Pressure water supplementing metering device for municipal pipeline water pressure test

Abstract

The present application relates to the technical field of pipeline hydrostatic test, and especially relates to a pressure water supplementing and metering device for municipal pipeline hydrostatic test, which comprises a water tank, a booster pump, a flowmeter and a control cabinet, a program controller is arranged in the control cabinet, further comprising a pressure sensor and a vibration sensor arranged along the axial direction of the pipeline to be tested, and an active tunable damper arranged between the pump outlet and the pipeline; the program controller synchronously collects pressure and vibration data, carries out correlation analysis on the data, generates a pressure wave space-time propagation graph, calculates a water hammer risk index and divides a risk level through a nonlinear fusion model according to the dominant frequency, amplitude, propagation speed change rate and vibration energy of the pressure wave, selects an inhibition strategy according to the risk level, calculates a pressure peak value reduction rate and a wave fluctuation decay time according to the data after inhibition, and generates an effect evaluation value. The present application realizes active sensing, graded inhibition and effect evaluation of water hammer risk, and improves the safety and intelligent level of the hydrostatic test.

Description

Technical Field

[0001] This invention relates to the field of pipeline water pressure testing technology, and in particular to a pressure water supply metering device for municipal pipeline water pressure testing. Background Technology

[0002] After construction, municipal pipelines (such as water supply pipelines and fire protection pipelines) must undergo a hydrostatic test in accordance with the requirements of the "Code for Construction and Acceptance of Water Supply and Drainage Pipeline Engineering" and the "Technical Code for Fire Water Supply and Fire Hydrant System" to verify the strength and tightness of the pipelines. The hydrostatic test requires precise and stable replenishment of pressurized water to the pipeline system, and real-time high-precision measurement of the replenishment volume to determine whether there is any leakage in the pipeline.

[0003] Existing technologies include various pressure water supply metering devices, typically comprising a water tank, booster pump, flow meter, and control cabinet, capable of automatic pressure stabilization and water supply metering. However, in actual hydrostatic testing operations, especially under conditions such as booster pump start-up and shutdown, and rapid valve adjustment, water hammer effects are easily triggered. This water hammer effect is a pressure shock wave generated in a pressure pipeline due to a sudden change in flow velocity. Water hammer effects can cause a sudden surge in pressure, which can range from minor pressure fluctuations affecting the accuracy of test data to serious consequences such as pipeline rupture, equipment damage, and even safety accidents. Traditional water hammer suppression methods mainly include: 1. Slow operation: This method slows down flow rate changes by extending the pump start-up and shutdown time and valve adjustment time. However, this method significantly prolongs the test time and cannot respond quickly in emergency situations. 2. Passive buffer devices: such as airbag buffer tanks or spring safety valves, but these devices can only passively absorb energy or relieve pressure after water hammer occurs. They are "post-event remedies" and cannot prevent damage to pipelines from instantaneous impacts. 3. Fixed parameter control: The pump speed or valve position is adjusted by a PID controller, but the PID parameters are fixed and it is difficult to adapt to the water hammer characteristics of different pipe materials and different working conditions. Therefore, there is an urgent need for an intelligent water hammer suppression solution that can proactively identify water hammer precursors, accurately predict water hammer development, and adopt a graded suppression strategy to solve the water hammer hazard problem during hydrostatic testing. Summary of the Invention

[0004] To address these issues, this invention provides a pressure water supply metering device for municipal pipeline water pressure testing, aiming to solve the problems of water hammer precursors not being identified in advance, limited and slow-responding suppression methods, inability to quantify and evaluate suppression effects, and lack of self-learning and adaptive capabilities in the system.

[0005] To achieve the above objectives, the present invention provides a pressure water supply metering device for municipal pipeline water pressure testing, comprising a water tank, a booster pump, a flow meter, and a control cabinet, wherein the control cabinet is equipped with a programmable controller, characterized in that it further comprises: Several pressure sensors and several vibration sensors are arranged at intervals along the axial direction of the pipeline to be tested; An active adjustable damper is installed between the outlet of the booster pump and the pipeline to be tested. The active adjustable damper includes a resonant cavity and a throttling valve. The program controller includes: The data acquisition module synchronously acquires the detection data of each pressure sensor and each vibration sensor at a sampling frequency of not less than 1000Hz, and the detection data includes pressure data and vibration data. The spectrum construction module performs spectrum analysis on the detection data to extract the dominant frequency and amplitude of the pressure fluctuation, and performs correlation analysis on the detection data at different sensor locations to generate a real-time spatiotemporal propagation spectrum that reflects the propagation direction, velocity and energy attenuation of the pressure wave along the pipeline. The index grading module calculates the water hammer hazard index of the current working condition based on the dominant frequency, amplitude, rate of change of propagation speed, and vibration energy of the pressure fluctuation through a preset nonlinear fusion model to determine the risk level of the current working condition. The strategy selection module determines the origin point of the pressure wave based on the spatiotemporal propagation map and determines the suppression strategy based on the risk level. The suppression strategy includes adjusting the volume of the resonant cavity and the opening of the throttle valve; The effect evaluation module calculates the pressure peak reduction rate and fluctuation decay time based on the detection data after executing the suppression strategy, and generates an effect evaluation value.

[0006] As a preferred technical solution for pressure replenishment metering devices used in municipal pipeline water pressure testing, active adjustable dampers include: The damper housing is connected to the main pipeline; A movable inertial block is disposed within the damper housing, which, together with the damper housing, forms the resonant cavity; An electromagnetic driver connected to the moving inertial block drives the moving inertial block to move according to the execution command to change the volume of the resonant cavity; The throttle valve is located at the connection between the damper housing and the main pipeline.

[0007] As a preferred technical solution for a pressure water replenishment metering device used in municipal pipeline water pressure testing, the nonlinear fusion model preset in the index grading module is a neural network model or a support vector machine model. Its input layer includes at least: the similarity between the current pressure waveform and the feature waveforms in the pre-stored water hammer precursor feature library, the instantaneous change rate of the pressure wave propagation speed, the energy integral of the vibration signal in the preset feature frequency band, and the maximum value of the pressure rise rate.

[0008] As a preferred technical solution for pressure water supply metering devices used in municipal pipeline water pressure testing, the index grading module is configured to determine the risk level based on the water hammer hazard index, wherein: If the water hammer hazard index is less than or equal to the low-risk threshold, it is determined to be a low-risk level. If the water hammer hazard index is between the low-risk threshold and the high-risk threshold, it is determined to be at a medium-risk level. If the water hammer hazard index is greater than or equal to the high-risk threshold, it is determined to be a high-risk level. The high-risk threshold is greater than the low-risk threshold.

[0009] As a preferred technical solution for a pressure water supply metering device used in municipal pipeline water pressure testing, the strategy selection module determines a suppression strategy based on the risk level and the pressure wave origin point identified in the spatiotemporal propagation spectrum, including: In response to the low-risk level, the suppression strategy is to adjust the speed change curve of the booster pump to a non-linear curve by using a frequency converter; In response to a medium-risk level, the suppression strategy is to tune the resonant frequency of the resonant cavity of the active tunable damper to a phase that is the same as but opposite to the dominant frequency of the pressure wave. In response to a high-risk level, the suppression strategy is to predict the time when the pressure wave arrives at the preset pressure relief valve based on the origin point and propagation speed of the pressure wave, and open the pressure relief valve at a preset opening degree before that time.

[0010] As a preferred technical solution for a pressure water supply metering device used in municipal pipeline water pressure testing, the strategy selection module calculates the propagation speed of the pressure wave based on the coordinates of each pressure sensor and the time difference between the arrival of the pressure wave at each pressure sensor, and determines the origin point of the pressure wave based on the location of the abrupt change in the propagation speed.

[0011] As a preferred technical solution for a pressure replenishment metering device for municipal pipeline water pressure testing, the effect evaluation module is configured to determine the pressure peak reduction rate as the ratio of the measured pressure peak after the suppression strategy is implemented to the predicted pressure peak predicted according to the spatiotemporal propagation map.

[0012] As a preferred technical solution for a pressure replenishment metering device used in municipal pipeline water pressure testing, the effect evaluation module is configured to determine the fluctuation attenuation time based on the time-domain waveform of the pressure data after implementing the suppression strategy, including: Starting from the moment the suppression strategy is completed, continuously monitor the pressure data detected by the pressure sensor; When the duration of the pressure fluctuation amplitude first decaying to the preset stability threshold reaches the preset time window, this moment is recorded as the fluctuation stability moment; The difference between the time when the fluctuation stabilizes and the time when the suppression strategy is completed is determined as the fluctuation decay time.

[0013] As a preferred technical solution for a pressure replenishment metering device used in municipal pipeline water pressure testing, the effect evaluation module is configured to calculate an effect evaluation value based on the pressure peak reduction rate and the fluctuation decay time, including: Obtain the first normalized value of the pressure peak reduction rate and the second normalized value of the fluctuation decay time; The first weighting factor and the second weighting factor are determined based on the risk level, wherein the weighting factor of the pressure peak reduction rate corresponding to the high risk level is higher than the weighting factor of the pressure peak reduction rate corresponding to the low risk level. The effect evaluation value is obtained by summing the product of the first normalized value and the first weight factor, and the product of the second normalized value and the second weight factor.

[0014] Compared with existing technologies, the beneficial effects of this invention are as follows: The pressure water replenishment metering device for municipal pipeline water pressure testing provided by this invention constructs a real-time spatiotemporal propagation map by deploying multiple high-frequency pressure sensors and vibration sensors along the pipeline. This allows for the extraction of characteristic information before water hammer occurs from multiple dimensions, such as the dominant frequency, amplitude, rate of change of propagation speed, and vibration energy of the pressure wave, thus elevating water hammer identification from post-event monitoring to pre-event early warning. A nonlinear fusion model (neural network or support vector machine) is used to calculate the water hammer hazard index, and combined with the effect evaluation value, the model parameters and strategy library are continuously optimized through reinforcement learning. This allows for automatic adjustment based on feedback from each event, adapting to changes in different pipe materials, water temperatures, and operating conditions, achieving continuous evolution and becoming increasingly intelligent with use. By actively suppressing the water hammer effect, damage to the pipeline caused by sudden pressure surges is avoided, ensuring the safety of the testing process. Simultaneously, a stable pressure environment ensures the accuracy of the water replenishment metering data, providing a reliable basis for judging pipeline tightness. In particular, by classifying the water hammer hazard index into low, medium, and high risk levels, a precise suppression strategy with graded matching is achieved: at low risk, the pump speed curve is optimized by the frequency converter to smooth operation from the source and avoid fluctuations; at medium risk, the damper frequency is tuned to the opposite phase to actively suppress waves on the propagation path; at high risk, the arrival time of the wave peak is predicted based on the origin point and propagation speed of the pressure wave, and the pressure relief valve is opened in advance to achieve precise peak reduction. This graded mechanism avoids excessive intervention or response lag of fixed control, and combined with the precise location of the origin point by the spatiotemporal propagation map, it ensures that each level of strategy is executed at the best time and in the most suitable way, thereby minimizing media loss and system disturbance while ensuring pipeline safety, and significantly improving the intelligence level and hammer suppression effect of the water pressure test. In particular, a quantitative closed-loop evaluation system for the suppression effect was constructed through the effect evaluation module: First, the ratio of the measured peak pressure to the predicted peak pressure is defined as the peak pressure reduction rate, which intuitively reflects the suppression strategy's ability to reduce pressure shocks; at the same time, the fluctuation decay time is determined based on the time it takes for the pressure time-domain waveform to first stabilize to a preset threshold, accurately characterizing the speed at which the system recovers stability from fluctuations; then, after normalizing the two, weights are dynamically allocated according to the risk level and weighted summation is performed to generate a comprehensive effect evaluation value; this evaluation value not only provides operators with an objective basis for the suppression effect, but more importantly, it serves as a reward signal for reinforcement learning, driving the system to continuously optimize strategy parameters and model weights based on the feedback of each event, achieving a leap from passive suppression to active evolution, and significantly improving the device's adaptability under different operating conditions and its long-term operational reliability. Attached Figure Description

[0015] Figure 1 This is a connection diagram of the pressure water supply metering device for municipal pipeline water pressure testing according to an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0017] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0018] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0019] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] Please see Figure 1 As shown, it is a connection diagram of the pressure water supply metering device for municipal pipeline water pressure testing according to an embodiment of the present invention.

[0021] The present invention provides a pressure water supply metering device for municipal pipeline water pressure testing. The device adopts a vertical frame structure. Its exterior is equipped with a structural frame and shell composed of square steel skeleton and cover plate. The bottom is equipped with a handling caster with brakes and an adjustable support foot, which facilitates movement and positioning on the construction site, while ensuring stability during the test.

[0022] The device integrates core components such as a water tank, booster pump, flow meter, and control cabinet. The water tank is made of 304 stainless steel, with an effective volume of 0.6 cubic meters and a wall thickness of not less than 1.5 mm. The tank is equipped with an inlet, an overflow outlet, a drain outlet, and a manhole with a cover. A magnetic level gauge is installed on the side wall for real-time observation of the water level in the tank. The booster pump is a multi-stage centrifugal booster pump with a rated outlet pressure of 2.5MPa, an adjustable flow range of 1m³ / h-2m³ / h, a motor power of approximately 3kW, and stainless steel for the flow-through components. The flow meter adopts a Coriolis mass flow meter or a high-precision electronic turbine flow meter, with a measurement accuracy of ±0.1%FS, a resolution of 0.01 liters, a range of 0L / h-2000L / h, and a pressure rating of not less than 4.0MPa. It is used to accurately measure and accumulate the amount of water added to the pipeline in real time. The control cabinet has an IP54 protection rating and is equipped with a built-in programmable logic controller (PLC), relays, switching power supply, etc. The cabinet has a human-machine interface to display parameters such as pipeline pressure, instantaneous and cumulative water replenishment, water tank level, and pump operating status. It also supports audible and visual alarms for overpressure, water shortage, and equipment failure. The control cabinet also integrates a frequency converter to realize soft start and stop and stepless speed regulation of the booster pump.

[0023] Regarding pipeline connections, a pipeline filter with a 100-mesh stainless steel screen is installed before the inlet of the booster pump to filter impurities in the water. A check valve and a high-precision flow meter are installed sequentially on the outlet pipeline of the booster pump. The check valve is used to prevent backflow of high-pressure water from impacting the pump and metering device. A precision pressure regulating valve / back pressure valve is also connected in parallel on the main pipeline for accurate setting and stabilizing the water supply pressure. A safety pressure relief valve is installed on the pump outlet main pipeline with a set pressure of 2.6MPa-2.8MPa, which automatically opens to relieve pressure when the system is over-pressurized. The internal components of the device are connected by high-pressure ball valves, needle valves, and stainless steel pipelines, and pressure gauges (0MPa-4MPa) are provided for on-site visual monitoring. The device outlet is quickly connected to the dedicated water supply interface of the pipeline to be tested via a high-pressure hose for convenient on-site operation. In addition, the device is equipped with a USB / RS485 data interface for exporting test data or connecting to a remote monitoring system.

[0024] In addition, this device also includes several pressure sensors and several vibration sensors arranged at intervals along the axial direction of the pipeline to be tested (in practice, 3 to 6 pressure sensors and 2 to 3 vibration sensors are usually installed), as well as an active adjustable damper installed on the main pipeline between the booster pump outlet and the pipeline to be tested; the active adjustable damper includes a damper housing, a moving inertial block, an electromagnetic actuator, and a throttle valve, wherein the moving inertial block and the housing form a resonant cavity, the electromagnetic actuator can drive the inertial block to move according to the control command to change the volume of the resonant cavity, and the throttle valve is used to adjust the fluid flow area.

[0025] The control cabinet's built-in programmable controller includes several functional modules: a data acquisition module, a spectrum construction module, an index grading module, a strategy selection module, and an effect evaluation module. 1. The data acquisition module synchronously acquires detection data from each pressure sensor and vibration sensor at a sampling frequency of at least 1000Hz. 2. The spectrum construction module performs spectral and correlation analysis on the detection data to generate a real-time spatiotemporal propagation spectrum reflecting the pressure wave propagation characteristics. 3. The index grading module calculates the water hammer hazard index of the current operating condition based on the dominant frequency, amplitude, rate of change of propagation velocity, and vibration energy of the pressure fluctuation using a preset nonlinear fusion model, thereby determining the risk level. 4. The strategy selection module determines the pressure wave origin point based on the spatiotemporal propagation spectrum and determines the corresponding suppression strategy according to the risk level. 5. After executing the suppression strategy, the effect evaluation module calculates the pressure peak reduction rate and fluctuation decay time based on the detection data, generating an effect evaluation value. Through this structure, the present invention can achieve active identification, graded suppression, and effect evaluation of the water hammer effect, significantly improving the safety and intelligence level of pipeline hydrostatic testing.

[0026] Specifically, the interior of the device is arranged according to functional zones: the bottom is the equipment and piping layer, the middle is the integrated water tank, and the top is the control panel and some precision instruments; this compact layout allows the overall size to be controlled within 1200mm in length, 800mm in width, and 1200mm in height, meeting the needs of movement and operation in a limited space.

[0027] Specifically, the inertial block is a metal block (preferably a high-density alloy) with a certain mass. A sealing ring is provided between its outer periphery and the inner wall of the shell, so that the inertial block can slide axially in the shell while maintaining a seal. The inertial block divides the inner cavity of the shell into two parts: a resonant cavity connected to the water inlet and a back pressure cavity connected to the inertial block drive mechanism. When the inertial block moves, the volume of the resonant cavity changes accordingly. The electromagnetic actuator is located outside the damper housing, and its drive rod passes through the housing and is fixedly connected to the movable inertial block. The electromagnetic actuator generates electromagnetic force according to the execution command issued by the program controller, and pushes the inertial block to move axially through the drive rod. The electromagnetic actuator can be a linear motor, a proportional electromagnet, or a lead screw mechanism driven by a servo motor to achieve precise control of the position of the inertial block. The throttle valve is an electrically controlled proportional throttle valve, whose opening degree can be adjusted in real time according to the instructions of the programmable controller. It is used to control the flow area of ​​the fluid flowing through the damper, thereby changing the damping characteristics of the damper.

[0028] In implementation, under normal operating conditions, the electromagnetic actuator keeps the inertial block in the neutral position, the throttle valve maintains a certain opening, and the water flows normally through the resonant cavity and the throttle valve. When the program controller determines that active suppression needs to be activated based on the water hammer hazard index, the strategy selection module calculates the fluctuation frequency that needs to be offset based on the pressure wave's dominant frequency. According to vibration theory, the resonant frequency of the resonant cavity is inversely proportional to its volume. The electromagnetic actuator drives the inertial block to move to the corresponding position according to the target resonant frequency, adjusting the volume of the resonant cavity to the target value, thereby making the damper's natural frequency and the detected pressure wave's dominant frequency the same. The system is matched; simultaneously, the throttle valve adjusts its opening according to the command, working in coordination with the resonant cavity; when the volume of the resonant cavity is adjusted to match the main frequency of the pressure wave, the pressure fluctuation flowing through the damper will generate an anti-phase pressure wave in the resonant cavity. This anti-phase wave superimposes and cancels out the positive pressure wave in the main pipeline, thereby achieving active dissipation of water hammer energy; through the above structure, the active adjustable damper can complete frequency tuning and damping adjustment in milliseconds, achieving active, rapid, and precise suppression of water hammer pressure, which is significantly better than traditional fixed buffer tanks or passive pressure relief valves.

[0029] Specifically, the input layer of the nonlinear fusion model for calculating the water hammer hazard index includes at least the following four feature parameters: 1. Similarity between the current pressure waveform and the characteristic waveforms in the pre-stored water hammer precursor feature library: In advance, pressure fluctuation segments before the occurrence of typical water hammer are extracted through a large amount of historical test data or simulation data, and a water hammer precursor feature waveform library is established through cluster analysis; During real-time monitoring, the current pressure waveform is truncated by a sliding window, and the maximum similarity between the current window waveform and each feature waveform in the library is calculated using dynamic time warping or cosine similarity algorithm, which is used as the feature value; 2. Instantaneous rate of change of pressure wave propagation speed: Based on the data collected by multiple pressure sensors installed along the pipeline, the pressure wave propagation speed between adjacent sensors is calculated through cross-correlation analysis; the speed values ​​of multiple consecutive sampling periods are taken, and the instantaneous rate of change of the speed relative to the reference speed is calculated (i.e., the difference between the current speed and the speed at the previous moment divided by the time interval), which reflects the degree of abrupt change in propagation speed. 3. Energy integration of vibration signal in preset characteristic frequency band: Perform fast Fourier transform on the signal collected by vibration sensor to extract characteristic frequency band (e.g., 20Hz-200Hz) related to the resonance characteristics of pipe structure, integrate the power spectral density in this frequency band to obtain the vibration energy per unit time, which is used to characterize the structural vibration intensity induced by pressure fluctuation of pipe wall. 4. Maximum pressure rise rate: Calculate the maximum pressure rise slope per unit time (i.e., the maximum value of dP / dt) on the time domain curve of the pressure sensor. This value directly reflects the severity of the pressure shock.

[0030] In implementation, the structural design when using a neural network model is as follows: Input layer: 4 nodes, corresponding to the four feature parameters mentioned above.

[0031] Hidden layers: Two hidden layers are set, with 8 nodes in the first hidden layer and 4 nodes in the second hidden layer; ReLU (Rectified Linear Unit) is selected as the activation function to enhance the nonlinear expression capability; Output layer: 1 node, Sigmoid is used as the activation function, and the output value is mapped to the range [0,1], which is the water hammer danger index. The larger the value, the higher the risk. It should be understood that: 1. Collect various working condition data recorded in historical hydrostatic tests (including normal operation, slight fluctuations, near-water hammer events, and actual water hammer events). For each event, experts manually label the hazard index (continuous value or discrete level between 0 and 1) based on indicators such as pressure peak, pressure rise rate, and whether damage was caused, forming a training sample set; 2. Use mean squared error (MSE) to measure the difference between the predicted hazard index and the manually labeled value; 3. Use the Adam optimizer with an initial learning rate of 0.001 and a batch size of 32, iteratively training until the loss converges; during real-time operation, input the four feature values ​​extracted at the current moment into the trained neural network, and calculate the water hammer hazard index of the current working condition through forward propagation, which is then used by the subsequent grading module.

[0032] In implementation, the parameter configuration for the support vector machine model is as follows: Kernel function: The radial basis function (RBF) is selected, and its expression is as follows: ; Parameter settings: The penalty coefficient C is set to 10, and the kernel function parameter γ is set to 0.1 (optimized by cross-validation based on feature dimension and data distribution). Output format: Continuous hazard index values ​​are output using Support Vector Regression (SVR); or discrete risk levels (such as low, medium, and high) are output using Support Vector Classification (SVC) and then mapped to continuous indices. It should be understood that the same training sample set as the neural network is used, with four features as input and manually labeled hazard index (or level) as output, to train the SVR / SVC model. The C and γ parameters are optimized through grid search. After extracting features in real time, the data is input into the trained SVR model to directly obtain the water hammer hazard index.

[0033] Understandably, the water hammer hazard index is a dimensionless value in the [0,1] interval that maps multiple physical quantities (pressure waveform similarity, wave velocity change rate, vibration energy, and pressure rise rate) to the nonlinear fusion model. Its value directly represents the probability and potential intensity of destructive water hammer under the current working conditions: Water hammer hazard index = 0 means that the current working conditions are completely consistent with the historical normal working conditions and there is no risk of water hammer; Water hammer hazard index = 1 means that the current working conditions are completely consistent with the historical typical destructive water hammer events and water hammer is about to occur with extremely high intensity; The higher the water hammer hazard index, the greater the probability and the stronger the potential intensity of destructive water hammer under the current working conditions. In practice, the low-risk threshold typically ranges from [0.2, 0.4], and is preferably set to 0.3; the high-risk threshold typically ranges from [0.7, 0.9], and is preferably set to 0.8. It should be understood that: 1. The pressure rise rate (dP / dt) is the core indicator for measuring the destructive force of water hammer. Through extensive experimental statistics: when dP / dt < 0.5 MPa / s, the pipeline system can naturally attenuate the fluctuations without active intervention due to its own damping, corresponding to a danger index of less than 0.3; when dP / dt is in the range of 0.5 MPa / s to 5 MPa / s, the fluctuations have exceeded the natural attenuation capacity, but have not yet reached the instantaneous strength limit of the pipeline, requiring active intervention but allowing for millisecond-level response time, corresponding to a danger index of 0.3 to 0.8; when dP / dt > 5 MPa / s, the pressure wave can reach the destructive peak within tens of milliseconds, requiring emergency measures with microsecond-level response, corresponding to a danger index greater than 0.8. 2. The energy integral of the vibration sensor in the characteristic frequency band reflects the dynamic stress induced by pressure fluctuations in the pipe wall. According to the fatigue curve of the pipe material, when the vibration energy is less than 30% of the material's fatigue limit, it can be considered safe (low risk); when the vibration energy reaches 30% to 80% of the fatigue limit, the fluctuation needs to be limited to continue (medium risk); when the vibration energy exceeds 80% of the fatigue limit, it is close to the instantaneous failure threshold and emergency pressure relief is necessary (high risk). 3. Abrupt changes in the propagation velocity of pressure waves are a precursor to water hammer: a propagation velocity change rate of <5% indicates a slow pressure adjustment caused by normal operation (low risk); a propagation velocity change rate of 5% to 20% indicates that a significant pressure wave has formed and is propagating (medium risk); a propagation velocity change rate of >20% indicates that a complete shock wave has formed and is about to reach the critical point (high risk).

[0034] Specifically, the characteristics of a low-risk level are small pressure fluctuation amplitude, slow rate of rise, and no obvious shock wave yet formed. At this time, the root cause of the fluctuation is usually the pressure change generated by the operation process itself (such as the start and stop of the pump, the opening and closing of the valve) rather than the shock wave that has already formed. However, even at the low-risk level, if left unchecked, the fluctuation may gradually amplify due to system resonance or operational accumulation, eventually evolving into a medium-to-high risk. Therefore, the low-risk stage is the best window of opportunity to prevent water hammer, and proactive intervention should be taken, but the intensity of intervention should not be too great. Adjusting the speed curve of the frequency converter (such as the S-curve or exponential curve) is essentially changing the excitation characteristics of the disturbance source, rather than passively responding to the fluctuation that has already occurred. Nonlinear curve control makes the speed change rate transition smoothly during the start-up and stop phases, avoiding sudden pressure changes, thereby reducing the injection of fluctuation energy from the source.

[0035] Specifically, at the medium-risk level, pressure waves have already formed and propagated along the pipeline, exhibiting obvious periodic fluctuations, abnormal propagation speed, and a significant increase in vibration energy. At this point, the energy of the fluctuations exceeds the control range of the source and intervention must be carried out along the propagation path. The active tunable damper is essentially a tunable mechanical-fluid resonance system. Its working principle is based on the anti-phase wave elimination theory in active vibration control: when the resonant frequency of the damper is consistent with the dominant frequency of the suppressed pressure wave, the fluid and inertial block inside the damper will vibrate at the same frequency as the incoming wave. By precisely controlling the position of the inertial block, the pressure fluctuations generated inside the damper are 180 degrees out of phase with the pressure fluctuations in the main pipeline. According to the superposition principle of waves, when two waves with the same frequency but opposite phases meet, they cancel each other out, thereby achieving energy dissipation.

[0036] Specifically, at high-risk levels, the pressure wave has already formed a shock wave with an extremely high rise rate and peak pressure, and is about to reach or has already reached the instantaneous strength limit of the pipeline. Because the wave propagation speed far exceeds the physical response limit of the damper tuning, any tuning or absorption methods are too late at this time. The core contradiction at this time is that the pressure wave peak is about to arrive, and some of the medium must be released before the peak arrives to reduce the peak pressure: Traditional pressure relief only opens when the pressure reaches the set value, at which point the pressure peak has already acted on the pipeline, which is a post-event remedy and cannot avoid the instantaneous impact; while predictive pressure relief calculates the time when the wave peak will arrive at the pressure relief valve in advance based on the origin point and propagation speed of the pressure wave, and opens the valve with a preset opening degree before the wave peak arrives (e.g., 10 to 50 milliseconds in advance). When the wave peak arrives, some of the medium has already been released through the pressure relief valve, and the peak pressure is flattened; in addition, the timing accuracy of the pressure relief valve opening directly affects the suppression effect. The precise opening time of the pressure relief valve requires knowing the exact moment when the wave peak arrives, which depends on the origin point coordinates and propagation speed data provided by the spatiotemporal propagation map; In implementation, the coordinates x0 of the pressure wave origin and the propagation velocity v are obtained from the spatiotemporal propagation spectrum, and the dominant frequency f of the pressure wave is obtained from the spectrum analysis. The installation coordinates xv of the pressure relief valve are known values. The wavefront arrival time tbase = |xv-x0| / v is the time for the pressure wave to propagate from the origin to the pressure relief valve. For an approximately sinusoidal pressure wave, the delay time of the wavefront relative to the peak is about one-quarter of the cycle, tpeak-delay = 1 / 4f. The absolute time when the wavefront arrives at the pressure relief valve with the current time tnow as the reference is tarrival = tnow + tbase + tpeak-delay. Considering the mechanical response delay of the valve, a lead time Δt is reserved before the wavefront arrives, and the opening time is topen = tarrival - Δt.

[0037] Specifically, the propagation speed of pressure waves in the pipeline medium (water) is approximately 1000m / s to 1500m / s. The specific value depends on factors such as pipe material, wall thickness, and water temperature. When a sudden pressure change occurs at a certain point in the pipeline (origin point) due to operation (such as rapid valve closure or pump start-up and shutdown), the pressure wave will propagate along the pipeline to both sides with that point as the center. By measuring the time difference of the pressure wave reaching different sensors and combining it with the spatial coordinates of each sensor, the origin point location and propagation speed of the pressure wave can be calculated in reverse.

[0038] Specifically, in the reinforcement learning framework, the performance evaluation value is input as a reward signal into the self-learning module; the self-learning module judges the quality of the current policy based on the magnitude of the reward value: a high reward value indicates a good inhibition effect, and the selection tendency of the current policy should be strengthened; a low reward value indicates a poor inhibition effect, and other policies should be explored or policy parameters should be adjusted. In implementation, the generated effect evaluation value is input into the self-learning module as part of the training sample: the detection data of this event, risk level, selected strategy, execution parameters, and effect evaluation value are combined into a complete training sample; using the effect evaluation value E as the reward signal, reinforcement learning algorithms such as Q-learning or policy gradient are used to update the state-action value function in the strategy selection module. When E is high, the probability of selecting the strategy in that state is increased, and when E is low, the probability is decreased; the sample is input into a neural network or support vector machine model, and the parameters of the first-layer nonlinear fusion model are optimized in reverse through supervised learning, so that the predicted water hammer risk index is more consistent with the subsequent actual suppression effect (E value); for strategies with excellent long-term performance, the parameters of the self-learning module can be solidified to form a gold medal strategy, and for strategies with poor performance, they can be automatically eliminated or mutant strategies can be generated for exploration.

[0039] In implementation, the focus of the suppression effect differs under different risk levels: at high risk, more emphasis is placed on whether the peak can be reduced quickly (the pressure peak reduction rate is more important), while at low risk, more emphasis is placed on whether the system can quickly return to stability (the fluctuation decay time is more important); the effect evaluation module dynamically adjusts the weight factors by risk level, realizing the adaptive matching of evaluation criteria and risk scenarios. The effectiveness evaluation value is a dimensionless comprehensive performance index, ranging from [0,1] (or mapped to this range after normalization). The larger the value, the better the overall effect of the suppression strategy: When the effectiveness evaluation value is between 0.8 and 1.0, the suppression strategy is significant, the pressure peak is greatly reduced, and the system recovers to stability in a very short time; when the effectiveness evaluation value is between 0.5 and 0.8, the suppression strategy is effective, the pressure peak is effectively controlled (reduction rate of 50% to 80%), and the fluctuation decay time is within an acceptable range; when the effectiveness evaluation value is between 0.2 and 0.5, the suppression strategy is partially effective, the pressure peak is reduced to some extent but the magnitude is insufficient, or the system takes too long to recover to stability; when the effectiveness evaluation value is between 0 and 0.2, the suppression strategy is basically ineffective, the pressure peak is not effectively controlled, or the fluctuation continues for a long time or may even intensify.

[0040] Specifically, based on the propagation pattern of pressure waves in the spatiotemporal propagation spectrum, the program controller can predict the peak pressure when the pressure wave reaches critical locations (such as valves and elbows) without any intervention before the suppression strategy is executed. The prediction method is as follows: 1. Wavefront extrapolation method: Based on the pressure wave rise segment data collected before the current moment, fit the pressure change curve over time (such as using polynomial fitting or exponential fitting), extrapolate to the pressure wave peak moment, and obtain the predicted peak value. 2. Waveform matching method: Match the current pressure waveform with the pressure waveforms under similar working conditions recorded in the historical database, and extract the peak value corresponding to the historical waveform as the predicted peak value; 3. Physical model method: Based on the pressure wave propagation equation (such as the basic equation of water hammer), combined with the pipe parameters and the current wave velocity and amplitude, the predicted peak value is obtained by solving the equation. Understandably, after the suppression strategy is executed, the program controller extracts the maximum value detected by the pressure sensor during the event from the stored pressure data, and records it as the measured pressure peak value; the pressure peak value reduction rate = (predicted pressure peak value - measured pressure peak value) ÷ predicted pressure peak value × 100%; In practice, the pressure fluctuation amplitude = measured pressure - target test pressure (target value during the stabilization phase); the preset stabilization threshold is set according to the test accuracy requirements, usually ±1% to ±5% of the target pressure; in one implementation, for a test pressure of 2.5MPa, the preset stabilization threshold can be set to 0.025MPa; the preset time window is 5s to 10s.

[0041] Understandably, the peak pressure reduction rate is a percentage value, which can be directly divided by 100% to map to the [0,1] interval to obtain the first normalized value. Based on engineering experience, a maximum acceptable decay time (15s~30s) is set. When the fluctuation decay time ≤ the maximum acceptable decay time, the second normalized value = 1 - when the fluctuation decay time / maximum acceptable decay time; when the fluctuation decay time > the maximum acceptable decay time, the second normalized value is 0. In implementation, the first weighting factor (i.e., the pressure peak reduction rate weight) and the second weighting factor (i.e., the volatility decay time weight) for low risk are 0.3 and 0.7, respectively; the first weighting factor and the second weighting factor for medium risk are 0.6 and 0.4, respectively; and the first weighting factor and the second weighting factor for high risk are 0.9 and 0.1, respectively. The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0042] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A pressure water supply metering device for municipal pipeline water pressure testing, comprising a water tank, a booster pump, a flow meter, and a control cabinet, wherein the control cabinet is equipped with a programmable controller, characterized in that, Also includes: Several pressure sensors and several vibration sensors are arranged at intervals along the axial direction of the pipeline to be tested; An active adjustable damper is installed between the outlet of the booster pump and the pipeline to be tested. The active adjustable damper includes a resonant cavity and a throttling valve. The program controller includes: The data acquisition module synchronously acquires detection data from each pressure sensor and each vibration sensor, including pressure data and vibration data. The spectrum construction module performs spectrum analysis on the detection data to extract the main frequency and amplitude of pressure fluctuations, and performs correlation analysis on the detection data at different sensor locations to generate a real-time spatiotemporal propagation spectrum. The index grading module calculates the water hammer hazard index of the current working condition based on the dominant frequency, amplitude, rate of change of propagation speed, and vibration energy of the pressure fluctuation through a preset nonlinear fusion model to determine the risk level of the current working condition. The strategy selection module determines the origin point of the pressure wave based on the spatiotemporal propagation map and determines the suppression strategy based on the risk level. The suppression strategy includes adjusting the volume of the resonant cavity and the opening of the throttle valve; The effect evaluation module calculates the pressure peak reduction rate and fluctuation decay time based on the detection data after executing the suppression strategy, and generates an effect evaluation value.

2. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The active tunable damper includes: The damper housing is connected to the main pipeline; A movable inertial block is disposed within the damper housing, which, together with the damper housing, forms the resonant cavity; An electromagnetic driver connected to the moving inertial block drives the moving inertial block to move according to the execution command to change the volume of the resonant cavity; The throttle valve is located at the connection between the damper housing and the main pipeline.

3. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The nonlinear fusion model preset in the index grading module is a neural network model or a support vector machine model, and its input layer includes at least: the similarity between the current pressure waveform and the feature waveforms in the pre-stored water hammer precursor feature library, the instantaneous change rate of the pressure wave propagation speed, the energy integral of the vibration signal in the preset feature frequency band, and the maximum value of the pressure rise rate.

4. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The index grading module is configured to determine the risk level based on the water hammer hazard index, wherein: If the water hammer hazard index is less than or equal to the low-risk threshold, it is determined to be a low-risk level. If the water hammer hazard index is between the low-risk threshold and the high-risk threshold, it is determined to be at a medium-risk level. If the water hammer hazard index is greater than or equal to the high-risk threshold, it is determined to be a high-risk level. The high-risk threshold is greater than the low-risk threshold.

5. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The strategy selection module determines a suppression strategy based on the risk level and the pressure wave origin point identified in the spatiotemporal propagation map, including: In response to the low-risk level, the suppression strategy is to adjust the speed change curve of the booster pump to a non-linear curve by using a frequency converter; In response to a medium-risk level, the suppression strategy is to tune the resonant frequency of the resonant cavity of the active tunable damper to a phase that is the same as but opposite to the dominant frequency of the pressure wave. In response to a high-risk level, the suppression strategy is to predict the time when the pressure wave arrives at the preset pressure relief valve based on the origin point and propagation speed of the pressure wave, and open the pressure relief valve at a preset opening degree before that time.

6. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The strategy selection module calculates the propagation speed of the pressure wave based on the coordinates of each pressure sensor and the time difference between the arrival of the pressure wave at each pressure sensor, and determines the origin point of the pressure wave based on the location of the abrupt change in the propagation speed.

7. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The effect evaluation module is configured to determine the pressure peak reduction rate as the ratio of the measured pressure peak after the suppression strategy is implemented to the predicted pressure peak based on the spatiotemporal propagation map.

8. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The effect evaluation module is configured to determine the fluctuation decay time based on the time-domain waveform of the pressure data after implementing the suppression strategy, including: Starting from the moment the suppression strategy is completed, continuously monitor the pressure data detected by the pressure sensor; When the duration of the pressure fluctuation amplitude first decaying to the preset stability threshold reaches the preset time window, this moment is recorded as the fluctuation stability moment; The difference between the time when the fluctuation stabilizes and the time when the suppression strategy is completed is determined as the fluctuation decay time.

9. The pressure water supply metering device for municipal pipeline water pressure testing according to claim 1, characterized in that, The effect evaluation module is configured to calculate an effect evaluation value based on the pressure peak reduction rate and the fluctuation decay time, including: Obtain the first normalized value of the pressure peak reduction rate and the second normalized value of the fluctuation decay time; The first weighting factor and the second weighting factor are determined based on the risk level, wherein the weighting factor of the pressure peak reduction rate corresponding to the high risk level is higher than the weighting factor of the pressure peak reduction rate corresponding to the low risk level. The effect evaluation value is obtained by summing the product of the first normalized value and the first weight factor, and the product of the second normalized value and the second weight factor.

Images