefficient and then added to the cumulative integral value multiplied by the second sliding mode coefficient as the sliding mode switching variable.

8. The pressure reducing valve according to claim 7, characterized in that, The controller performs a superspiral second-order sliding mode operation on the sliding mode switching variable: It sets up a first superspiral state register and a second superspiral state register; within each control cycle, it takes the absolute value of the sliding mode switching variable and calculates its square root; multiplies this square root by the first superspiral gain and then by the sign value of the sliding mode switching variable to obtain the first superspiral component; it adds the current stored value of the first superspiral state register to the first superspiral component to obtain the basic control quantity and writes it to the first superspiral state register; it multiplies the sign value of the sliding mode switching variable by the second superspiral gain to obtain the second superspiral component; it adds the current stored value of the second superspiral state register to the product obtained by multiplying the second superspiral component by the control cycle and writes it to the second superspiral state register.

9. The pressure reducing valve according to claim 8, characterized in that, The controller synchronously operates a nonlinear disturbance observer, which takes the damping stabilizing chamber pressure value and the current drive current value as input. In each control cycle, the current drive current value is converted into the nominal output force according to the drive current-output force lookup table obtained in advance through static calibration tests and stored in the controller. The nominal net force is obtained by subtracting the damping stabilizing chamber pressure value from the nominal output force and multiplying it by the equivalent force area parameter obtained in advance through valve disc geometry calibration. The difference between the disturbance estimate value currently stored in the disturbance estimate register and the nominal net force is multiplied by the observer gain and then by the control cycle and accumulated in the disturbance estimate register to update the disturbance estimate value. The controller adds the disturbance estimate value to the basic control quantity to obtain the compensated total control quantity, which is then converted into a drive current command output to the solenoid excitation coil after amplitude limiting.

10. The pressure reducing valve according to claim 9, characterized in that, When the absolute value of the pressure tracking error remains within the preset steady-state error band for more than the preset number of exit confirmation cycles, the controller will reduce the drive current command to zero cycle by cycle according to the preset decreasing slope. Under the restoring force of the transmission connection structure, the lifting reverse flow valve disc will reseat on the conical sealing surface of the reverse flow valve seat, and the reverse flow bypass channel will be closed. After the reverse flow bypass channel is closed, the controller enters a preset lock monitoring period. During the lock monitoring period, the controller will only continuously collect pressure signals and will not respond to the trigger judgment of the reverse flow entry threshold. After the lock monitoring period ends, the controller will restore the reverse flow conduction condition judgment function.

Images