Water hammer prediction and coordination method for multi-pump water supply system

Abstract

The present application relates to a kind of water hammer prediction coordination methods of multi-pump water supply system, comprising: water pump system initialization and characteristic curve calibration;Real-time data acquisition and acceleration calculation;Threshold judgment and event identification;Control execution;Multi-pump collaborative control;Parameter online learning and optimization.The present application can give early warning and start response in advance before water hammer or flood peak arrives by flow-pressure acceleration joint determination, response time is shortened by more than 50% than traditional PID control, so as to effectively inhibit water hammer peak pressure, reduce the risk of pipeline burst, prolong the service life of pipe network.The present application avoids frequent adjustment by dynamic threshold control, combined with energy optimization response strategy, can reduce system energy consumption by 10%-20%.The multi-pump collaborative control taken effectively avoids the resonance problem when operating in parallel, improves system stability, and has self-adaptive ability of control parameters, can learn and adjust online according to actual working condition, applicable to various complex water supply scenarios.

Description

Technical Field

[0001] This invention relates to a method for predicting and coordinating water hammer in a multi-pump water supply system. It is a coordination and control method for hydraulic engineering, applicable to a water supply system composed of multiple variable frequency water pumps, and a method for predicting water hammer and coordinating the various pumps to prevent water hammer from occurring. Background Technology

[0002] Currently, the control methods for multi-pump water supply systems are mainly divided into two categories: constant pressure water supply control and constant flow water supply control. Constant pressure water supply control uses the outlet pressure of the pipeline network as the controlled variable, adjusting the output frequency of the pump inverter through a PID controller to maintain a constant pipeline pressure. This method only focuses on the single variable of pressure. When the flow rate changes abruptly, due to fluid inertia and the lag of PID control, the system response is slow, easily leading to drastic pressure fluctuations or even pipe bursts. Constant flow water supply control uses the pipeline flow rate as the controlled variable, adjusting the pump speed through flow meter feedback. This method ignores pressure changes and cannot effectively cope with pressure anomalies when the pipeline resistance characteristics change (such as changes in valve opening or pipe blockage). Both constant pressure and constant flow control are single-variable closed-loop control, lacking the ability to comprehensively judge the dynamic behavior of the fluid and unable to predict water hammer or sudden flow increases. Conventional PID control cycles are in the millisecond to second range, while the pressure wave propagation speed of water hammer can reach over 1000 m / s, making it difficult for existing PID control systems to achieve timely responses. Existing systems are all passive response controls, meaning they only adjust after pressure or flow deviates from the setpoint, failing to anticipate water hammer or flood peaks and take preventative measures. When multiple pumps operate in parallel, their similar frequencies can cause pipeline resonance, exacerbating pressure pulsations; current technologies lack effective resonance identification and avoidance mechanisms. Traditional control systems often employ conservative strategies (such as rapidly increasing to the highest frequency) when dealing with transient conditions, leading to unnecessary energy consumption. Improving the transient response and predictive waterproofing capabilities of pump systems, enhancing the stability of multi-pump coordination, and reducing energy consumption are problems that need to be addressed. Summary of the Invention

[0003] To overcome the problems of existing technologies, this invention proposes a method for predicting and coordinating water hammer in multi-pump water supply systems. This method employs a series of measures, including a bivariate acceleration joint judgment mechanism, a dynamic threshold adaptive control mechanism, a flow-frequency characteristic curve model, a water hammer and flood peak energy estimation model, and a multi-pump collaborative control and resonance avoidance mechanism. These measures enable early prediction of water hammer or flood peak arrivals, improve the transient response of the pump system, solve the problem of pipeline resonance caused by multi-pump operation, and reduce energy consumption.

[0004] The objective of this invention is achieved as follows: a method for predicting and coordinating water hammer in a multi-pump water supply system. The coordinated pump system includes multiple parallel variable frequency pumps and their respective frequency converters. Pressure transmitters are installed on the inlet and outlet pipes of each pump, and an electromagnetic flow meter is installed on the outlet pipe of each pump. Each electromagnetic flow meter, pressure transmitter, and frequency converter is electrically connected to a controller via a communication network. The method includes the following steps:

[0005] Step 1, Initialization and Characteristic Curve Calibration of the Water Pump System:

[0006] Each water pump was tested for its frequency conversion characteristics, and the flow rate, head, and power data at different frequencies were recorded.

[0007] The flow-frequency characteristic curve is fitted and stored in the controller's memory. Two operating conditions are set for each water pump: Condition 1 and Condition 2. Condition 1 is the reference condition, which is the known operating state of the water pump at the rated frequency. Condition 2 is the target condition, which is the operating state of the water pump after the frequency is changed to other frequencies.

[0008] Establish a flow-frequency characteristic curve model:

[0009]

[0010]

[0011]

[0012] Q1 is the flow rate under operating condition 1; Q2 is the flow rate under operating condition 2; H1 is the head under operating condition 1; H2 is the head under operating condition 2; P1 is the power under operating condition 1; P2 is the power under operating condition 2; n1 is the pump speed under operating condition 1; n2 is the pump speed under operating condition 2; f1 is the inverter output frequency under operating condition 1; f2 is the inverter output frequency under operating condition 2.

[0013] Considering the effects of fluid viscosity and pipe resistance, a correction factor is introduced:

[0014]

[0015] in: For the target frequency; The current frequency; For target traffic; K represents the current traffic. μ This is the viscosity correction factor. , The dynamic viscosity at a standard water temperature of 20°C; l K represents the dynamic viscosity of the fluid under actual operating conditions.R This is the tube resistance correction factor;

[0016] Measure the pipe parameters at the pump output end, including pipe length L, pipe inner diameter D, and calculate the pressure wave velocity c and the pipe natural frequency.

[0017] Calculate the acceleration coefficients α and β based on the pipeline parameters:

[0018]

[0019]

[0020] in: Reference pipe diameter For reference pipe length, Reference wave velocity for the pipeline;

[0021] Step 2, Real-time data acquisition and acceleration calculation: Collect the flow rate of each pump with a sampling period Δt. and pressure The signal is filtered using a moving average to eliminate noise interference.

[0022]

[0023] in: To eliminate noise interference with pump flow; t is the sampling time; N is the number of historical sampling points for moving average filtering;

[0024] Real-time traffic collection and pressure Signal, calculate flow acceleration and pressure acceleration :

[0025]

[0026] ;

[0027] Define the criteria for water hammer or flood peak warning:

[0028] when and At that time, a water hammer / flood peak warning was triggered, among which, The rated pressure of the water pump system, The rated flow rate of the water pump system;

[0029] Step 3, Threshold determination and event recognition: Calculate the dynamic threshold Th(Q) corresponding to the current traffic Q;

[0030]

[0031] in: This represents the allowable fluctuation threshold for the current flow rate Q. Minimum threshold; The maximum threshold is k; the attenuation coefficient is k. This is the maximum design flow rate of the water pump system;

[0032] Determine whether the actual fluctuation exceeds the threshold. If it does, further determine whether the acceleration meets the water hammer warning conditions. Based on the determination result, the water pump system enters the corresponding control mode.

[0033] Step 4, Control Execution: Control execution includes three modes: Normal mode: fluctuations are within the threshold, and the current frequency remains unchanged; Warning mode: fluctuations exceed the threshold but do not trigger a water hammer warning, and regular PID regulation is initiated; Emergency response mode: a water hammer warning is triggered.

[0034] Trigger the water hammer warning and perform the following actions:

[0035] Water hammer pressure increment :

[0036]

[0037] Where: ρ is the fluid density; This refers to the change in flow velocity.

[0038] Water hammer energy Estimate:

[0039]

[0040] in: The average flow rate during the water hammer event;

[0041] Water hammer duration Estimate:

[0042]

[0043] Based on the energy estimation results, determine the aggressiveness of the system response strategy;

[0044] Calculate the target frequency based on the flow-frequency curve, increase the frequency at the matched rate, and track the water hammer energy release curve;

[0045] Step 5, Multi-pump coordinated control: Each pump controller shares frequency information through the communication bus. The controller calculates the frequency interval of each pump in real time and determines whether there is a resonance risk. If there is a resonance risk, phase misalignment control is executed to adjust the target frequency of each pump. After the emergency response ends, each pump synchronously and smoothly reduces its frequency to the steady-state target frequency.

[0046] Multi-pump coordination: This embodiment employs multi-pump collaborative control and resonance avoidance mechanisms.

[0047] Resonance identification criteria:

[0048] When multiple water pumps are running in parallel, monitor the operating frequency of each pump. ...If the following conditions are met, a resonance risk is considered to exist:

[0049] (For any i ≠ j)

[0050] Furthermore, the pressure signal exhibits periodic pulsations, with the pulsation frequency approaching the pipe's natural frequency. :

[0051]

[0052] Where: i and j are the pump number indices. ; Minimum frequency interval;

[0053] Phase reversal control strategy:

[0054] After triggering the resonance warning, phase misalignment control is executed to maintain sufficient interval between the pump frequencies:

[0055]

[0056] in: The average frequency of each pump; n is the number of pumps in operation. For safe frequency spacing;

[0057] Synchronous stability adjustment strategy:

[0058] After a water hammer or flood peak event ends, all pumps synchronously reduce their frequency to the target frequency, with the reduction rate controlled in a coordinated manner.

[0059]

[0060] Frequency information is shared among the pumps via a communication bus to ensure a stable synchronous transition.

[0061] Step 6, Online Parameter Learning and Optimization: Record the actual response data for each water hammer or flood peak event, and based on the actual results, use the least squares method to correct the pipe resistance coefficient K online. R Regularly evaluate the control effect and optimize the acceleration coefficients α, β and threshold curve parameters.

[0062] The advantages and beneficial effects of this invention are as follows: This invention, through joint judgment of flow rate and pressure acceleration, can provide early warning and initiate response before water hammer or flood peaks arrive, shortening the response time by more than 50% compared to traditional PID control. This effectively suppresses peak water hammer pressure, reduces the risk of pipeline rupture, and extends the service life of the pipeline network. This invention avoids frequent adjustments through dynamic threshold control, and combined with an energy-optimized response strategy, can reduce system energy consumption by 10%–20%. The multi-pump collaborative control effectively avoids resonance problems during parallel operation, improves system stability, and has adaptive control parameter capabilities, enabling online learning and adjustment based on actual operating conditions, making it suitable for various complex water supply scenarios. Attached Figure Description

[0063] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0064] Figure 1 This is a flowchart of the method described in an embodiment of the present invention. Detailed Implementation

[0065] Example:

[0066] This embodiment is a method for predicting and coordinating water hammer in a multi-pump water supply system. The pump system coordinated by the method includes: multiple parallel variable frequency pumps and each pump's own frequency converter. Pressure transmitters are installed on the inlet and outlet pipes of each pump, and electromagnetic flow meters are installed on the outlet pipe of each pump. Each electromagnetic flow meter, pressure transmitter, and frequency converter is electrically connected to a controller via a communication network. The selection and configuration of various equipment are as follows:

[0067] Electromagnetic flowmeter selection:

[0068] Range selection principle: The design flow rate should be within 30% to 80% of the flow metering range. Maximum design range for flow metering. for:

[0069]

[0070] in: For design range.

[0071] Electromagnetic flowmeter water level accuracy requirements: 0.5 grade or higher, response time: ≤100ms (fast response type), output signal: 4-20mA or RS485 digital communication.

[0072] The electromagnetic flow meter should be installed at a location 5 to 10 times the pipe diameter downstream of the water pump outlet to ensure stable flow.

[0073] Pressure transmitter selection:

[0074] Pressure transmitter range selection principle: The rated pressure should be within 30% to 70% of the transmitter's range, with a margin for water hammer peak. Maximum design range of the pressure transmitter. for:

[0075] (Considering water hammer peak value)

[0076] Pressure transmitter accuracy requirements: 0.25 grade or higher, response time: ≤10ms (high-speed response type).

[0077] Installation locations: Install one at the water pump outlet and one at the inlet for pressure differential detection.

[0078] Controller configuration:

[0079] Controller: It adopts a PLC and has a high-speed analog signal acquisition module (sampling period ≤10ms) and floating-point operation capability.

[0080] Frequency converter: Each water pump is equipped with an independent frequency converter, which supports analog / communication control and has vector control or direct torque control functions.

[0081] Communication network: Industrial Ethernet or Profibus-DP bus is used to realize high-speed data exchange between the controller and each frequency converter (communication cycle ≤20ms).

[0082] The method specifically includes the following steps, the process of which is as follows: Figure 1 As shown:

[0083] Step 1, Pump System Initialization and Characteristic Curve Calibration: Perform frequency conversion characteristic tests on each pump, recording flow rate, head, and power data at different frequencies; fit the flow-frequency characteristic curve and store it in the controller memory; set two operating conditions for each pump: Condition 1 and Condition 2; Condition 1 is the reference condition, which is the known operating state of the pump at its rated frequency (e.g., 50Hz power frequency). All values ​​are measured or known values ​​on the nameplate; Condition 2 is the target condition; it represents the operating state of the water pump after the frequency converter is set to other frequencies. These are predicted values ​​calculated using proportional relationships. Operating condition 1 is "known," while operating condition 2 is "to be calculated." The difference lies in the frequency / speed, which causes the flow rate, head, and power to change proportionally. For example, the flow rate at 50Hz... (Condition 1), then Downloading traffic means (Operating Condition 2). In the actual calibration process: first, the measured value is measured at the rated frequency. As a baseline (condition 1), the controller then uses the similarity law formula to calculate the flow rate, head, and power at the target frequency when frequency adjustment is required (condition 2). Therefore, condition 1 is determined once during initialization calibration, while condition 2 is dynamically calculated during operation.

[0084] Establish a flow-frequency characteristic curve model:

[0085]

[0086] Where: Q is the flow rate; H is the head; P is the power; n is the rotational speed; f is the inverter output frequency;

[0087] According to the similarity law of water pumps:

[0088]

[0089]

[0090]

[0091] Flow rate under operating condition 1 ; Flow rate under operating condition 2 ; The head (m) under operating condition 1; The head (m) under operating condition 2; The power (kW) is under operating condition 1. The power (kW) is under operating condition 2. The pump speed (rpm, revolutions per minute) is the pump speed under operating condition 1. The pump speed (rpm, revolutions per minute) is under operating condition 2. The inverter output frequency under operating condition 1 ( ); The inverter output frequency under operating condition 2 ( ).

[0092] Considering the effects of fluid viscosity and pipe resistance, a correction factor is introduced:

[0093]

[0094] in: For the target frequency ( ), which is the inverter output frequency that needs to be adjusted to; For the current frequency ( This refers to the current output frequency of the inverter. For target traffic ( (i) the traffic volume that the system needs to achieve; For the current traffic ( This refers to the current actual traffic flow of the system. This is the viscosity correction factor. , The dynamic viscosity at a standard water temperature of 20°C; The dynamic viscosity of the fluid under actual working conditions ( (), depends on the current water temperature; The tube resistance correction factor is obtained through online learning based on actual operating data, with an initial value of 1.0.

[0095] The flow-frequency characteristic curve originates from pump similarity laws: a set of mathematical relationships used to predict the performance changes of centrifugal pumps at different speeds or impeller diameters. These laws are based on the similarity principle of fluid dynamics and are an important theoretical basis for pump selection, speed regulation, and energy-saving retrofitting. The key to establishing the flow-frequency correspondence based on pump similarity laws is the correction factor. , The introduction of [the technology] and the joint early warning mechanism of bivariate acceleration.

[0096] Measure the pipe parameters (length L, inner diameter D) at the pump output end, calculate the pressure wave velocity c and the pipe's natural frequency; calculate the acceleration coefficients α and β based on the pipe parameters.

[0097] α and β are acceleration coefficients, determined based on pipe characteristics. Recommended range of values: , The specific values ​​of α and β, along with the pipe diameter D and pipe length L, can be determined using the following empirical formulas:

[0098]

[0099]

[0100] in, (Reference pipe diameter) (Reference to the supervisor) (Reference wave speed).

[0101] Step 2, Real-time data acquisition and acceleration calculation: based on the sampling period (Recommended 50ms) Collect flow rate of each pump and pressure The signal is filtered using a moving average to eliminate noise interference.

[0102]

[0103] in: To eliminate noise interference with pump flow; t is the sampling time; N is the number of historical sampling points for moving average filtering (based on the sampling period). Sure).

[0104] Calculate flow acceleration and pressure acceleration :

[0105] Since this embodiment adopts a dual-variable acceleration joint judgment mechanism, it is necessary to install an electromagnetic flowmeter and a pressure transmitter on each water pump outlet pipe to collect flow data in real time. and pressure Signal, calculate flow acceleration and pressure acceleration :

[0106]

[0107]

[0108] in: The sampling period is recommended to be between 10ms and 100ms.

[0109] Define the criteria for water hammer / flood peak warning:

[0110] when and At that time, a water hammer / flood peak warning was triggered, among which, The rated pressure of the water pump system, This refers to the rated flow rate of the water pump system.

[0111] Step 3, Threshold Determination and Event Recognition: Calculate the dynamic threshold corresponding to the current traffic Q. ;

[0112] To avoid the rigidity of traditional techniques with fixed threshold values, this embodiment employs a dynamic threshold adaptive control mechanism:

[0113] To address the varying control accuracy requirements under different flow conditions, this embodiment proposes a dynamic threshold curve, enabling the control threshold to adaptively adjust with the flow rate.

[0114]

[0115] in: The allowable fluctuation threshold (percentage) for the current flow Q; The minimum threshold is 3%, with a recommended value of 3% (higher accuracy is required for high flow rates). The maximum threshold is 10% (larger fluctuations are allowed at low flow rates); k is the attenuation coefficient, with a recommended value of 2.5–3.5. This is the maximum design flow rate of the water pump system.

[0116] When the actual fluctuation is within the threshold range, the current frequency remains unchanged to avoid frequent adjustments. When the fluctuation exceeds the threshold, acceleration judgment and response control are initiated.

[0117] Determine whether the actual fluctuation exceeds the threshold. If it does, further determine whether the acceleration meets the water hammer warning conditions. Based on the judgment result, the water pump system executes the corresponding control mode: normal mode, warning mode, or emergency response mode.

[0118] Step 4, Control Execution: Control execution includes three modes: Normal mode: fluctuations are within the threshold, and the current frequency remains unchanged; Warning mode: fluctuations exceed the threshold but do not trigger a water hammer warning, and regular PID regulation is initiated; Emergency response mode: a water hammer warning is triggered.

[0119] Trigger the water hammer warning and perform the following actions:

[0120] The total energy and duration of this transient event were estimated using a water hammer flood peak energy estimation model.

[0121] Based on the Joukowsky formula, the water hammer pressure increment ΔP is:

[0122]

[0123] Where: ρ is the fluid density (water approximately 1000 kg / m³); c is the pressure wave velocity (approximately 1000–1400 m / s). This represents the change in flow velocity.

[0124] Water hammer energy Estimate:

[0125]

[0126] in: The average flow rate during the water hammer event ( or From an energy calculation perspective, water hammer energy is the integral of pressure increment and flow rate over time. Since real-time integration is difficult in practical calculations, an approximation method is used: the total energy is estimated by multiplying the average flow rate during the water hammer period by the water hammer pressure increment and duration. More precisely: =(Q_water hammer start time + Q_water hammer end time) / 2, or in other words This represents the average flow rate during a water hammer event.

[0127] Water hammer duration Estimate:

[0128]

[0129] Based on the energy estimation results, the aggressiveness of the system response strategy is determined.

[0130] The target frequency is calculated based on the flow-frequency curve, the frequency is increased by matching the rate, and the water hammer energy release curve is tracked.

[0131] Step 5, Multi-pump coordinated control: Each pump controller shares frequency information through the communication bus. The controller calculates the frequency interval of each pump in real time and determines whether there is a resonance risk. If there is a resonance risk, phase misalignment control is executed to adjust the target frequency of each pump. After the emergency response ends, each pump synchronously and smoothly reduces its frequency to the steady-state target frequency.

[0132] Multi-pump coordination: This embodiment employs multi-pump collaborative control and resonance avoidance mechanisms.

[0133] Resonance identification criteria:

[0134] When multiple water pumps are running in parallel, monitor the operating frequency of each pump. ,...,..., ...If the following conditions are met, a resonance risk is considered to exist:

[0135] (For any) )

[0136] Furthermore, the pressure signal exhibits periodic pulsations, with the pulsation frequency approaching the pipe's natural frequency f. pipe :

[0137]

[0138] Where i and j are the pump number indices. ; The minimum frequency interval is recommended to be 2Hz.

[0139] Phase reversal control strategy:

[0140] After triggering the resonance warning, phase misalignment control is executed to maintain sufficient interval between the pump frequencies:

[0141]

[0142] in: The average frequency of each pump; n is the number of pumps in operation. For a safe frequency interval, a value of 5Hz is recommended.

[0143] Synchronous stability adjustment strategy:

[0144] After a water hammer or flood peak event ends, all pumps synchronously reduce their frequency to the target frequency, with the reduction rate controlled in a coordinated manner.

[0145] ,

[0146] Frequency information sharing between pumps is achieved through a communication bus (such as Modbus / Profibus), ensuring a stable and synchronous transition.

[0147] Step 6, Online Parameter Learning and Optimization: Record the actual response data for each water hammer or flood peak event, and based on the actual results, use the least squares method to correct the pipe resistance coefficient online. Regularly evaluate the control effect and optimize the acceleration coefficients α, β and threshold curve parameters.

[0148] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred arrangement, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solution of the present invention (such as the form of the water pump, the arrangement, the application of various formulas, the order of steps, etc.) without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for predicting and coordinating flood hammer in a multi-pump water supply system, wherein the pump system coordinated by the method includes: The water pump system coordinated by the method includes: multiple parallel variable frequency water pumps and each water pump's own frequency converter; pressure transmitters are installed on the inlet and outlet pipes of each water pump; electromagnetic flow meters are installed on the outlet pipe of each water pump; and each electromagnetic flow meter, pressure transmitter, and frequency converter is electrically connected to a controller via a communication network. The method is characterized by the following steps: Step 1, Initialization and Characteristic Curve Calibration of the Water Pump System: Each water pump was tested for its frequency conversion characteristics, and the flow rate, head, and power data at different frequencies were recorded. The flow-frequency characteristic curve is fitted and stored in the controller's memory. Two operating conditions are set for each water pump: Condition 1 and Condition 2. Condition 1 is the reference condition, which is the known operating state of the water pump at the rated frequency. Condition 2 is the target condition, which is the operating state of the water pump after the frequency is changed to other frequencies. Establish a flow-frequency characteristic curve model: The flow rate under operating condition 1; The flow rate under operating condition 2; The head under operating condition 1; The head under operating condition 2; This refers to the power output under operating condition 1. This refers to the power output under operating condition 2. This refers to the pump speed under operating condition 1. This refers to the pump speed under operating condition 2. This is the inverter output frequency under operating condition 1. This refers to the inverter output frequency under operating condition 2. Considering the effects of fluid viscosity and pipe resistance, a correction factor is introduced: in: For the target frequency; The current frequency; For target traffic; Current traffic; This is the viscosity correction factor. , Standard water temperature Dynamic viscosity at that time; This represents the dynamic viscosity of the fluid under actual operating conditions. This is the tube resistance correction factor; Measure the pipe parameters at the pump output end, including pipe length L, pipe inner diameter D, and calculate the pressure wave velocity c and the pipe natural frequency. Calculate the acceleration coefficient based on the pipeline parameters. , : in: This is a reference pipe diameter. For reference pipe length, Reference wave velocity for the pipeline; Step 2, Real-time data acquisition and acceleration calculation: based on the sampling period The flow rate Q(t) and pressure P(t) signals of each pump are collected, and moving average filtering is used to eliminate noise interference. in: To eliminate noise interference with pump flow; t is the sampling time; N is the number of historical sampling points for moving average filtering; Real-time acquisition of flow rate Q(t) and pressure P(t) signals, and calculation of flow acceleration aQ and pressure acceleration aP: ； Define the criteria for water hammer or flood peak warning: when and At that time, a water hammer / flood peak warning was triggered, among which, The rated pressure of the water pump system, This refers to the rated flow rate of the water pump system. Step 3, Threshold determination and event recognition: Calculate the dynamic threshold Th(Q) corresponding to the current traffic Q; Where: Th(Q) is the allowable fluctuation threshold under the current flow Q; Minimum threshold; The maximum threshold is k; the attenuation coefficient is k. This is the maximum design flow rate of the water pump system; Determine whether the actual fluctuation exceeds the threshold. If it does, further determine whether the acceleration meets the water hammer warning conditions. Based on the determination result, the water pump system enters the corresponding control mode. Step 4, Control Execution: Control execution includes three modes: Normal mode: fluctuations are within the threshold, and the current frequency remains unchanged; Warning mode: fluctuations exceed the threshold but do not trigger a water hammer warning, and regular PID regulation is initiated; Emergency response mode: a water hammer warning is triggered. Trigger the water hammer warning and perform the following actions: Water hammer pressure increment : in: For fluid density; This refers to the change in flow velocity. Water hammer energy Estimate: in: The average flow rate during the water hammer event; The duration of the water hammer; Water hammer duration Estimate: Based on the energy estimation results, determine the aggressiveness of the system response strategy; Calculate the target frequency based on the flow-frequency curve, increase the frequency at the matched rate, and track the water hammer energy release curve; Step 5, Multi-pump collaborative control: Each pump controller shares frequency information through the communication bus. The controller calculates the frequency interval of each pump in real time and determines whether there is a risk of resonance. If there is a risk of resonance, phase misalignment control is executed to adjust the target frequency of each pump. After the emergency response ends, each pump synchronously and smoothly reduces its frequency to the steady-state target frequency. Multi-pump coordination: This embodiment employs multi-pump collaborative control and resonance avoidance mechanisms. Resonance identification criteria: When multiple water pumps are running in parallel, monitor the operating frequency of each pump. ...If the following conditions are met, a resonance risk is considered to exist: Furthermore, the pressure signal exhibits periodic pulsations, with the pulsation frequency approaching the pipe's natural frequency. : Where i and j are the pump number indices. ; Minimum frequency interval; Phase reversal control strategy: After triggering the resonance warning, phase misalignment control is executed to maintain sufficient interval between the pump frequencies: in: The frequency interval for each pump; The average frequency of each pump; n is the number of pumps in operation. For safe frequency spacing; Synchronous stability adjustment strategy: After a water hammer or flood peak event ends, all pumps synchronously reduce their frequency to the target frequency, with the reduction rate controlled in a coordinated manner. Where: f is the output frequency of the frequency converter; Frequency information is shared among the pumps via a communication bus to ensure a stable synchronous transition. Step 6, Online Parameter Learning and Optimization: Record the actual response data for each water hammer or flood peak event, and based on the actual results, use the least squares method to correct the pipe resistance coefficient online. Regularly evaluate the control effectiveness and optimize the acceleration coefficient. , and threshold curve parameters.

Images