Intelligent flow control valve for water treatment and self-calibration method

Abstract

The application relates to the technical field of valves, in particular to an intelligent flow control valve suitable for water treatment and a self-calibration method, which comprises a valve body, a valve seat arranged at the bottom of the valve body, a water inlet and a water outlet arranged at the two sides of the valve body respectively, an electric actuator arranged at the upper part of the valve body and fixedly connected with the valve body, a motor driving shaft connected with the output end of the electric actuator, a valve core assembly comprising a valve core and a valve rod, a torque sensor arranged between the motor driving shaft and a transmission shaft or arranged at the connection position of the transmission shaft and the valve rod / valve core, and an electronic control module arranged at one side of the valve body. Compared with the prior art, the water inlet pressure port and the water outlet pressure port are directly arranged on the valve body, and the pressure is taken on site by an upstream pressure sensor and a downstream pressure sensor, so that the leakage points, installation errors and dynamic lags of external pressure taking pipelines are reduced.

Description

Technical Field

[0001] This invention relates to the field of valve technology, and in particular to an intelligent flow control valve and self-calibration method suitable for water treatment. Background Technology

[0002] In water treatment and distribution systems, valves are commonly used for constant flow water supply, process flow distribution, and pressure regulation. In existing projects, common practices include installing flow meters outside the valves as closed-loop feedback, or setting up pressure taps and differential pressure transmitters on the pipelines to calculate flow and determine operating conditions.

[0003] In existing technologies, for example, Choi proposed in his paper "Flow control system design without flowmeter sensor" (Sensors and Actuators A: Physical, 2012, 185: 127-131) to achieve flow control using only pressure gauges and valve flow coefficients (related to opening degree); this reflects the existing approach of "using pressure and valve characteristics to calculate flow rate and replace flow meters"; another example is the Chinese patent document CN119617174A, which focuses on reducing water hammer effect through a multi-stage valve closing strategy, reflecting the existing direction of "suppressing water hammer through valve closing strategy"; however, it emphasizes the valve closing control strategy itself, but consistent with traditional methods, the water treatment medium often contains flocs, sand, and other solid particles as well as air bubbles. External flow meters and external pressure tapping lines are prone to scaling, blockage, air trapping, and frequent maintenance. Furthermore, external pressure tapping points and lines can introduce installation errors, leaks, and dynamic lag, making differential pressure measurements unstable. Consequently, differential pressure-based calculations and controls are susceptible to noise interference. In addition, rapid opening and closing of valves or pump start-up and shutdown in the pipeline network can easily induce water hammer and pressure shocks. Traditional methods that rely on manually setting fixed opening and closing speeds or a single linear valve closing strategy often fail to balance response and safety under different pipeline lengths, pump operating conditions, and air content conditions. Therefore, this application discloses an intelligent flow control valve and self-calibration method suitable for water treatment. Summary of the Invention

[0004] In view of this, the purpose of this invention is to propose an intelligent flow control valve suitable for water treatment, in order to solve the problems that the presence of particles and air bubbles in the water treatment medium makes the external flow meter / pressure tapping pipeline prone to scaling and blockage, and introduces installation and dynamic errors, resulting in noise interference in differential pressure measurement. At the same time, rapid valve action or pump start-up and shutdown can easily induce water hammer, and traditional fixed opening and closing strategies are difficult to balance response and safety under varying operating conditions.

[0005] To achieve the above objectives, the present invention provides an intelligent flow control valve suitable for water treatment, comprising: a valve body and a valve seat disposed at the bottom of the valve body; an inlet and an outlet are respectively disposed on both sides of the valve body; an inlet pressure port communicating with the inlet and an outlet pressure port communicating with the outlet are respectively disposed on the valve body; an upstream pressure sensor is connected to the inlet pressure port, and a downstream pressure sensor is connected to the outlet pressure port.

[0006] An electric actuator is disposed on the upper part of the valve body and fixedly connected to the valve body. The output end of the electric actuator is connected to a motor drive shaft. The motor drive shaft is driven by a transmission shaft. The transmission shaft is used to drive the valve stem / valve core inside the valve body to move relative to the valve seat to change the valve opening.

[0007] A valve core assembly, comprising a valve core and a valve stem, wherein the valve core and a valve seat cooperate to form a variable throttling channel, and the valve stem is connected to a drive shaft to drive the valve core to move relative to the valve seat under the drive of an electric actuator, thereby changing the valve opening degree;

[0008] A torque sensor is provided, which is disposed between the motor drive shaft and the transmission shaft or at the connection between the transmission shaft and the valve stem / valve core, and is used to output a torque signal characterizing the load of the transmission chain.

[0009] An electronic control module is disposed on one side of the valve body and is electrically connected to an upstream pressure sensor, a downstream pressure sensor, a torque sensor, and an electric actuator.

[0010] The electronic control module includes a control PCB board for carrying circuit and signal lines, a microprocessor mounted on the control PCB board, a storage unit connected to the microprocessor, a pressure signal processing unit connected to the pressure sensor, and a current and torque monitoring unit connected to the electric actuator and torque sensor.

[0011] The electronic control module is configured as follows:

[0012] A. Based on the pressure difference ΔP across the valve obtained from the outputs of the upstream and downstream pressure sensors, and combined with the valve opening information, the estimated flow rate is output through the valve flow-opening model. This allows for closed-loop flow regulation without the need for an external flow meter.

[0013] B. Apply constraints or trajectory shaping to the valve opening change process based on the pressure transient characteristics output by the pressure signal processing unit to reduce the risk of water hammer;

[0014] C. Based on the actuator current and torque signals output by the current and torque monitoring unit, and combined with the valve opening and differential pressure signals, valve status indicators are generated for valve health monitoring and abnormal alarms.

[0015] Preferably, the inlet pressure port and / or outlet pressure port includes a pressure tapping channel and a buffer chamber. The pressure tapping channel is set at an angle or offset relative to the mainstream direction, and the inlet end of the pressure tapping channel is provided with one or more of a labyrinth flow resistance structure, a throttling orifice, or a filter element to reduce the deposition of solid particles into the pressure tapping channel and reduce the impact of air bubble entrainment on the stability of pressure sampling.

[0016] Preferably, the valve flow-opening model is one or more of a piecewise function model, a lookup table model, or a parameterized model; the model parameters of the valve flow-opening model include the flow coefficient Cv, the equivalent flow resistance coefficient, the local loss coefficient, or the correction coefficient; and the electronic control module stores multiple sets of model parameter sets corresponding to different medium temperature ranges, different pressure difference ranges, or different valve diameters, so as to improve the accuracy and availability of flow estimation in a wide range of water treatment conditions.

[0017] Preferably, the transient characteristics of the pressure signal of the pressure signal processing unit include the upstream pressure change rate dP_up / dt, the downstream pressure change rate dP_down / dt, the differential pressure change rate d(ΔP) / dt, the pressure fluctuation energy, or a combination thereof; the electronic control module applies an upper limit constraint on at least one of the valve opening change rate, opening acceleration, actuator output speed, or output torque to suppress water hammer or pressure shock caused by rapid valve action.

[0018] Preferably, when the electronic control module detects that the transient characteristics meet the water hammer risk criteria, it transforms the valve opening command from the original target command into a smooth trajectory command. The smooth trajectory command is one or more of the following: an S-shaped trajectory, a piecewise linear trajectory, a fifth-order polynomial trajectory, or a band-limited slope trajectory. Furthermore, the maximum slope of the smooth trajectory command is adaptively adjusted by the transient characteristic amplitude.

[0019] Preferably, the electronic control module is configured to extract torque-opening curve features or torque fluctuation spectrum features output by the torque sensor, the features including friction increment, hysteresis, jamming threshold, number of abnormal spikes or combinations thereof, and output valve health level, maintenance recommendations or protection strategies based on the features.

[0020] Preferably, the electronic control module constructs soft measurement parameters of the medium operating condition based on torque signal, actuator current signal, valve opening degree and ΔP. The soft measurement parameters include equivalent viscosity index, gas content risk index, solid content disturbance index or a combination thereof. Furthermore, the electronic control module adaptively corrects the flow closed-loop control parameters, the upper limit of the opening change rate or the water hammer risk threshold based on the soft measurement parameters.

[0021] Preferably, when any signal from the upstream pressure sensor, downstream pressure sensor, or torque sensor is detected to be abnormal, exceeding limits, or failing, the electronic control module switches to a degraded control mode. The degraded control mode includes one or more of the following: open-loop amplitude limiting control, approximate flow control based on unilateral pressure, or conservative control based on torque / current. Furthermore, the electronic control module is equipped with a communication interface with the pump station controller or frequency converter to output a linkage signal to coordinate pump speed and valve opening when water hammer risk or degraded mode is triggered.

[0022] Preferably, the pressure signal processing unit includes an analog front-end and a digital signal processing module. The analog front-end includes one or more of amplification, filtering, anti-aliasing, or isolation functions. The digital signal processing module is configured to perform sampling synchronization, noise reduction, drift compensation, and transient feature extraction.

[0023] The current and torque monitoring unit includes a current sampling circuit and a torque sampling circuit. The current sampling circuit is used to collect the drive current or equivalent load current of the electric actuator, and the torque sampling circuit is used to collect the output of the torque sensor. The microprocessor generates at least one monitoring quantity based on the drive current and torque signals. The monitoring quantity includes one or more of the following: overload criterion, jamming criterion, hysteresis criterion, or transmission abnormality criterion.

[0024] This invention also discloses a self-calibration method for an intelligent flow control valve suitable for water treatment, applied to the aforementioned intelligent flow control valve for water treatment, comprising the following steps:

[0025] S1. Acquire valve opening degree x, upstream pressure P_up, downstream pressure P_down, and torque signal T within the control cycle, and calculate the pressure difference. ;

[0026] S2. When the preset triggering conditions are met, a limited opening perturbation command is superimposed on the electric actuator (8) without changing the overall control objective, so that the valve opening changes slightly.

[0027] S3. During the perturbation process, cache the changes in the opening degree x, pressure difference ΔP, and torque signal T according to the timestamp to form a sample segment;

[0028] S4. Perform a data validity determination based on torque constraints on the sample segment. When an abnormal torque peak is detected or the torque is continuously higher than the threshold, remove the corresponding sample segment or postpone the self-calibration.

[0029] S5. Update the valve flow-opening model parameters based on the effective sample segments after elimination, and use the updated model parameters for subsequent flow estimation and control.

[0030] The beneficial effects of this invention are:

[0031] This intelligent flow control valve and self-calibration method for water treatment directly places the inlet and outlet pressure ports on the valve body, and uses upstream and downstream pressure sensors to take pressure locally, reducing leakage points, installation errors, and dynamic hysteresis in external pressure tapping pipelines. At the same time, the pressure tapping channel adopts an angle / offset setting relative to the mainstream direction, and is equipped with a labyrinth flow resistance structure, throttling orifice, or filter element with a buffer chamber at the inlet end, so that particles and air bubbles tend to pass with the mainstream rather than enter the pressure tapping port, reducing pressure drift and sampling fluctuations caused by blockage and air entrainment. This provides a stable and repeatable signal reference for the calculation of pressure difference ΔP, directly supporting the consistency of subsequent flow estimation, transient feature extraction, and health diagnosis.

[0032] The electronic control module takes the valve opening degree and pressure difference ΔP as inputs in each control cycle, and outputs a flow rate estimate through a valve flow-opening model. It performs closed-loop regulation to achieve flow control without the need for an external flow meter; the model configuration allows for the use of piecewise functions, lookup tables, or parameterization methods individually or in combination, and explicitly specifies Cv, equivalent flow resistance, local loss coefficients, or correction coefficients, making the model both engineering-ready and easy to maintain; simultaneously, it stores multiple parameter sets corresponding to different medium temperature ranges, different pressure difference ranges, or different valve diameters, selecting the parameter set according to the operating conditions during runtime, reducing the systematic bias caused by "full-range hard fitting" of a single parameter, thus... It maintains availability and stability even under conditions of fluctuating water treatment temperatures, large pressure differentials, and diverse diameters.

[0033] The transient characteristics output by the pressure signal processing unit are used to impose upper limit constraints on the rate of change of opening / acceleration, actuator speed, or output torque. When the water hammer risk criterion is met, the target command is shaped into an S-shaped, piecewise linear, fifth-order polynomial, or band-limited slope trajectory, and the maximum slope is adaptively adjusted according to the amplitude of the transient characteristics to suppress water hammer at the source. At the same time, the current and torque monitoring unit correlates the actuator current and torque signals with the opening and ΔP to generate valve status indicators, realizing health monitoring and abnormal alarm. When any key sensor signal is abnormal / failed, it switches to open-loop amplitude limiting control, single-sided pressure approximate flow control, or conservative control based on torque / current, and outputs linkage signals through the communication interface with the pump station controller / frequency converter to coordinate pump speed and valve position, improving fault tolerance and system-level safety. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a three-dimensional structural diagram of the present invention;

[0036] Figure 2 This is a schematic diagram of the workflow of the present invention;

[0037] Figure 3 This is a schematic diagram of the intelligent flow control of the present invention;

[0038] Figure 4 This is a schematic diagram of the water hammer / active damping control structure of the present invention;

[0039] Figure 5 This is a schematic diagram of the soft measurement under operating conditions of the present invention.

[0040] The diagram is marked as follows:

[0041] 1. Valve body; 2. Valve seat; 3. Inlet; 4. Outlet; 5. Inlet pressure port; 6. Outlet pressure port; 7. Upstream pressure sensor; 8. Electric actuator; 9. Motor drive shaft; 10. Transmission shaft; 11. Torque sensor; 12. Electronic control module; 13. Downstream pressure sensor. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0043] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0044] like Figures 1 to 5 As shown, an intelligent flow control valve suitable for water treatment includes a valve body 1 and a valve seat 2 located at the bottom of the valve body 1. An inlet 3 and an outlet 4 are respectively located on both sides of the valve body 1. An inlet pressure port 5 connected to the inlet 3 and an outlet pressure port 6 connected to the outlet 4 are respectively located on the valve body 1. An upstream pressure sensor 7 is connected to the inlet pressure port 5, and a downstream pressure sensor 13 is connected to the outlet pressure port 6. An electric actuator 8 is located on the upper part of the valve body 1 and fixedly connected to it. The output end of the electric actuator 8 is connected to a motor drive shaft 9, which is driven by a transmission shaft 10. The transmission shaft 10 is used to drive the valve stem / valve core inside the valve body 1 to move relative to the valve seat 2 to change the valve opening. A valve core assembly includes a valve core and a valve stem. The valve core and valve seat 2 cooperate to form a variable throttling channel. The valve stem and transmission shaft 1... The valve body 1 is connected to the electric actuator 8 to drive the valve core to move relative to the valve seat 2, thereby changing the valve opening. A torque sensor 11 is located between the motor drive shaft 9 and the transmission shaft 10, or at the connection between the transmission shaft 10 and the valve stem / valve core, and is used to output a torque signal characterizing the load on the transmission chain. An electronic control module 12 is located on one side of the valve body 1 and is electrically connected to the upstream pressure sensor 7, the downstream pressure sensor 13, the torque sensor 11, and the electric actuator 8. The electronic control module 12 includes a control PCB board for carrying circuit and signal lines, a microprocessor on the control PCB board, a storage unit connected to the microprocessor, a pressure signal processing unit connected to the pressure sensor, and a current and torque monitoring unit connected to the electric actuator 8 and the torque sensor 11.

[0045] Electronic control module 12 is configured as follows:

[0046] A. Based on the pressure difference ΔP across the valve obtained from the outputs of upstream pressure sensor 7 and downstream pressure sensor 13, and combined with the valve opening information, the estimated flow rate is output through the valve flow-opening model. This allows for closed-loop flow regulation without the need for an external flow meter.

[0047] B. Apply constraints or trajectory shaping to the valve opening change process based on the pressure transient characteristics output by the pressure signal processing unit to reduce the risk of water hammer;

[0048] C. Based on the actuator current signal and torque signal output by the current and torque monitoring unit, and combined with the valve opening and differential pressure signals, valve status indicators are generated for valve health monitoring and abnormal alarms.

[0049] The inlet pressure port 5 and outlet pressure port 6 include a pressure tapping channel and a buffer chamber. The pressure tapping channel is set at an angle or offset relative to the mainstream direction. The inlet end of the pressure tapping channel is provided with one or more of the following: a labyrinth flow resistance structure, a throttling orifice, or a filter element, to reduce the deposition of solid particles into the pressure tapping channel and to reduce the impact of air bubble entrainment on the stability of pressure sampling.

[0050] The valve body 1 has an inlet 3 and an outlet 4 on both sides. An inlet pressure port 5 and an outlet pressure port 6, connected to the inlet 3 and outlet 4 respectively, are also installed on the valve body 1. This allows the upstream pressure sensor 7 and the downstream pressure sensor 13 to take pressure locally on the valve body. The electronic control module 12 obtains the pressure difference ΔP between the two ends of the valve through the pressure tapping points within the same valve body 1. The direct effect of this "integrated pressure tapping on valve body 1 + dual pressure sensing" layout is to solidify the key process quantities (pre-valve pressure and post-valve pressure) required for flow control into the valve structure, avoiding leaks, installation errors, and hysteresis caused by external pressure tapping pipelines. Simultaneously, it facilitates subsequent flow estimation... The system provides a unified and repeatable signal reference for water hammer resistance and health diagnosis. Especially for media commonly encountered in water treatment processes, including those containing sand, flocs, small solid particles, and entrained air bubbles, the inlet pressure port 5 and outlet pressure port 6 employ a "pressure tapping channel + buffer chamber" structure. The pressure tapping channel is angled or offset relative to the mainstream direction, effectively causing particles and air bubbles to flow along the mainstream under inertia and buoyancy rather than directly into the pressure tapping inlet. This significantly reduces the probability of sediment entering the pressure tapping channel and causing blockage or zero-point drift. Simultaneously, one or more of the following are incorporated at the inlet end of the pressure tapping channel: a labyrinthine flow resistance structure, a throttling orifice, or a filter element. The flow resistance system weakens particle kinetic energy and suppresses eddy currents carrying sand by extending the flow path and multiple reversals. The throttling orifice limits the pressure sampling flow rate to a small range to reduce bubble intake and pressure pulsation amplification effects. The filter element further traps larger particles and forms a maintainable "pre-protection." Combined with the volumetric averaging effect of the buffer chamber on transient pulsations, the pressure signal becomes more stable and less susceptible to interference. The benefit is that after the pressure sampling stability is improved, the flow estimation model based on ΔP is less likely to be misled by noise. Thus, usable closed-loop flow regulation can still be achieved without installing an external flow meter, reducing the common problems of flow meter scaling / clogging / maintenance in water treatment sites. This reduces costs and decreases on-site modification space and straight pipe section requirements, making it suitable to make valves into "plug-and-play" intelligent actuators. In the actuator part, the electric actuator 8 is fixed on the upper part of the valve body 1. The output end drives the valve stem / valve core to move relative to the valve seat 2 through the motor drive shaft 9 and the transmission shaft 10 to change the valve opening. The valve core and the valve seat 2 cooperate to form a variable throttling channel. This structure ensures that the mechanical throttling characteristics of the valve are stable and can be modeled. The torque sensor 11 is arranged between the motor drive shaft 9 and the transmission shaft 10 or at the connection between the transmission shaft 10 and the valve stem / valve core, so that the torque measured by it can more directly characterize the load of the transmission chain and the change of resistance inside the valve.By introducing the torque signal and actuator current signal into the current and torque monitoring unit of the electronic control module 12, load information of the "motor input side" and "mechanical output side" can be obtained synchronously in the same control closed loop. The advantage is that it can not only identify friction increase or opening and closing abnormalities caused by valve core jamming, valve stem wear, valve seat 2 scaling, foreign object blockage, etc. earlier, but also distinguish "load increase caused by ΔP increase under normal working conditions" and "load increase caused by abnormal fault" through the correlation relationship of current-torque-opening degree-differential pressure, thereby reducing false alarms and improving the interpretability and maintainability of health monitoring.

[0051] The valve's workflow can be understood as follows: After the system is powered on, the electronic control module 12 initializes and performs self-tests on the upstream pressure sensor 7, downstream pressure sensor 13, torque sensor 11, and electric actuator 8. The pressure signal processing unit first samples, filters, and performs necessary temperature drift / zero point compensation on the two pressure channels (since the labyrinthine flow resistance / throttle orifice / filter and buffer chamber at the pressure tapping channel inlet have already suppressed air trapping and particle disturbance, the electronic filter does not need to be excessively "dulled," and can significantly reduce noise while maintaining response speed). The microprocessor calculates the pressure before and after the valve in real time and obtains the pressure difference ΔP. At the same time, it obtains the current valve opening (the position of the valve core relative to the valve seat 2) from the actuator position feedback or internal opening information. Then, it calls the pre-calibrated or self-learned valve flow-opening model in the storage unit, and substitutes ΔP and the opening into the output flow estimate. In scenarios requiring constant flow or water supply according to a process curve, the control module receives the target flow setpoint. (Can be from local settings, host computer, or process controller), will and The deviation is compared and the control algorithm outputs an opening adjustment command to the electric actuator 8, causing the motor drive shaft 9 to drive the transmission shaft 10 to push the valve stem / valve core to change the throttling channel area, thereby making the actual flow rate approach the target flow rate and completing the flow closed-loop regulation without an external flow meter. Even if the on-site flow meter causes measurement distortion due to scaling, silt, or air bubbles, the valve can still form a usable process quantity closed loop based on the pressure difference and opening, improving the continuous operation capability of the water treatment system. During the entire process of valve opening change, the pressure signal processing unit simultaneously extracts the pressure transient characteristics (e.g., ΔP). (Including mutation rate, downstream pressure fluctuation amplitude, specific frequency band pulsation energy, or rapid drop / rebound characteristics, etc.), the microprocessor uses these factors to constrain or shape the valve opening changes. For example, when a "rapid shut-off trend" that may cause water hammer is detected, the valve closing speed is automatically limited, an S-shaped acceleration / deceleration curve is used, segmented small-step approximation is employed, and a brief pause is made to release the pipeline pressure. Alternatively, when a sharp drop or oscillation in downstream pressure is detected, the valve opening is slowed down and damped control is introduced. This directly incorporates the "pressure transient risk" into the actuator command generation process. The advantage of this is that it reduces water hammer suppression from... The manual parameter tuning based on experience has been transformed into adaptive constraints based on real-time signals, ensuring consistent safety across different pipeline lengths, pump operating conditions, and gas content in different media. Simultaneously, the current and torque monitoring unit continuously collects actuator current and torque signals, and generates valve status indicators in conjunction with the current opening degree, ΔP, and the rate of opening change. For example, "torque baseline deviation during opening and closing" characterizes scaling trends, "abnormal phase / amplitude of torque and current pulsations" identifies jamming or backlash, and "abnormal torque drop under high ΔP but with pressure noise" indicates a potential problem. "Rising sound" indicates potential cavitation / flash risk or two-phase flow disturbance within the valve. "Changes in opening command with sluggish position / torque response" indicates actuator step loss or valve stem loosening. Alarms are output and events and operational data are recorded for traceability when indicators exceed limits. Because the torque sensor 11 is positioned closer to the critical location in the transmission chain, it can react more sensitively to early friction growth or foreign object intrusion. Combined with the pressure port structure ensuring pressure sampling stability, "flow control—water hammer prevention—health diagnosis" can mutually verify each other within the same hardware system: pressure increases smoothly. Credibility, Increased reliability makes control actions smoother, reducing pressure shocks at the source. Smoother control reduces mechanical shock and wear and extends the life of valve core / seat 2. Health monitoring detects abnormalities in time and avoids forced actuation in a stuck state, which could cause greater water hammer or mechanical damage.

[0052] As a supplementary approach, the flow-opening model can employ lookup table curves, piecewise polynomials, or parameterized models based on Cv / Kv characteristics. It can also store temperature / viscosity correction coefficients for different media in the storage unit or perform online fine-tuning based on operating history to compensate for changes in flow resistance caused by wear and scaling of the valve seat 2. The electronic control module 12 can also be optionally equipped with a communication interface (such as RS / Modbus, Ethernet, or wireless) for uploading data. The system includes ΔP, opening degree, torque / current status indicators, and alarm records, facilitating predictive maintenance in water plants or stations. Further enhancements include power failure protection strategies (power failure memory of current opening degree, slow closing to a safe opening degree, or mechanical return) and pressure tap maintenance strategies (replaceable filters, periodic backflushing, or self-cleaning flow field generated by short-duration opening pulses). This allows the intelligent flow control valve to achieve lower measurement and maintenance costs, higher control stability, and enhanced safety and diagnostics under complex water treatment media and long-term operating conditions.

[0053] like Figures 1 to 3 As shown, the valve flow-opening model is one or more of the following: piecewise function model, lookup table model, or parameterized model; the model parameters of the valve flow-opening model include the flow coefficient Cv, the equivalent flow resistance coefficient, the local loss coefficient, or the correction coefficient; and the electronic control module 12 stores multiple sets of model parameter sets corresponding to different medium temperature ranges, different pressure difference ranges, or different valve diameters, so as to improve the accuracy and availability of flow estimation in a wide range of water treatment conditions.

[0054] In this embodiment, the electronic control module 12 performs flow estimation and control decisions based on the "valve flow-opening model" during operation. Its core process is as follows: within each control cycle, the microprocessor uses the current valve opening information and the pressure difference ΔP across the valve as model inputs, and calls the valve flow-opening model to output the estimated flow value. To ensure the availability and accuracy of this estimation across a wide range of operating conditions in water treatment scenarios, the valve flow-opening model is not limited to a single form. Instead, it can employ one or more of the following: a piecewise function model, a lookup table model, or a parameterized model. The electronic control module 12 can select the implementation method according to engineering needs and store the corresponding data structure in the storage unit, thereby achieving a better compromise under different hardware resources, control cycles, and accuracy requirements. When using a piecewise function model, the microprocessor can quickly locate the corresponding segment according to the opening interval or pressure difference interval and perform calculations, ensuring the model remains accurate even in sections with significantly nonlinear throttling characteristics (such as small opening intervals or high pressure difference intervals). To maintain fitting accuracy and avoid systematic biases caused by using a single linear or single function, when using a lookup table model, the microprocessor can index a two-dimensional or multi-dimensional table (containing temperature or diameter dimensions) according to "current opening degree - current ΔP" and perform interpolation. This solidifies complex throttling characteristics, local loss variations, and valve core characteristics of different structures in data form, reducing online computation and improving model interpretability. This is particularly suitable for scenarios in water treatment where long-term stable operation is required and model drift is undesirable. When using a parametric model, the microprocessor can express the valve's flow characteristics within a specific range using a small number of parameters, facilitating rapid calculation even with limited storage and computing resources. Furthermore, the correction coefficient can be used as an adjustable quantity to adapt to different installation conditions or different media states; the direct effect of the above three model forms allowing "single or combined use" settings is that different models can be configured for different specifications, different controller computing power and cost constraints on the same set of valve products, or the model can be dynamically selected according to the working conditions in the same control process (e.g., priority lookup table, segmented correction for boundary working conditions, or use of parameterized model as a quick estimate and then supplemented with correction terms), thereby improving the versatility in complex water treatment systems;

[0055] Furthermore, the model parameters of the valve flow-opening model include the flow coefficient Cv, equivalent flow resistance coefficient, local loss coefficient, or correction coefficient. This setting allows the model to decompose the "valve's own throttling capacity" and "flow path resistance / local loss" in parametric form and incorporate them into the same estimation framework: in the calculation At this time, the microprocessor can map ΔP to a flow rate estimate based on the effective Cv or equivalent flow resistance corresponding to the current opening degree. Simultaneously, it can compensate for non-ideal factors through local loss coefficients or correction coefficients, avoiding deviations caused by relying solely on ideal throttling relationships. Cv expresses the valve's flow capacity at a given opening degree; the equivalent flow resistance coefficient expresses the combined resistance caused by the valve core-seat 2 throttling channel and the valve body 1 flow channel; the local loss coefficient expresses the impact of local energy losses such as at the inlet, bends, and narrowing / expanding diameters; and the correction coefficient is used to uniformly compensate for factors such as manufacturing tolerances, assembly differences, and characteristic drift caused by long-term operation. By making these parameters explicit, the electronic control module 12 outputs... At the same time, it actually establishes an "adjustable and maintainable" model framework: when the estimation has a systematic deviation under a certain working condition, the maintainer or the upper system does not need to overturn the entire model, but only needs to replace or correct a certain set of parameters to restore the accuracy; at the same time, explicit parameters also facilitate subsequent quality traceability and batch management. For example, different batches of valves of the same diameter can use different Cv curve parameter sets to reduce estimation errors caused by batch differences.

[0056] In one embodiment, the valve flow-opening model can be implemented using a standard liquid flow equation based on the valve flow coefficient: for incompressible liquids, the estimated flow rate is... It can be represented as =Cv(x)·F_c(T,μ)·√(ΔP / SG), where Cv(x) is the flow coefficient corresponding to the valve opening x, ΔP is the pressure difference across the valve, SG is the relative density (SG=ρ / ρ0, ρ0 is the reference density), and F_c(T,μ) is the medium temperature / viscosity correction factor (used to correct for density changes caused by temperature and the Reynolds number effect under high viscosity), and can be converted according to engineering units. The output is m³ / h or L / s; preferably, the electronic control module stores the Cv(x) curve in a data structure of "discrete opening points + parameter table": the opening interval [0, 100%] is discretized into x_i at a preset resolution (e.g., 1% or 2% step), and the Cv_i corresponding to each discrete point and the optional set of correction coefficients (e.g., Cv_i(T_j, DN_m) grouped by temperature interval, diameter, or pressure difference interval) are stored in the storage unit. During operation, linear / spline interpolation is performed on the current opening x to obtain Cv(x) and it is substituted into the above equation for output. Furthermore, the Cv(x) curve can be obtained through factory calibration: the valve is subjected to multi-point opening gradient sweep on a standard test bench, and reference flow rate Q_ref and pressure difference ΔP are collected at each opening point. Cv_i is back-calculated according to Cv_i=Q_ref·√(SG / ΔP) / F_c(T,μ) and smoothed (e.g., monotonic constraint fitting to ensure that Cv does not decrease with opening). The fitted Cv table is written into the storage unit. When deployed in the field, a first-order scaling method is allowed for rapid recalibration (e.g., the scaling factor s is calculated based on 1~2 stable operating points so that Cv(x)=s·Cv0(x)) to compensate for the overall deviation caused by installation conditions and long-term wear.

[0057] To adapt to the characteristics of water treatment operations, such as "temperature variations, large differential pressure ranges, and diverse valve diameters," the electronic control module 12 stores multiple sets of model parameter sets corresponding to different medium temperature ranges, different differential pressure ranges, or different valve diameters. During operation, it completes a closed loop of "parameter set selection—model call—flow estimation": When the control cycle begins, the microprocessor first determines the current differential pressure range based on the current ΔP, then determines the parameter family corresponding to the valve diameter based on the valve diameter information (this diameter information can be stored as equipment configuration parameters in the storage unit), and can select the corresponding temperature range parameter set according to the medium temperature range. Subsequently, the selected parameter set is loaded into the model calculation path and output. The benefit of this "parameter set switching by interval / diameter" process is that it covers the nonlinearity and variability under a wide range of operating conditions through a discretized parameter set. This eliminates the need for estimation to rely on hard fitting of a single parameter across the entire range, thus significantly improving the accuracy and availability of flow estimation. It also avoids closed-loop regulation instability or slow response caused by excessive estimation deviations under conditions of low or high pressure differentials, low or high temperatures, or small or large diameters. Especially in water treatment systems, where pipeline resistance varies with operating conditions, valves may operate at different opening ranges for extended periods, and pressure differentials may jump due to pump speed or branch switching, the multi-parameter set mechanism ensures the system remains continuously available even when operating conditions vary. This output ensures that closed-loop control does not become "blind" or frequently misjudge due to sudden changes in operating conditions. Furthermore, when a lookup table model or piecewise function model is combined with a multi-parameter set, the microprocessor can directly look up the table or calculate segment by segment after selecting the parameter set, which reduces the complexity of online calculations and improves real-time performance, making the control cycle more stable, facilitating smoother valve actions and reducing control jitter. When a parameterized model is combined with a multi-parameter set, different intervals can be quickly adapted by switching parameter vectors, avoiding the computational burden caused by using complex high-order models over a wide range.

[0058] The aforementioned model format and parameter set storage configuration also bring additional value in terms of "engineering implementation and maintenance": On the one hand, the model and parameters are stored as data in the storage unit of the electronic control module 12, allowing valves to be pre-configured with corresponding parameter sets for different diameters, structures, or application scenarios at the factory, enabling adaptation without hardware changes after on-site installation; on the other hand, when the operating conditions of the water treatment system or the valve's operating environment change (e.g., the same valve is reused in different pipe sections), the estimated performance can be restored by replacing the parameter sets corresponding to the temperature range, pressure difference range, or diameter, thereby reducing on-site commissioning costs and improving deployment efficiency; through the above process, the electronic control module 12 can stably obtain [data / values] in each control cycle. Furthermore, by maintaining the continuity and consistency of estimation under a wide range of operating conditions, the "closed-loop flow regulation without external flowmeters" becomes feasible and engineering stable. The optional / combinable implementation methods of piecewise functions, lookup tables, or parameterized models, as well as the parameter system and multi-parameter set management mechanism that includes Cv, equivalent flow resistance, local loss, and correction coefficients, together constitute the key support for improving the accuracy and usability of flow estimation in water treatment scenarios.

[0059] like Figure 1 , Figure 2 , Figure 4 As shown, the transient characteristics of the pressure signal of the pressure signal processing unit include the upstream pressure change rate dP_up / dt, the downstream pressure change rate dP_down / dt, the differential pressure change rate d(ΔP) / dt, the pressure fluctuation energy, or a combination thereof; the electronic control module 12 applies an upper limit constraint to at least one of the valve opening change rate, opening acceleration, actuator output speed, or output torque to suppress water hammer or pressure shock caused by rapid valve action; when the electronic control module 12 detects that the transient characteristics meet the water hammer risk criterion, it reshapes the valve opening command from the original target command into a smooth trajectory command, which is one or more of the following: S-shaped trajectory, piecewise linear trajectory, fifth-order polynomial trajectory, or band-limited slope trajectory; and the maximum slope of the smooth trajectory command is adaptively adjusted by the transient characteristic amplitude;

[0060] The electronic control module 12 synchronously acquires the upstream and downstream pressures and forms a pressure difference ΔP in each control cycle. After pressure signal processing, it further calculates the transient characteristics of the pressure signal from the pressure signal processing unit, including the upstream pressure change rate dP_up / dt, the downstream pressure change rate dP_down / dt, the pressure difference change rate d(ΔP) / dt, the pressure fluctuation energy, or a combination thereof. The introduction of these transient characteristics enables the system to not only "see the pressure value" but also "see the speed and intensity of pressure changes," thereby capturing the precursors before water hammer actually forms and propagates, avoiding the hysteresis suppression caused by traditional adjustments based solely on steady-state pressure or opening error. Based on the aforementioned transient characteristics, the electronic control module 12 first sets a boundary for the valve's action capability, that is, it applies an upper limit constraint to at least one of the valve opening change rate, opening acceleration, actuator output speed, or output torque, so that even if the upper-level control or external command gives a large opening step, the actual valve execution is still limited to a "safe action envelope" that will not easily trigger pressure shock. The purpose of doing so is to change the water hammer risk from "post-event remedy" to "pre-action constraint", directly reducing the possibility of pressure difference change and pressure wave propagation along the pipeline caused by rapid valve opening and closing. At the same time, it also avoids excessive mechanical shock and load peaks generated by the actuator under high pressure difference conditions due to following rapid commands, thereby improving the life of the valve core-valve seat 2 mating surface and the stability of the transmission chain.

[0061] Subsequently, the electronic control module 12 compares the calculated transient characteristics with the preset water hammer risk criteria. When the transient characteristics are detected to meet the water hammer risk criteria, instead of simply "cropping" or "freezing" the opening command, it reshapes the valve opening command from the original target command into a smooth trajectory command, so that the transition of the opening from the current value to the target value is a continuous and controllable change process. The smooth trajectory command can be selected as one or more of the following: S-shaped trajectory, piecewise linear trajectory, fifth-order polynomial trajectory, or limited slope trajectory. This multi-trajectory selection setting allows the system to adopt different shaping methods according to different field response characteristics and actuator capabilities. For example, when a smoother start and stop is required, an S-shaped or fifth-order polynomial trajectory is used to reduce acceleration abrupt changes. When it is easy to implement and verify in engineering, a piecewise linear or limited slope trajectory is used to constrain the slope and response time, thereby balancing implementation difficulty and adjustment efficiency without sacrificing controllability.

[0062] Meanwhile, the maximum slope of the smooth trajectory command is adaptively adjusted by the transient characteristic amplitude, enabling trajectory shaping to adapt to changing conditions: when dP_up / dt, dP_down / dt, d(ΔP) / dt, or pressure fluctuation energy is small and the system is in a mild operating condition, a larger maximum slope is allowed to ensure valve response speed and flow control performance; when the transient characteristic amplitude increases, indicating that the pipeline network is more sensitive to valve action or that a pressure fluctuation trend has appeared, the maximum slope is automatically reduced and the transition time is extended, making the valve action slower and thus bringing the pressure change rate back to the safe range. The advantage of this adaptive mechanism is that it eliminates the need for manual selection of a conservative value between "fast response" and "safer," allowing the valve to maintain efficiency under most normal operating conditions and automatically switch to a more conservative action mode under high-risk conditions, thus balancing process stability and pipeline safety. The complete workflow is as follows: the electronic control module 12 continuously calculates transient characteristics → sets upper limits for the rate of change of opening / acceleration / speed / torque based on the transient characteristics → when the criterion is triggered, the original target opening command is shaped into a trajectory and a smooth trajectory is generated → the maximum slope of the trajectory is adjusted in real time based on the amplitude of the transient characteristics → the shaped command is output to the actuator for execution, and the transient characteristics are recalculated using the updated pressure signal in the next cycle to form a closed loop. In this process, transient characteristics provide a basis for risk quantification, upper limit constraints provide a "hard boundary" for the action, trajectory shaping provides a "soft transition" for the action, and adaptive slope provides "elastic adjustment" according to the working conditions. The combination of these four factors enables the valve in the water treatment system to significantly reduce water hammer or pressure shock caused by rapid action, while maintaining the highest possible regulation efficiency and stability under the premise of safety.

[0063] In this embodiment, the electronic control module samples the upstream pressure P_up(k), downstream pressure P_down(k), and valve opening x(k) at a fixed control period Δt, where k is the sampling number, and the pressure difference ΔP(k) is defined as P_up(k) minus P_down(k). To characterize the transient changes related to water hammer, the electronic control module calculates the rate of change of pressure and the rate of change of pressure difference: the rate of change of upstream pressure is approximately equal to... The rate of change of downstream pressure is approximately equal to The rate of change of pressure difference is approximately equal to Simultaneously, to reflect the intensity of pressure fluctuations within a certain time window, the electronic control module calculates the pressure fluctuation energy E(k) within a sliding window of length N, where E(k) is equal to the energy from i=0 to... Summation: The square of ; ΔP_avg(k) is the mean pressure difference within the window, equal to Multiply by from i=0 to right Summation of .

[0064] In this embodiment, the water hammer risk criterion is given using a multi-feature fusion approach. Thresholds A_up, A_down, A_Δ, and A_E are pre-set, and a normalized risk factor R(k) is calculated, where R(k) takes the maximum value of the following four ratios: the first ratio is the upstream pressure change rate. The second ratio is the rate of change of downstream pressure. The third ratio is the rate of change of pressure difference. The fourth ratio is When R(k) is greater than or equal to 1, the system is considered to be in a water hammer risk state. To avoid frequent switching near the threshold, an exit threshold R_out (e.g., 0.8) can be optionally set. When R(k) is less than or equal to R_out, the system exits the water hammer risk state. The thresholds A_up, A_down, A_Δ, and A_E can be determined by the upper limit of the allowable pressure change of the equipment, on-site commissioning test data, or process safety specifications, and can be stored as a parameter set for use under different operating conditions.

[0065] When the water hammer risk state has not yet been reached, the electronic control module issues target opening commands. (k) Perform routine amplitude and speed limiting processing, such as applying upper limit constraints to the opening change rate dx / dt, opening acceleration d2x / dt2, actuator speed, or output torque before outputting to the electric actuator; when entering the water hammer risk state, the electronic control module performs trajectory shaping on the target opening command to reduce the impact of opening and closing actions on the pipeline. Specifically, first set the basic maximum opening slope v0 and minimum movable slope v_min under the risk-free state, and adaptively determine the maximum allowable slope v_max(k) under the risk state according to the risk factor R(k), for example, first calculate the temporary value. , where α is the sensitivity coefficient; then, v_tmp(k) is limited so that v_max(k) is not greater than v0 and not less than v_min, so that v_max(k) decreases as R(k) increases.

[0066] In this embodiment, trajectory shaping employs a fifth-order polynomial to smooth the trajectory. Let the current opening be x0, and the target opening after shaping be... ,make And define the normalized time τ = t / T, where t is the trajectory running time and T is the total trajectory duration, then output the reference trajectory within the time interval t from 0 to T:

[0067]

[0068] The velocity and acceleration of this five-fold trajectory are zero at the start and end points, which reduces abrupt changes in the opening command. To satisfy the maximum slope constraint, the maximum slope approximation relationship of this trajectory can be used:

[0069]

[0070] Therefore, T is chosen to satisfy:

[0071]

[0072] The electronic control module discretely outputs x_ref(t) as x_ref(k) according to the control cycle and sends it to the electric actuator. This automatically slows down the valve action and achieves smooth opening and closing when a rapid pressure change or increased pressure fluctuation is detected, reducing the risk of water hammer. Optionally, when the risk condition persists and R(k) continues to increase, the electronic control module can further reduce v_max(k) or extend T to achieve stronger water hammer protection.

[0073] To ensure the real-time performance and noise immunity of transient feature extraction, the control period Δt is selected as 5ms to 20ms (corresponding to a sampling frequency of 50Hz to 200Hz) and upstream and downstream pressure sampling is synchronized. The sliding window length N and Δt together determine the feature extraction window duration Tw = N·Δt, with Tw preferably being 0.5s to 3s to cover the main pressure fluctuation process caused by valve action. The pressure change rate dP / dt in the transient features can be obtained by the first-order difference within the window and combined with digital low-pass filtering or Savitzky-Golay smoothing to suppress measurement noise. The pressure fluctuation energy E(k) and mean ΔP_avg(k) are used to characterize the "fluctuation intensity" and "steady-state bias". The water hammer risk thresholds A_up, A_down, A_Δ, and A_E can be set using a combination of "safety upper limit + adaptive noise margin": when the valve's stable operation conditions are met, the baseline mean μ and standard deviation σ of each feature are statistically analyzed, and the threshold is set to max(process allowable upper limit). μ+β·σ), where β is preferably 3 to 6 to suppress false triggering, and a hysteresis strategy with entry threshold R_in and exit threshold R_out is adopted to avoid frequent switching; when entering the water hammer risk state, trajectory shaping not only restricts the opening change rate (slope) v_max, but also restricts the opening acceleration a_max and the jump j_max (i.e. the opening acceleration change rate). For example, when generating S-shaped / fifth-order polynomial trajectories, it explicitly satisfies |dx / dt|≤v_max, |d²x / dt²|≤a_max, and |d³x / dt³|≤j_max, thereby further reducing the impact of command mutation on the pipeline while ensuring controllable response time, and making the anti-water hammer behavior under different pipeline lengths, different diameters and different pump valve combinations have configurable and verifiable engineering boundaries.

[0074] like Figure 1 , Figure 2 , Figure 4 As shown, Figure 5The electronic control module 12 is configured to extract the torque-opening curve characteristics or torque fluctuation spectrum characteristics output by the torque sensor 11. The characteristics include friction increment, hysteresis, jamming threshold, number of abnormal spikes or combinations thereof, and output valve health level, maintenance suggestions or protection strategies based on the characteristics. The electronic control module 12 constructs soft measurement quantities of the medium operating conditions based on the torque signal, actuator current signal, valve opening and ΔP. The soft measurement quantities include equivalent viscosity index, gas content risk index, solid content disturbance index or combinations thereof. Furthermore, the electronic control module 12 adaptively corrects the flow closed-loop control parameters, the upper limit of the opening change rate or the water hammer risk threshold based on the soft measurement quantities.

[0075] In this embodiment, while the valve is operating normally, the electronic control module 12 continuously and synchronously acquires the torque signal output by the torque sensor 11, the actuator current signal, the valve opening, and the pressure difference ΔP across the valve. These quantities are then input as "state observations" under the same time reference into the feature extraction and soft measurement calculation process. Specifically, the module first maps the torque signal with the valve opening as the independent variable to obtain a torque-opening curve, from which it extracts features such as friction increment, hysteresis, jamming threshold, and the number of abnormal spikes. Alternatively, it performs frequency domain / spectral feature analysis on the torque change over time to obtain torque fluctuation spectrum features. The direct effect of this setup is to transform the originally difficult-to-quantify mechanical health state into calculable and comparable indicators, such as the friction increment reflecting the valve core / The gradual increase in valve stem / seal or transmission chain resistance, hysteresis, and other characteristics can indicate different responses to the same command caused by transmission clearance, elastic deformation, or frictional hysteresis. The jamming threshold and the number of abnormal spikes can capture instantaneous load changes caused by foreign object clamping, scaling jamming, or abnormal meshing. These characteristics can form trend judgments during operation without additional disassembly and inspection. Based on this, the electronic control module 12 can output the valve health level and provide more on-site maintenance suggestions (such as prompting cleaning, checking seals, or checking the transmission mechanism) or protection strategies (such as limiting opening changes, reducing the frequency of operation, or switching to a more conservative operation mode). The advantage is that maintenance is shifted from "post-fault repair" to "trend warning", reducing the risk of sudden jamming, malfunction, and the resulting process fluctuations and shocks.

[0076] Based on feature extraction, the electronic control module 12 further utilizes the load characterization capabilities of torque and actuator current signals, combined with the operating condition characterization capabilities of opening degree and ΔP, to construct a soft measurement of media operating conditions and output equivalent viscosity index, gas content risk index, and solid content disturbance index. Essentially, this process treats the response changes of the valve as an actuator and throttling element as an indirect measurement of the media and pipeline state: when the equivalent viscosity of the medium increases or the solid content disturbance increases, the valve often exhibits different load characteristics and fluctuation patterns under the same opening degree and pressure difference conditions. The combination of torque / current and ΔP / opening degree can form a sensitive quantity for these changes; when the gas content increases, the fluctuation patterns of pressure and load also show typical changes, and the soft measurement can condense these changes into a risk index for decision-making. The purpose of this is that even without additional viscometers, gas content sensors, or online solid content instruments, the system can still obtain "usable prompts" for changes in media state, thereby improving control robustness in scenarios with large fluctuations in water treatment operating conditions and avoiding misjudging media changes as valve failures or valve load changes as simple control errors.

[0077] After the soft measurement is output, the electronic control module 12 uses it to adaptively correct the flow closed-loop control parameters, the upper limit of the opening change rate, or the water hammer risk threshold, forming a closed loop of "observation-evaluation-correction-reoperation". For example, when the equivalent viscosity index increases or the solid content disturbance index increases, the system can adjust the closed-loop control parameters to be more robust and appropriately tighten the upper limit of the opening change rate to reduce overshoot and oscillation caused by changes in medium resistance. When the gas content risk index increases or the pressure fluctuation pattern becomes more sensitive, the water hammer risk threshold can be reduced accordingly or a more conservative trajectory shaping can be triggered in advance to reduce the probability of pressure shock. The advantage of this setting is that the control strategy is no longer a fixed threshold and fixed parameters, but automatically "becomes more suitable" as the medium and load state change. It can maintain the regulation efficiency under normal operating conditions and actively tend to be conservative when the risk increases, thereby truly transforming the results of health diagnosis and operating condition identification into safety and stability benefits on the control side, and improving the availability and reliability of the valve under long-term operation, complex media, and variable pipeline network conditions.

[0078] like Figure 1 , Figure 2 , Figure 5 As shown, when any signal from the upstream pressure sensor 7, downstream pressure sensor 13, or torque sensor 11 is detected to be abnormal, exceeds the limit, or fails, the electronic control module 12 switches to a degraded control mode. The degraded control mode includes one or more of the following: open-loop amplitude limiting control, approximate flow control based on unilateral pressure, or conservative control based on torque / current. Furthermore, the electronic control module 12 is equipped with a communication interface with the pump station controller or frequency converter to output a linkage signal to coordinate the pump speed and valve opening when water hammer risk or degraded mode is triggered.

[0079] In this embodiment, while the valve is operating under normal closed-loop regulation, the electronic control module 12 continuously monitors the signals from the upstream pressure sensor 7, the downstream pressure sensor 13, and the torque sensor 11. When any signal becomes abnormal, exceeds the limit, or fails, the control process will not be directly interrupted or continue to use the original closed-loop algorithm for "blind control." Instead, it will first enter the fault identification and safety judgment logic and immediately switch to the degraded control mode. This setting aims to bring unavoidable sensor failures, line interference, short-term distortion, and other situations into the controllable range, avoiding the amplification of opening commands due to incorrect input data, which could lead to valve malfunctions or pressure surges. This ensures that the system can still maintain predictable and constrained behavior when critical measurements are unavailable. After entering the degraded control mode, the electronic control module 12 can select one or more of the following based on the remaining available signals: open-loop limiting control, approximate flow control based on unilateral pressure, or conservative control based on torque / current. When pressure signals are unavailable or unreliable, open-loop limiting control is preferred. The valve opening is limited to a preset safe range and the rate of change of opening is restricted, keeping the valve in a relatively conservative throttling state to reduce the risk of sudden pressure changes and water hammer in the pipeline network. When only one side of the pressure is still available, approximate flow control based on the single-side pressure is adopted, so that the valve can still make approximate adjustments to the flow rate or at least maintain the flow rate within an acceptable range without relying on the complete pressure difference ΔP, thereby reducing the degree of process deviation. When the pressure signal is unreliable but the torque / current on the actuator side is still available, conservative position control based on torque / current is adopted. By using torque or current as a load constraint, the valve is prevented from continuing to force operation under high load conditions, which would lead to increased jamming, transmission shock, or damage to the valve core and valve seat. Since the three degradation modes essentially correspond to different scenarios of "lack of observables", the modules can be selected or combined as needed to maximize the use of still reliable observables, so that the valve can shift from "pursuing precise flow control" to "prioritizing safety and controllability" in the event of a failure. The benefits are reduced failure propagation and secondary damage, and improved system fault tolerance and continuous operation capability.

[0080] When the degradation mode is triggered or water hammer risk is detected, the electronic control module 12 outputs a linkage signal through the communication interface with the pump station controller or frequency converter to coordinate the pump speed and valve opening to form a system-level collaborative control process. That is, once the valve side enters a more conservative opening strategy (such as limiting the position or reducing the opening change rate), it simultaneously sends a corresponding linkage command or status signal to the pump station side, causing the pump station controller / frequency converter to adjust the pump speed, reduce the flow step, or reduce the pressure rise rate, thereby reducing pressure fluctuations from the source. Conversely, when changes in the operating conditions of the pump station side may lead to sudden changes in differential pressure or an increase in water hammer risk, the valve side can also enter a more stable mode in advance through this interface. The sliding action mode or a more conservative opening constraint; the purpose of this setting is to extend the safety control of a single valve to a holistic safety strategy of "pump-valve linkage", avoiding the inability of the valve itself to offset the pressure shock caused by sudden changes in pump speed when the valve speed limit alone is insufficient. At the same time, in the event of sensor failure, the probability of extreme operating conditions is reduced by pump-side cooperation, making it easier to maintain system stability through degraded control. Overall, this process ensures that when critical sensor signals are abnormal or water hammer risks occur, the system can quickly and clearly switch from the normal closed loop to a controllable safety mode, and control pressure shocks and process disturbances within acceptable ranges through pump speed and valve position coordination.

[0081] like Figure 1 , Figure 2 As shown, the pressure signal processing unit includes an analog front-end and a digital signal processing module. The analog front-end includes one or more of amplification, filtering, anti-aliasing, or isolation functions. The digital signal processing module is configured to perform sampling synchronization, noise reduction, drift compensation, and transient feature extraction.

[0082] The current and torque monitoring unit includes a current sampling circuit and a torque sampling circuit. The current sampling circuit is used to collect the drive current or equivalent load current of the electric actuator 8, and the torque sampling circuit is used to collect the output of the torque sensor 11. The microprocessor generates at least one monitoring quantity based on the drive current and torque signals. The monitoring quantity includes one or more of the following: overload criterion, jamming criterion, hysteresis criterion, or transmission abnormality criterion.

[0083] The pressure signal processing unit adopts a combined architecture of "analog front-end + digital signal processing module". The workflow begins when the sensor signal enters the electronic control module 12: the analog signals from the upstream pressure sensor 7 and the downstream pressure sensor 13 first enter the analog front-end. The analog front-end amplifies the signal according to the sensor output amplitude and the system sampling range, so that the pressure signal can make full use of the sampling range and improve the effective resolution. At the same time, it filters and suppresses power frequency interference, high-frequency noise, and spikes caused by valve action. When necessary, anti-aliasing processing is introduced to prevent high-frequency components from folding to low frequencies after sampling and causing spurious fluctuations. For the long cables, high-power equipment, and frequency converter interference environment commonly found in water treatment sites, the isolation measures of the analog front-end can reduce the impact of ground loops and electromagnetic interference on the sampling baseline, making the pressure signal "cleaner and usable". After the pressure signal from the analog front-end enters the digital signal processing module, the digital module... Sampling synchronization is performed at fixed intervals to ensure that upstream and downstream pressures are aligned on the same time base, thus preventing errors in subsequent pressure difference ΔP and rate of change calculations due to phase differences. Noise reduction and drift compensation are performed on top of synchronization to avoid sensor zero-point drift, temperature drift, or long-term offset from deflecting the control system. Simultaneously, the processed pressure sequence is used for transient feature extraction, enabling the system to stably obtain characteristic quantities such as dP / dt, d(ΔP) / dt, or pressure fluctuation energy, providing reliable input for water hammer risk assessment and opening trajectory shaping. The advantage of this setup is that it moves the "pressure signal quality" problem forward to the hardware and underlying algorithm level, preventing upper-level estimation flow control and water hammer suppression from frequent false triggers or adjustments due to random noise, sampling asynchrony, or drift. This improves control stability and judgment consistency, making it particularly suitable for water treatment systems operating long-term, in environments with strong interference, and with long maintenance cycles.

[0084] In parallel with the pressure link, the current and torque monitoring unit monitors the actuator load in real time through current sampling circuits and torque sampling circuits: the current sampling circuit collects the drive current or equivalent load current of the electric actuator 8, and the torque sampling circuit collects the output of the torque sensor 11. These two signals enter the microprocessor and are used to generate at least one monitoring quantity, including one or more of the following: overload criterion, jamming criterion, hysteresis criterion, or transmission anomaly criterion. Its workflow is "sampling—characterization—criteria generation—output monitoring quantity": after collecting the current and torque, the microprocessor can calculate the peak value, average value, rate of change, number of spikes, or the correspondence with the opening degree change, and compare them with preset thresholds or logical conditions to form a criterion output. For example, if the drive current or torque is continuously higher than the normal range, the overload criterion is met; if the current / torque shows a significant lag or jump when the opening degree command changes, the jamming criterion is met; and if the current during the opening degree reciprocation process... A significant asymmetry or cumulative difference in torque satisfies the hysteresis criterion, while a mismatch in the relationship between current and torque, or the occurrence of periodic abnormal fluctuations, indicates a transmission anomaly. This configuration transforms the operating status of the actuator and transmission chain into quantifiable and alarm-enabled monitoring quantities. This allows the system to identify and implement protective measures in advance when the valve is still operational but exhibits an "abnormal load trend," preventing it from developing into jamming, burnout, or mechanical damage. Simultaneously, current and torque are two observations of the same load, mutually verifying each other and reducing the probability of misjudgment from a single measurement, making monitoring more reliable. Since these monitoring quantities originate directly from the sampling circuit and the microprocessor's local computation, without relying on external diagnostic equipment, they can output health status and risk warnings in real time during the continuous operation of the water treatment valve. This provides a basis for subsequent limiting control, deceleration, or maintenance decisions, thereby comprehensively improving the safety, maintainability, and long-term stable operation capability of the valve system.

[0085] like Figures 1 to 3 As shown, a self-calibration method for an intelligent flow control valve suitable for water treatment, applied to the aforementioned intelligent flow control valve for water treatment, includes the following steps:

[0086] S1. Acquire valve opening degree x, upstream pressure P_up, downstream pressure P_down, and torque signal T within the control cycle, and calculate the pressure difference. ;

[0087] S2. When the preset triggering conditions are met, a limited opening perturbation command is superimposed on the electric actuator (8) without changing the overall control objective, so that the valve opening changes slightly.

[0088] S3. During the perturbation process, cache the changes in the opening degree x, pressure difference ΔP, and torque signal T according to the timestamp to form a sample segment;

[0089] S4. Perform a data validity determination based on torque constraints on the sample segment. When an abnormal torque peak is detected or the torque is continuously higher than the threshold, remove the corresponding sample segment or postpone the self-calibration.

[0090] S5. Update the valve flow-opening model parameters based on the effective sample segments after elimination, and use the updated model parameters for subsequent flow estimation and control.

[0091] In the online self-calibration embodiment, when the valve enters a relatively stable operating window or reaches a preset calibration interval, the electronic control module superimposes a small opening perturbation within a safe amplitude onto the actuator, causing the valve core to produce a tiny opening change with minimal impact on the process. The cache unit records the dynamic trajectory of the opening, pressure difference ΔP, and torque signal before and after the perturbation in a timestamp-aligned manner. After the perturbation ends, the microprocessor first filters the validity of the samples based on the torque constraint. When the torque shows an abnormal peak or is continuously high, indicating valve jamming / foreign object clamping / abnormal load, the corresponding data segment is removed or the calibration is postponed to avoid "calibrating with defects". Subsequently, only the valid samples that have passed the screening are used to perform small-step updates on the flow-opening model parameters, and the updated parameter set is written back to the storage unit for subsequent flow estimation and closed-loop regulation, thereby maintaining the long-term availability and stability of the model without adding an external flow meter.

[0092] In one embodiment, "updating the valve flow-opening model parameters" can be achieved using weighted least squares or recursive least squares (RLS): the valve model parameters are denoted as θ (e.g., including the overall scaling factor s of the Cv curve, the zero-point offset b, or several coefficients from piecewise polynomial / table lookup interpolation), and an objective function is constructed using the sample segment as the dataset. Where f(x,ΔP;θ) is the flow-opening model, w_i is the weight (higher weights can be assigned to samples with more stable torque and lower pressure noise); Q_ref can be the target flow command value when the system is stable in the closed loop and the deviation converges (when the closed loop meets the steady-state criterion, Q≈Q_set is considered, and Q_set is used as the "pseudo-true value" for online calibration), or the flow estimate obtained by converting the pump speed and pump characteristic curve when pump station controller / frequency converter data is available can be used as a reference value to achieve online identification without external flow meters; parameter updates adopt "small steps + projection constraints" to achieve the following: To ensure stability and physical plausibility: For example, θ_new = Proj(θ_old + η·Δθ), where η is the update step size with an upper limit (to avoid model abrupt changes caused by a single calibration), and Proj(·) is used to force Cv(x) to be monotonically non-decreasing with opening degree and for parameters to fall within a preset physical range. Updates are paused when the residual of a sample segment exceeds a threshold or the torque criterion indicates jamming / abnormal friction. Furthermore, a forgetting factor λ (0.95–0.995) can be used to progressively track long-term drift, and a "consistency check before and after update" (such as for the same opening degree interval) can be performed. (Set an upper limit on the amount of change) to prevent convergence and divergence, so that the model remains usable for a long time under conditions of wear, scaling and media changes.

[0093] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in the details for the sake of brevity.

[0094] This invention is intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An intelligent flow control valve suitable for water treatment, characterized in that, include: The valve body (1) and the valve seat (2) are located at the bottom of the valve body (1). The valve body (1) is provided with an inlet (3) and an outlet (4) on both sides. The valve body (1) is provided with an inlet pressure port (5) connected to the inlet (3) and an outlet pressure port (6) connected to the outlet (4). The inlet pressure port (5) is connected to an upstream pressure sensor (7), and the outlet pressure port (6) is connected to a downstream pressure sensor (13). An electric actuator (8) is disposed on the upper part of the valve body (1) and fixedly connected to the valve body (1). The output end of the electric actuator (8) is connected to a motor drive shaft (9). The motor drive shaft (9) is connected to a transmission shaft (10). The transmission shaft (10) is used to drive the valve stem / valve core inside the valve body (1) to move relative to the valve seat (2) to change the valve opening. The valve core assembly includes a valve core and a valve stem. The valve core and valve seat (2) cooperate to form a variable throttling channel. The valve stem is connected to a drive shaft (10) to drive the valve core to move relative to the valve seat (2) under the drive of an electric actuator (8), thereby changing the valve opening. A torque sensor (11) is provided between the motor drive shaft (9) and the transmission shaft (10) or at the connection between the transmission shaft (10) and the valve stem / valve core, and is used to output a torque signal characterizing the load of the transmission chain. Electronic control module (12) is disposed on one side of the valve body (1) and is electrically connected to upstream pressure sensor (7), downstream pressure sensor (13), torque sensor (11) and electric actuator (8). The electronic control module (12) includes a control PCB board for carrying circuit and signal lines, a microprocessor mounted on the control PCB board, a storage unit connected to the microprocessor, a pressure signal processing unit connected to the pressure sensor, and a current and torque monitoring unit connected to the electric actuator (8) and the torque sensor (11). The electronic control module (12) is configured as follows: A. Based on the output of the upstream pressure sensor (7) and the downstream pressure sensor (13), the pressure difference ΔP between the two ends of the valve is obtained, and combined with the valve opening information, the flow rate estimate is output through the valve flow-opening model. This allows for closed-loop flow regulation without the need for an external flow meter. B. Apply constraints or trajectory shaping to the valve opening change process based on the pressure transient characteristics output by the pressure signal processing unit to reduce the risk of water hammer; C. Based on the actuator current signal and torque signal output by the current and torque monitoring unit, and combined with the valve opening and differential pressure signals, valve status indicators are generated for valve health monitoring and abnormal alarms. The inlet pressure port (5) and / or outlet pressure port (6) include a pressure tapping channel and a buffer chamber. The pressure tapping channel is set at an angle relative to the mainstream direction, and the inlet end of the pressure tapping channel is provided with one or more of a labyrinth flow resistance structure, a throttling orifice or a filter element to reduce the deposition of solid particles into the pressure tapping channel and reduce the impact of bubble entrainment on the stability of pressure sampling. When the electronic control module (12) detects that the transient characteristics meet the water hammer risk criteria, it transforms the valve opening command from the original target command into a smooth trajectory command. The smooth trajectory command is one or more of the following: S-shaped trajectory, piecewise linear trajectory, fifth-order polynomial trajectory, or band-limited slope trajectory. The maximum slope of the smooth trajectory command is adaptively adjusted by the transient characteristic amplitude. The electronic control module (12) is configured to extract the torque-opening curve features or torque fluctuation spectrum features output by the torque sensor (11), the features including friction increment, hysteresis, jamming threshold, number of abnormal spikes or combinations thereof, and output valve health level, maintenance recommendations or protection strategies based on the features. The pressure signal processing unit includes an analog front end and a digital signal processing module. The analog front end includes one or more of amplification, filtering, anti-aliasing or isolation. The digital signal processing module is configured to perform sampling synchronization, noise reduction, drift compensation and transient feature extraction.

2. The intelligent flow control valve for water treatment according to claim 1, characterized in that, The valve flow-opening model is one or more of a piecewise function model, a lookup table model, or a parameterized model; the model parameters of the valve flow-opening model include the flow coefficient Cv, the equivalent flow resistance coefficient, the local loss coefficient, or the correction coefficient; and the electronic control module (12) stores multiple sets of model parameter sets corresponding to different medium temperature ranges, different pressure difference ranges, or different valve diameters, so as to improve the accuracy and availability of flow estimation in a wide range of water treatment conditions.

3. The intelligent flow control valve for water treatment according to claim 1, characterized in that, The transient characteristics of the pressure signal from the pressure signal processing unit include the upstream pressure change rate. Downstream pressure change rate Pressure difference change rate Pressure fluctuation energy or a combination thereof; The electronic control module (12) applies an upper limit constraint on at least one of the valve opening change rate, opening acceleration, actuator output speed or output torque to suppress water hammer or pressure shock caused by rapid valve action.

4. The intelligent flow control valve for water treatment according to claim 1, characterized in that, The electronic control module (12) constructs a soft measurement of the medium condition based on the torque signal, actuator current signal, valve opening degree and ΔP. The soft measurement includes the equivalent viscosity index, gas content risk index, solid content disturbance index or a combination thereof. Furthermore, the electronic control module (12) adaptively corrects the flow closed-loop control parameters, the upper limit of the opening change rate, or the water hammer risk threshold based on the soft measurement parameters.

5. The intelligent flow control valve for water treatment according to claim 1, characterized in that, When any signal from the upstream pressure sensor (7), downstream pressure sensor (13), or torque sensor (11) is detected to be abnormal, exceeds the limit, or fails, the electronic control module (12) switches to a degraded control mode. The degraded control mode includes one or more of the following: open-loop limiting control, approximate flow control based on single-sided pressure, or conservative control based on torque / current. The electronic control module (12) is provided with a communication interface with the pump station controller or frequency converter to output a linkage signal to coordinate the pump speed and valve opening when water hammer risk or degraded mode is triggered.

6. The intelligent flow control valve for water treatment according to claim 1, characterized in that, The current and torque monitoring unit includes a current sampling circuit and a torque sampling circuit. The current sampling circuit is used to collect the drive current or equivalent load current of the electric actuator (8), and the torque sampling circuit is used to collect the output of the torque sensor (11). The microprocessor generates at least one monitoring quantity based on the drive current and torque signal. The monitoring quantity includes one or more of the following: overload criterion, jamming criterion, hysteresis criterion, or transmission abnormality criterion.

7. A self-calibration method for an intelligent flow control valve suitable for water treatment, applied to the intelligent flow control valve for water treatment as described in any one of claims 1-6, characterized in that, Includes the following steps: S1. Obtain valve opening degree x and upstream pressure within the control cycle. Downstream pressure Calculate the pressure difference using the torque signal T. ; S2. When the preset triggering conditions are met, a limited opening perturbation command is superimposed on the electric actuator (8) without changing the overall control objective, so that the valve opening changes slightly. S3. During the perturbation process, cache the changes in the opening degree x, pressure difference ΔP, and torque signal T according to the timestamp to form a sample segment; S4. Perform a data validity determination based on torque constraints on the sample segment. When an abnormal torque peak is detected or the torque is continuously higher than the threshold, remove the corresponding sample segment or postpone the self-calibration. S5. Update the valve flow-opening model parameters based on the effective sample segments after elimination, and use the updated model parameters for subsequent flow estimation and control.

Images