Pressure regulation methods, devices, equipment and media for multi-branch self-pressurized irrigation networks

Abstract

This invention relates to the field of agricultural water conservancy engineering and intelligent control technology, specifically to a method, device, equipment, and medium for pressure regulation of multi-branch self-pressurized irrigation networks. In this invention, the relationship between valve opening and hydraulic loss coefficient is accurately modeled through experiments, providing a precise physical characteristic basis for pressure regulation, enabling precise quantification of the impact of valve adjustment on network pressure. A multi-objective regulation model is constructed with pressure balance and reliability as its objectives, taking into account both the uniformity of network pressure distribution and safety redundancy, thereby improving the hydraulic performance of the irrigation system. A serial hybrid intelligent algorithm is used for solving the problem, combining the global search of NSGA-II with the fast convergence advantage of MOPSO to efficiently obtain the Pareto optimal solution set, achieving precise regulation of valve opening, making the network pressure distribution more reasonable, reducing pressure fluctuations and energy consumption, and improving the stability, reliability, and water resource utilization efficiency of the irrigation system.

Description

Technical Field

[0001] This invention relates to the field of agricultural water conservancy engineering and intelligent control technology, specifically to a method, device, equipment and medium for pressure regulation of multi-branch self-pressurized irrigation networks. Background Technology

[0002] Self-pressurized irrigation systems rely on the gravitational potential energy generated by terrain elevation differences to drive water flow, offering advantages such as energy saving and low operating costs. However, the pressure distribution of its pipe network is entirely dependent on terrain conditions. Under conditions such as changes in water load, pipe aging, or changes in local resistance, severe uneven pressure distribution can easily occur, leading to excessively high pressure in some pipe sections, causing risks of leaks and pipe bursts, while other pipe sections suffer from insufficient pressure, affecting the uniformity of irrigation.

[0003] Traditional control methods typically rely on manual experience to adjust key valves or optimize pipe diameter and layout during the design phase. They lack dynamic, precise, and systematic pressure regulation capabilities for existing pipe networks with fixed topologies. Summary of the Invention

[0004] This invention provides a method, device, equipment, and medium for pressure regulation of multi-branch self-pressurized irrigation networks, in order to solve the problem in the prior art of lacking dynamic, precise, and systematic pressure regulation capabilities for existing networks with fixed topologies.

[0005] In a first aspect, the present invention provides a method for pressure regulation in a multi-branch self-pressurized irrigation network, the method comprising: The hydraulic loss coefficient of the valve under different valve openings was determined by experiment. With the pressure balance and pressure reliability of each node in a multi-branch self-pressurized irrigation network as the objectives, a multi-objective pressure control model is constructed with valve opening as the decision variable. The pressure balance is measured by the standard deviation of the pressure at each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy at each node in the network. Based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model and obtain the Pareto optimal solution set. Based on the preset engineering control requirements, one set of solutions is selected from the Pareto optimal solution set as the valve opening control scheme to control the valve opening.

[0006] In this invention, the relationship between valve opening and hydraulic loss coefficient is precisely modeled through experiments, providing a precise physical characteristic basis for pressure regulation in multi-branch self-pressurized irrigation networks, enabling precise quantification of the impact of valve adjustment on network pressure. A multi-objective regulation model is constructed with pressure balance and reliability as objectives, taking into account both the uniformity of network pressure distribution and safety redundancy, thereby improving the hydraulic performance of the irrigation system. The NSGA-II-MOPSO serial hybrid intelligent algorithm is used for solving the problem, combining the global search of NSGA-II with the fast convergence advantage of MOPSO to efficiently obtain the Pareto optimal solution set, ensuring both solution diversity and accelerating convergence speed, thus improving optimization efficiency. From the Pareto solution set, a regulation scheme is selected according to engineering requirements to achieve precise control of valve opening, making the network pressure distribution more reasonable, reducing pressure fluctuations and energy consumption, improving the stability, reliability, and water resource utilization efficiency of the irrigation system, while reducing the difficulty and cost of manual regulation.

[0007] In one optional implementation, the multi-objective pressure control model further includes node pressure constraints, energy conservation constraints, pipeline flow velocity constraints, and valve opening constraints. The node pressure constraints include minimum and maximum node pressure constraints. The energy conservation constraint includes the head difference between any two nodes being equal to the friction head loss and local head loss of the pipe section between them, wherein the local head loss is determined by the valve hydraulic loss coefficient. The pipeline flow velocity constraints include minimum and maximum flow velocity constraints. The valve opening constraint includes minimum and maximum valve opening constraints.

[0008] In this invention, by adding constraints on node pressure, energy conservation, pipeline flow velocity, and valve opening to the multi-objective pressure control model, multi-dimensional precise constraints and guarantees for irrigation network pressure control are achieved. Specifically, node pressure constraints ensure that the pressure at each node is within a safe range, avoiding overpressure or underpressure, and improving irrigation uniformity and equipment safety; energy conservation constraints, combined with valve hydraulic loss coefficients, ensure consistency between hydraulic calculations and actual energy transfer in the network, improving model accuracy and control reliability; pipeline flow velocity constraints maintain reasonable flow rates, preventing siltation or scouring, extending network lifespan, and reducing head loss; and valve opening constraints limit the operating range, avoiding ineffective or dangerous adjustments, and improving system stability and execution efficiency. Thus, through the synergistic effect of multiple constraints, the accuracy, safety, economy, and engineering practicality of pressure control in multi-branch self-pressurized irrigation networks are significantly improved, effectively ensuring the stable operation of the irrigation system and the efficient utilization of water resources.

[0009] In one optional implementation, the valve hydraulic loss coefficient corresponding to different valve opening degrees is determined experimentally, including: Experiments were conducted to obtain data on inlet and outlet pressures and flow rates under different inlet pressures and valve openings. The actual head loss is calculated based on the inlet and outlet pressures and flow data. Based on the actual head loss, determine the valve hydraulic loss coefficient under different inlet pressures and different valve openings; The valve hydraulic loss coefficient corresponding to different valve openings is determined by the average value of the valve hydraulic loss coefficient at different inlet pressures under any valve opening.

[0010] In this invention, data on inlet and outlet pressures and flow rates are collected under different inlet pressures and valve openings, comprehensively covering actual operating conditions and making the obtained loss coefficients more consistent with real-world scenarios. By calculating the actual head loss to determine the loss coefficients, and combining this with the method of averaging under multiple inlet pressures, the error and randomness of single pressure tests can be effectively reduced, improving the accuracy and robustness of the valve hydraulic characteristic model. Furthermore, the high-precision valve loss coefficient model provides a reliable basis for the construction of subsequent multi-objective pressure control models (such as the calculation of local head loss under energy conservation constraints), making the solution of hydraulic balance in the pipeline network and the optimization and control of pressure distribution more accurate. This, in turn, improves the accuracy, stability, and reliability of pressure control in the irrigation pipeline network, ensuring the efficient and safe operation of the irrigation system, while reducing control failures or resource waste caused by model errors.

[0011] In one optional implementation, based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model to obtain the Pareto optimal solution set, including: Based on the valve hydraulic loss coefficient corresponding to different valve openings, and combined with preset boundary conditions, the pressure value of each node is calculated. Based on the pressure values ​​of each node, the NSGA-II algorithm is used to solve the multi-objective pressure control model to obtain the initial Pareto optimal solution set. Select an initial candidate elite set from the initial Pareto optimal solution set; The initial candidate elite set is optimized using the MOPSO algorithm to obtain the Pareto optimal solution set.

[0012] In this invention, the pressure values ​​of each node are first calculated based on the valve hydraulic loss coefficient and preset boundary conditions. Then, the NSGA-II algorithm is used to solve the multi-objective model to obtain an initial Pareto optimal solution set. This fully utilizes the global search capability of NSGA-II, ensuring the diversity and coverage of the initial solution set. Selecting an initial candidate elite set from the initial solution set allows focusing on high-quality solution regions and improving optimization efficiency. Subsequently, the MOPSO algorithm is used to optimize the elite set. Leveraging MOPSO's fast convergence and local fine-search characteristics, the quality and accuracy of the solution are further optimized. This hybrid strategy combines the advantages of both algorithms, ensuring the comprehensiveness of the Pareto optimal solution set while accelerating convergence and improving the quality of the optimal solution. This makes the solution of the multi-objective pressure control model more accurate and practical, effectively guiding valve opening control and improving the accuracy, stability, and reliability of pressure control in multi-branch self-pressurized irrigation networks, achieving an optimized balance between network pressure uniformity and reliability.

[0013] In one alternative implementation, screening an initial candidate elite set from the initial Pareto optimal solution set includes: When the number of solutions in the initial Pareto optimal solution set does not exceed a preset threshold, the initial Pareto optimal solution set is used as the initial candidate elite set. When the number of solutions in the initial Pareto optimal solution set exceeds a preset threshold, a preset threshold number of solutions are selected from the initial Pareto optimal solution set as an initial candidate elite set.

[0014] In this invention, an initial candidate elite set is selected from the initial Pareto optimal solution set. By handling different cases (using the initial solution set directly when the number of solutions is less than or equal to a preset threshold, and selecting a preset threshold number of solutions when the number exceeds the threshold), the input quality for subsequent optimizations is optimized. When the number of solutions does not exceed the threshold, the initial solution set is directly used as the elite set, avoiding unnecessary selection overhead and improving efficiency. When the number of solutions exceeds the threshold, selecting a preset threshold number of solutions as the elite set effectively eliminates inferior solutions, focuses on the high-quality solution region, reduces the search space and computational load of the subsequent MOPSO algorithm, and retains information on high-quality solutions, providing a more accurate optimization starting point for the MOPSO algorithm, improving optimization efficiency and the quality of the final solution. This makes the Pareto optimal solution set of the multi-objective pressure control model more targeted and practical, ensuring the efficiency and reliability of the valve opening control scheme, and contributing to the precise optimization of pressure control in multi-branch self-pressurized irrigation networks.

[0015] In one optional implementation, selecting a preset threshold number of solutions from the initial Pareto optimal solution set as an initial candidate elite set includes: The convergence index is determined based on the distance from each solution in the initial Pareto optimal solution set to the preset ideal point; The distribution index is determined based on the crowding distance of each solution in the initial Pareto optimal solution set; The solutions in the initial Pareto optimal solution set are scored based on the convergence index and the distribution index, and a preset threshold number of solutions are selected as the initial candidate elite set based on the scoring results.

[0016] In this invention, when selecting the initial candidate elite set from the initial Pareto optimal solution set, solutions are scored using a convergence index (distance from the solution to the preset ideal point) and a distribution index (crowding distance of solutions). A preset threshold number of solutions are then selected as the elite set. The convergence index ensures that solutions approach the ideal optimal solution, improving the optimization level; the distribution index ensures that solutions are evenly distributed in the target space, maintaining the diversity of the solution set. This combined scoring and selection mechanism can control the size of the elite set while considering both the quality (convergence) and coverage (distribution) of the solutions. This provides a high-quality initial optimization population for the subsequent MOPSO algorithm, reducing invalid searches, improving optimization efficiency, and ensuring the diversity and convergence of the final Pareto optimal solution set. This makes the solutions of the multi-objective pressure control model more accurate and comprehensive, helping valve opening control schemes to more efficiently and reliably optimize the pressure balance and reliability of the pipeline network.

[0017] In one alternative implementation, the method is triggered by a preset period or a preset operating condition event.

[0018] In this invention, triggering according to a preset cycle allows the pressure regulation process to be updated and optimized periodically, adapting to changes in pressure characteristics caused by factors such as pipe aging and gradual changes in water usage patterns during long-term operation of the pipeline network, ensuring the timeliness and continuity of the regulation strategy. Triggering according to preset operating conditions (such as peak / valley water usage, equipment failure, changes in pipeline topology, etc.) enables rapid response to sudden changes in operating conditions, timely adjustment of valve openings and pressure regulation schemes, avoiding problems such as overpressure, underpressure, or poor water flow caused by changes in operating conditions, and improving the stability and reliability of pipeline network operation. The combination of these two triggering methods takes into account both periodic maintenance and handling of sudden operating conditions, enabling the pressure regulation system of a multi-branch self-pressurized irrigation network to achieve both long-term stable optimization and flexible response to sudden changes, effectively ensuring the water supply quality, equipment safety, and water resource utilization efficiency of the irrigation system.

[0019] Secondly, the present invention provides a pressure regulating device for a multi-branch self-pressurized irrigation network, the device comprising: The test module is used to determine the valve hydraulic loss coefficient corresponding to different valve opening degrees through experiments; The model building module is used to construct a multi-objective pressure control model with valve opening as the decision variable, with the pressure balance and pressure reliability of each node in a multi-branch self-pressurized irrigation network as the objectives. The pressure balance is measured by the standard deviation of the pressure of each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy of each node in the network. The solution module is used to solve the multi-objective pressure control model based on the valve hydraulic loss coefficient using the NSGA-II-MOPSO serial hybrid algorithm to obtain the Pareto optimal solution set; The control module is used to select one set of solutions from the Pareto optimal solution set as the valve opening control scheme based on the preset engineering control requirements, and to control the valve opening.

[0020] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the multi-branch self-pressurized irrigation network pressure regulation method described in the first aspect or any corresponding embodiment.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the multi-branch self-pressurized irrigation network pressure regulation method described in the first aspect or any corresponding embodiment.

[0022] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the multi-branch self-pressurized irrigation network pressure regulation method described in the first aspect or any corresponding embodiment. Attached Figure Description

[0023] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the first process of a pressure regulation method for a multi-branch self-pressurized irrigation network according to an embodiment of the present invention. Figure 2 This is a flowchart of the NSGA-II-MOPSO serial hybrid algorithm for the pressure regulation method of multi-branch self-pressurized irrigation network according to an embodiment of the present invention; Figure 3 This is a structural block diagram of a multi-branch self-pressurized irrigation network pressure regulating device according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0027] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0028] According to an embodiment of the present invention, a method for regulating pressure in a multi-branch self-pressurized irrigation network is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] This embodiment provides a method for pressure regulation in a multi-branch self-pressurized irrigation network. Figure 1 This is a flowchart of a pressure regulation method for a multi-branch self-pressurized irrigation network according to an embodiment of the present invention, as shown below. Figure 1 As shown, the process includes the following steps: Step S101 involves experimentally determining the valve hydraulic loss coefficient for different valve opening degrees. Specifically, in a multi-branch self-pressurized irrigation network, valves, as directly adjustable components, can effectively adjust local head loss by changing their opening degree, thereby altering the pressure distribution and flow allocation of the network. The valve hydraulic loss coefficient, as a key parameter affecting local head loss, needs to be characterized before adjustment. When the valve opening degree changes, the valve hydraulic loss coefficient also changes accordingly. This embodiment uses experimental methods to pre-determine and model the relationship between valve opening degree and valve hydraulic loss coefficient, enabling accurate determination of local head loss under different valve opening degrees.

[0030] Step S102: With the pressure balance and pressure reliability of each node in the multi-branch self-pressurized irrigation network as the objectives, a multi-objective pressure control model is constructed with valve opening as the decision variable. The pressure balance is measured by the standard deviation of the pressure at each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy at each node in the network.

[0031] Specifically, the multi-objective pressure control model constructed in this embodiment includes two conflicting optimization objectives: pressure balance and pressure reliability. The pressure balance objective is measured by the standard deviation of pressure at all key nodes in the pipeline network. A smaller standard deviation indicates a more uniform pressure distribution across the network nodes, resulting in higher system pressure balance. Therefore, this pressure balance objective... F 1 is expressed using the following formula:

[0032] In the formula, P x Represents the pressure at node x. This represents the average pressure of all pipeline nodes, and n represents the total number of nodes in the pipeline network.

[0033] The pressure reliability target is measured by the average node pressure redundancy. This indicator reflects the extent to which the actual pressure of a node exceeds its minimum required pressure. The higher the average value, the stronger the system's buffering capacity when facing fluctuations or partial failures, and the higher the reliability. Therefore, the pressure reliability target is expressed by the following formula:

[0034] In the formula, and These represent the minimum and maximum pressure allowed for node i, respectively.

[0035] Based on the two pressure control objectives mentioned above, during the control process, when adjusting the decision variable, namely the valve opening, the pressure balance objective and pressure reliability objective corresponding to different valve openings can be calculated, and the final valve opening can be determined by the changes in the objectives.

[0036] Step S103: Based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model to obtain the Pareto optimal solution set.

[0037] Specifically, to efficiently solve the aforementioned high-dimensional multi-objective pressure regulation model, this embodiment constructs a two-stage serial hybrid optimization algorithm. This algorithm organically integrates the Non-dominated Sorting Genetic Algorithm II (NSGA-II) with the Multi-objective Particle Swarm Optimization (MOPSO) algorithm, forming a clear two-stage optimization framework. The core process of the algorithm is as follows: First, NSGA-II is used for global exploration to obtain a well-distributed initial Pareto front; then, a precise elite solution screening mechanism is used to extract high-quality solutions from the results of NSGA-II; finally, this elite solution set is used as the core to initialize the MOPSO population for local fine-tuning search. This serial strategy fully leverages the advantages of NSGA-II in global exploration and solution set distribution, as well as the strengths of MOPSO in fast convergence, achieving complementary advantages.

[0038] The first stage involves global exploration and initial frontier construction. Specifically, the NSGA-II algorithm is used for global search. The algorithm initializes a randomized valve opening population and generates offspring through genetic operations such as simulated binary crossover and polynomial mutation. Fast non-dominated sorting and crowding distance calculation are used to evaluate and select individuals in the population to guide the search toward the Pareto front while maintaining the breadth of solution distribution. In this stage, the focus is on exploring the entire decision space to obtain a widely distributed but potentially not fully converged initial set of Pareto optimal solutions.

[0039] In the second stage, local fine-grained search and front optimization are implemented. Specifically, elite solutions selected from the Pareto front obtained in the first stage are used as the initial population and injected into the MOPSO algorithm. In this stage, each particle represents a valve opening combination, and its flight velocity is updated based on the individual's historical best position and the global best position of the population. To effectively handle constraints, the "constraint domination" principle is adopted: feasible solutions always dominate infeasible solutions. Simultaneously, an external archive is designed to store non-dominated solutions discovered during iteration, and techniques such as adaptive grids are used to maintain the distribution of the archive. The MOPSO stage fully utilizes the high-quality initial information provided by elite solutions to perform a fast local fine-grained search, aiming to improve the convergence accuracy and uniformity of the Pareto front distribution.

[0040] Optimal solution set output. After the hybrid algorithm finishes running, the final external archive is output, which is a set of Pareto optimal solutions. Each solution corresponds to a specific combination of valve openings and its corresponding objective function value (…). F 1, F 2).

[0041] In the process of solving the multi-objective pressure control model using the NSGA-II-MOPSO serial hybrid algorithm, when calculating the two objectives corresponding to each group of valve openings, the characteristics of the correspondence between valve opening and valve hydraulic loss coefficient are called to determine the pressure corresponding to each group of valve openings, thereby determining the corresponding objectives based on the above two objective calculation formulas.

[0042] Step S104: Based on preset engineering control requirements, select one set of solutions from the Pareto optimal solution set as the valve opening control scheme to control the valve opening. Specifically, the Pareto optimal solution set includes multiple valve opening combinations. In actual engineering applications, based on predetermined engineering control requirements (such as prioritizing the pressure balance of the pipeline network or prioritizing the reliability of system operation), select one set of valve opening combinations as the final valve opening control scheme (for example, prioritizing the highest pressure balance while meeting the minimum reliability threshold; or maximizing the system's reliability redundancy while ensuring basic balance).

[0043] Meanwhile, after selecting a set of valve opening degrees, these are encapsulated into standardized control commands and sent via a communication network to the electric valve actuators deployed at the pipeline site. Upon receiving the commands, the actuators drive the valves to adjust them to the target opening degree, thereby achieving systematic and optimized control of the pressure distribution in the self-pressurized irrigation network.

[0044] This invention systematically coordinates the two major objectives of "pressure balance" and "operational reliability" in the valve control of self-pressurized irrigation networks. It provides a series of optimal trade-off solutions through Pareto optimality theory, rather than a single solution, resulting in high decision-making flexibility. Simultaneously, the innovative NSGA-II-MOPSO serial hybrid architecture effectively combines the excellent global exploration and distribution preservation capabilities of NSGA-II with the fast local convergence characteristics of MOPSO. Compared to a single algorithm, it obtains a Pareto optimal solution set with better convergence and a more uniform distribution in a shorter computation time. Furthermore, this method optimizes based on the fixed topology of an existing pipeline network, requiring no modification to existing hardware, making it suitable for a wide range of irrigation district renovation and intelligent upgrade scenarios. The system architecture supports cloud-edge collaboration, enabling both offline deep optimization and online rolling optimization at a certain frequency, adapting to dynamically changing operating conditions. Furthermore, this method optimizes the distribution of pipeline pressure to be more balanced, effectively eliminating local overpressure points, reducing the risk of pipe bursts, and extending the life of the pipeline network; at the same time, it avoids unnecessary energy loss (for systems with pressurization links) or water head waste caused by maintaining excessively high pressure overall to ensure the pressure at the end.

[0045] This embodiment provides a method for pressure regulation in a multi-branch self-pressurized irrigation network, which includes the following steps: Step S201: Determine the valve hydraulic loss coefficient corresponding to different valve opening degrees through experiments.

[0046] Specifically, step S201 includes: Step S2011: Obtain valve inlet pressure, valve outlet pressure, and flow rate data under different inlet pressures and valve openings using experiments.

[0047] Step S2012: Calculate the actual head loss based on the inlet pressure, outlet pressure, and flow rate data.

[0048] Step S2013: Determine the valve hydraulic loss coefficient under different inlet pressures and different valve openings based on the actual head loss.

[0049] Step S2014: Determine the valve hydraulic loss coefficient corresponding to different valve openings based on the average value of the valve hydraulic loss coefficients at different inlet pressures under any valve opening.

[0050] Specifically, this embodiment employs a controllable physical experiment to determine the hydraulic loss function of the target regulating valve at different opening degrees and establishes a continuous functional relationship between the two. This physical experiment is conducted through a closed-loop test pipeline system, which includes a constant-pressure water supply device simulating different inlet pressures, an electrically operated regulating valve as the test object, high-precision sensors for measuring the pressure and flow rate before and after the valve, and an integrated data acquisition and control system. The entire experimental process is coordinated by a central controller.

[0051] During the test, under several set stable water supply pressure levels, the valve opening was adjusted step by step according to a preset gradient. After the pipeline flow rate stabilized at each opening level, the pressure before and after the valve, as well as the flow rate data, were simultaneously collected and recorded. The above adjustment and measurement process was repeated by changing the water supply pressure level to obtain a complete dataset covering the actual operating pressure range.

[0052] After obtaining the raw data, calculations are performed based on hydraulic formulas. The head loss and corresponding local loss coefficient of the valve under various test conditions can be calculated using relevant techniques by measuring the pressure difference and flow rate before and after the valve. The actual head loss h can then be calculated using the hydraulic formula. Furthermore, to eliminate random errors introduced by fluctuations in water supply pressure, the arithmetic mean of the loss coefficients calculated for the same valve opening under different water supply pressures is taken to characterize the steady-state hydraulic characteristics of the valve at that opening.

[0053] Finally, a curve fitting method (such as the least squares method) was used to mathematically model the "opening degree - loss coefficient" data points obtained from the experiment, obtaining a continuous function expression describing the relationship between the two. To ensure model accuracy, the goodness of fit must reach a specified threshold, thereby providing accurate and reliable hydraulic parameter inputs for subsequent model optimization.

[0054] Step S202 involves constructing a multi-objective pressure control model with valve opening as the decision variable, aiming at the pressure balance and reliability of each node in the multi-branch self-pressurized irrigation network. The pressure balance is measured by the standard deviation of the pressure at each node in the network, and the pressure reliability is measured by the average pressure redundancy at each node. Specifically, the multi-objective pressure control model also includes node pressure constraints, energy conservation constraints, pipeline flow velocity constraints, and valve opening constraints.

[0055] Among them, node pressure constraints include minimum and maximum node pressure constraints, meaning that the pressure of each node must be between its allowable minimum and maximum values. Pipeline flow velocity constraints include minimum and maximum pipeline flow velocity constraints, meaning that the flow velocity of each pipe section must be maintained between the minimum allowable flow velocity (to prevent siltation) and the maximum allowable flow velocity (to prevent cavitation and erosion). Valve opening constraints include minimum and maximum valve opening constraints, meaning that the opening of each valve needs to be limited within its effective mechanical adjustment range according to the valve's physical characteristics, for example [20%, 90%].

[0056] The energy conservation constraint states that the head difference between any two nodes is equal to the friction head loss and local head loss of the pipe section between them. The local head loss is determined by the valve hydraulic loss coefficient. Specifically, the energy conservation constraint means that the pipe network must satisfy the basic hydraulic energy equation, that is, for any two nodes i and j, the head balance equation along the connection path is expressed by the following formula:

[0057] In the formula, and They are nodes and Total head; ∑ , This is the sum of the head losses along the pipe sections; ∑ , This is the sum of all local losses in the pipeline section, including losses caused by valves, elbows, tees, reducers, etc.

[0058] The head loss along the friction path is determined using the following formula:

[0059] In the formula: The coefficient of friction; This refers to the length of the pipe section. This refers to the inner diameter of the pipe. The flow velocity in the pipe; This is the acceleration due to gravity.

[0060] The local head loss is determined using the following formula:

[0061] In the formula, K represents the local loss coefficient, including the valve hydraulic loss coefficient and the loss coefficients of pipe fittings such as elbows, tees, reducers, inlets, and outlets. The valve hydraulic loss coefficient is determined by the continuous functional relationship γ = f(ξ) between the valve relative opening ξ (%) and the valve hydraulic loss coefficient γ. Other loss coefficients can be obtained by consulting a manual.

[0062] Step S203: Based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model to obtain the Pareto optimal solution set; Specifically, step S203 includes: Step S2031: Based on the valve hydraulic loss coefficients corresponding to different valve opening degrees and combined with preset boundary conditions, calculate the pressure value of each node; specifically, taking a tree-like self-pressurized irrigation network as an example, assuming it has m branch pipes, there are a total of m control valves at the inlet of the branch pipes. The decision variable vector is denoted as x = [ξ1, ξ2, …, ξ]. m ], where ξ m ∈ [0.2, 0.9]. The calculation of objective functions F1 and F2 depends on the hydraulic adjustment calculation of the pipeline network. Given a valve opening combination U, the loss coefficient of each valve can be determined according to γ=f(ξ), and then used as a known condition. Combined with the boundary conditions such as pipeline network topology, pipe diameter, pipe length, node elevation, and water source pressure, the pressure value P of all nodes is calculated by solving the pipeline network hydraulic equations based on energy conservation and flow conservation, using the Newton-Raphson method of pipeline network hydraulic adjustment. x The specific calculation process can be found in relevant technical documents and will not be elaborated here.

[0063] Step S2032: Based on the pressure values ​​of each node, the NSGA-II algorithm is used to solve the multi-objective pressure control model to obtain the initial Pareto optimal solution set.

[0064] Specifically, such as Figure 2 As shown, the process includes the following steps: (1) Initialization: Set the NSGA-II parameters (population size N, maximum number of generations G, crossover probability Pc, mutation probability Pm) and MOPSO parameters (population size N, maximum number of generations G, inertia weight w, learning factors c1, c2).

[0065] (2) NSGA-II Global Exploration Phase: a. Population generation: Randomly generate an initial population within the valve opening decision space.

[0066] b. Fitness assessment: For each individual in the population (i.e., a set of valve opening combinations), perform network hydraulic balance calculations to obtain the pressure values ​​at each node, and then calculate the objective function value and assess the degree of constraint violation.

[0067] c. Non-dominated ordination and crowding calculation: A fast non-dominated ordination is performed on the population, dividing all individuals into multiple non-dominated levels (frontiers). For individuals within the same non-dominated level, their crowding distance is calculated, which measures the distribution density of the individual in the target space.

[0068] d. Selection operation: A binary tournament selection operator is used. Two individuals are randomly selected from the population, prioritizing the individual with the higher non-dominated ranking; if the rankings are the same, the individual with the greater crowding distance is selected.

[0069] e. Crossover and mutation operations: Perform crossover and mutation operations on the selected parent individuals to generate the offspring population.

[0070] f. Population renewal: The parent and offspring populations are merged, and the non-dominated ranking and crowding distance are recalculated for the merged population. Then, the best individuals are selected based on the non-dominated ranking and crowding distance to form a new generation of population.

[0071] g. Iterative loop: Repeat the above evaluation, selection, crossover, mutation and update steps until the preset maximum number of generations is reached to obtain the initial Pareto optimal solution set.

[0072] Step S2033: Select an initial candidate elite set from the initial Pareto optimal solution set.

[0073] Specifically, step S2033 includes: Step a1: When the number of solutions in the initial Pareto optimal solution set does not exceed a preset threshold, the initial Pareto optimal solution set is used as the initial candidate elite set. Specifically, when the number of all solutions located in the first non-dominated layer in the initial Pareto optimal solution set (i.e., the final population of NSGA II) does not exceed a preset threshold K, it is directly output as the initial candidate elite solution S.

[0074] Step a2: When the number of solutions in the initial Pareto optimal solution set exceeds a preset threshold, a preset threshold number of solutions are selected from the initial Pareto optimal solution set as an initial candidate elite set. Specifically, when the number of solutions in the initial Pareto optimal solution set exceeds a preset threshold K, it needs to be selected.

[0075] The screening process specifically includes the following steps: Step a21: Determine the convergence index based on the distances from each solution in the initial Pareto optimal solution set to a preset ideal point; specifically, the preset ideal point... It consists of the optimal values ​​in each target direction in the current frontier, that is:

[0076] Therefore, the convergence index of solution x is:

[0077] in, The smaller the value, the closer the solution is to the ideal point, and the better the convergence.

[0078] Step a22: Determine the distribution index based on the crowding distance of each solution in the initial Pareto optimal solution set; wherein, the larger the crowding distance, the sparser the solution is in the target space and the better the distribution.

[0079] Step a23: Scoring each solution in the initial Pareto optimal solution set based on the convergence index and the distribution index, and selecting a preset threshold number of solutions as the initial candidate elite set based on the scoring results. Specifically, for each solution... ∈ S The normalized convergence score and distribution score are calculated, and a comprehensive score is synthesized according to the weights, as expressed by the following formula:

[0080] In the formula, , This is to prevent division by zero for extremely small positive numbers.

[0081] After obtaining the sum of scores for multiple solutions, the overall score is calculated using the Score. x Sort the solutions from highest to lowest and select the top K solutions to form the initial candidate elite set S.

[0082] Step S2034: The initial candidate elite set is optimized using the MOPSO algorithm to obtain the Pareto optimal solution set.

[0083] Specifically, the process includes the following steps: a. Population Initialization: All elite solutions from the initial candidate elite set are directly injected into the MOPSO initial population as initial particle positions. For each elite solution in the initial candidate elite set, a small perturbation is applied to generate a corresponding perturbed solution, which is then added to the initial candidate elite set. New solutions are randomly generated within the decision space to replenish the initial candidate elite set to the set population size. The initial velocity of each particle is set to zero vector, and its current position is set to its historical best position.

[0084] In this context, the small perturbation refers to adding a small offset to the elite solutions. Its purpose is to avoid over-concentration of the initial population, maintain population diversity, and prevent MOPSO from prematurely falling into local optima. Furthermore, supplementing with random new solutions increases the population's coverage in the decision space, enhancing the algorithm's exploration capability. In the hybrid algorithm framework, the NSGA-II stage already provides a batch of high-quality solutions (elite solutions), and the task of the MOPSO stage is to perform a refined local search based on these. Therefore, the initial population mainly consists of elite solutions, supplemented by perturbation solutions and random solutions, which maintains both a high-quality starting point and search vitality.

[0085] b. External Archive Initialization: Evaluate the objective function and constraint violation rates of all particles in the initial population. Compare solutions using the constraint priority domination criterion: feasible solutions always dominate infeasible solutions; if both are feasible, compare Pareto domination relationships; if both are infeasible, compare the sum of constraint violations, with the solution with the smaller violation rate dominating the solution with the larger violation rate. Select all non-dominated solutions and store them in the external archive.

[0086] c. Iterative updates: i. Global Leader Selection: An adaptive grid method is used to manage the external archive. The target space is divided into several hypergrids. Based on the distribution density of solutions in each grid in the archive, a roulette wheel selection method is used, which tends to select a solution from the grid with fewer particles as the global optimal leader for this iteration.

[0087] ii. Velocity and position update: For each particle in the population, update its velocity according to the velocity update formula, and then update its position according to the position update formula.

[0088] iii. Boundary handling: For the updated out-of-bounds positional components, correct them to the boundary of the domain of the decision variable.

[0089] iv. Fitness reassessment: Recalculate the objective function values ​​and constraint violations of all particles in the updated population.

[0090] v. Update Individual Optimum: For each particle, compare its new position with its historical best position. According to the constraint priority domination criterion, if the new position dominates the historical best position, then replace the historical best position with the new position; if the two do not dominate each other, then randomly select one to retain, or select the one with the greater crowding distance to retain.

[0091] vi. Update the external archive: Add non-dominated solutions from the current population to the external archive. Remove all old solutions dominated by the newly added solutions from the archive. If the archive size exceeds its capacity, truncate it: prioritize retaining all feasible solutions; if the number of feasible solutions exceeds the capacity, sort the feasible solutions by crowding distance and remove the solution with the smallest crowding distance until the archive size equals the capacity; if there are insufficient feasible solutions, retain all feasible solutions and supplement infeasible solutions in ascending order of constraint violation until the archive is full.

[0092] d. Iterative loop: Repeat the above update steps until the preset maximum number of iterations is reached.

[0093] Output: The algorithm ends, and the final external archive is the Pareto optimal solution set, which contains multiple non-dominated valve opening control schemes.

[0094] It should be noted that the algorithm parameters are pre-calibrated before applying the hybrid optimization algorithm. Specifically, a systematic parameter tuning method is used to calibrate the key operating parameters of the NSGA-II and MOPSO algorithms to obtain a stable parameter combination that optimizes the overall performance of the algorithm when solving similar problems, thereby improving the robustness and solution efficiency of the algorithm.

[0095] Step S204: Based on the preset engineering control requirements, select one set of solutions from the Pareto optimal solution set as the valve opening control scheme, and control the valve opening.

[0096] For steps S202 to S204 above, the pressure of the multi-branch self-pressurized irrigation network can be controlled according to a preset cycle or a preset working condition event, that is, the pressure of the multi-branch self-pressurized irrigation network can be controlled by the above method every preset time; or when a preset working condition event occurs (such as a significant change in working condition), the pressure of the multi-branch self-pressurized irrigation network can be controlled.

[0097] This embodiment also provides a pressure regulating device for a multi-branch self-pressurized irrigation network. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0098] This embodiment provides a pressure regulating device for a multi-branch self-pressurized irrigation network, such as... Figure 3 As shown, it includes: Test module 31 is used to determine the valve hydraulic loss coefficient corresponding to different valve opening degrees through experiments; The model building module 32 is used to build a multi-objective pressure control model with valve opening as the decision variable, with the pressure balance and pressure reliability of each node in the multi-branch self-pressurized irrigation network as the objectives. The pressure balance is measured by the standard deviation of the pressure of each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy of each node in the network. Solver module 33 is used to solve the multi-objective pressure control model based on the valve hydraulic loss coefficient using the NSGA-II-MOPSO serial hybrid algorithm to obtain the Pareto optimal solution set; The control module 34 is used to select one set of solutions from the Pareto optimal solution set as the valve opening control scheme based on the preset engineering control requirements, and to control the valve opening.

[0099] In one alternative implementation, the device specifically includes: (1) Data acquisition and sensing layer: It consists of pressure sensors, flow meters and valve opening feedback devices deployed at key nodes of the pipeline network to collect pipeline network operation status data in real time.

[0100] (2) Edge computing and control layer: including local controllers or industrial gateways, responsible for receiving sensor data, executing the issued control commands, driving valve actions, and having basic local safety interlock control functions.

[0101] (3) Cloud-based optimization and decision-making layer: Deployed on a server or cloud platform, containing the following core software modules: a. Data Management Module: Stores and manages pipeline topology data, historical operation data, valve characteristic parameters, and optimization case library.

[0102] b. Model Management Module: Maintains and configures multi-objective pressure control models.

[0103] c. Optimization Engine Module: Integrates and executes the NSGA-II-MOPSO hybrid optimization algorithm.

[0104] d. Communication Service Module: Responsible for secure data interaction and command issuance with the edge control layer.

[0105] e. Human-Computer Interaction Layer: Provides a web interface or mobile application for visually displaying the pipeline network status, optimization process, and final decision-making scheme, and allows engineers to set parameters, select schemes, and manually intervene.

[0106] Specifically, while precisely adjusting each valve to the target opening degree, the data acquisition and sensing layer continuously feeds back the adjusted pipeline pressure and flow status to the cloud, forming a closed-loop monitoring system and providing a data foundation for the next optimization cycle. This ultimately achieves a systematic and optimized adjustment of the pipeline pressure distribution. This forms a closed-loop intelligent control circuit of "perception-optimization-decision-execution," capable of repeatedly initiating the optimization process based on real-time data or preset cycles, realizing dynamic and adaptive management of the pressure in the self-pressurized irrigation pipeline network.

[0107] The pressure regulating device for multi-branch self-pressurized irrigation networks provided in this embodiment of the invention can execute the pressure regulating method for multi-branch self-pressurized irrigation networks provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0108] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0109] The following is a detailed reference. Figure 4 This diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 11, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 12 or a program loaded from memory 18 into random access memory (RAM) 13. The RAM 13 also stores various programs and data required for the operation of the electronic device. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0110] Typically, the following devices can be connected to I / O interface 15: input devices 16 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 17 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 18 including, for example, magnetic tapes, hard disks, etc.; and communication devices 19. Communication device 19 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0111] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 19, or installed from a memory 18, or installed from a ROM 12. When the computer program is executed by the processor 11, it performs the functions defined in the pressure regulation method for multi-branch self-pressurized irrigation networks according to embodiments of the present invention.

[0112] Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0113] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the pressure regulation method for multi-branch self-pressurized irrigation networks shown in the above embodiments is implemented.

[0114] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0115] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for pressure regulation in a multi-branch self-pressurized irrigation network, characterized in that, The method includes: The hydraulic loss coefficient of the valve under different valve openings was determined by experiment. With the pressure balance and pressure reliability of each node in a multi-branch self-pressurized irrigation network as the objectives, a multi-objective pressure control model is constructed with valve opening as the decision variable. The pressure balance is measured by the standard deviation of the pressure at each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy at each node in the network. Based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model and obtain the Pareto optimal solution set. Based on the preset engineering control requirements, one set of solutions is selected from the Pareto optimal solution set as the valve opening control scheme to control the valve opening.

2. The method according to claim 1, characterized in that, The multi-objective pressure control model also includes node pressure constraints, energy conservation constraints, pipeline flow velocity constraints, and valve opening constraints. The node pressure constraints include minimum and maximum node pressure constraints. The energy conservation constraint includes the head difference between any two nodes being equal to the friction head loss and local head loss of the pipe section between them, wherein the local head loss is determined by the valve hydraulic loss coefficient. The pipeline flow velocity constraints include minimum and maximum flow velocity constraints. The valve opening constraint includes minimum and maximum valve opening constraints.

3. The method according to claim 1, characterized in that, The hydraulic loss coefficient of the valve was determined experimentally for different valve opening degrees, including: Experiments were conducted to obtain data on inlet and outlet pressures and flow rates under different inlet pressures and valve openings. The actual head loss is calculated based on the inlet and outlet pressures and flow data. Based on the actual head loss, determine the valve hydraulic loss coefficient under different inlet pressures and different valve openings; The valve hydraulic loss coefficient corresponding to different valve openings is determined by the average value of the valve hydraulic loss coefficient at different inlet pressures under any valve opening.

4. The method according to claim 1, characterized in that, Based on the valve hydraulic loss coefficient, the NSGA-II-MOPSO serial hybrid algorithm is used to solve the multi-objective pressure control model to obtain the Pareto optimal solution set, including: Based on the valve hydraulic loss coefficient corresponding to different valve openings, and combined with preset boundary conditions, the pressure value of each node is calculated. Based on the pressure values ​​of each node, the NSGA-II algorithm is used to solve the multi-objective pressure control model to obtain the initial Pareto optimal solution set. Select an initial candidate elite set from the initial Pareto optimal solution set; The initial candidate elite set is optimized using the MOPSO algorithm to obtain the Pareto optimal solution set.

5. The method according to claim 4, characterized in that, Selecting an initial candidate elite set from the initial Pareto optimal solution set includes: When the number of solutions in the initial Pareto optimal solution set does not exceed a preset threshold, the initial Pareto optimal solution set is used as the initial candidate elite set. When the number of solutions in the initial Pareto optimal solution set exceeds a preset threshold, a preset threshold number of solutions are selected from the initial Pareto optimal solution set as an initial candidate elite set.

6. The method according to claim 5, characterized in that, Selecting a predetermined threshold number of solutions from the initial Pareto optimal solution set as an initial candidate elite set includes: The convergence index is determined based on the distance from each solution in the initial Pareto optimal solution set to the preset ideal point; The distribution index is determined based on the crowding distance of each solution in the initial Pareto optimal solution set; The solutions in the initial Pareto optimal solution set are scored based on the convergence index and the distribution index, and a preset threshold number of solutions are selected as the initial candidate elite set based on the scoring results.

7. The method according to claim 1, characterized in that, The method is triggered by a preset period or a preset operating condition event.

8. A pressure regulating device for a multi-branch self-pressurized irrigation network, characterized in that, The device includes: The test module is used to determine the valve hydraulic loss coefficient corresponding to different valve opening degrees through experiments; The model building module is used to construct a multi-objective pressure control model with valve opening as the decision variable, with the pressure balance and pressure reliability of each node in a multi-branch self-pressurized irrigation network as the objectives. The pressure balance is measured by the standard deviation of the pressure of each node in the network, and the pressure reliability is measured by the average value of the pressure redundancy of each node in the network. The solution module is used to solve the multi-objective pressure control model based on the valve hydraulic loss coefficient using the NSGA-II-MOPSO serial hybrid algorithm to obtain the Pareto optimal solution set; The control module is used to select one set of solutions from the Pareto optimal solution set as the valve opening control scheme based on the preset engineering control requirements, and to control the valve opening.

9. An electronic device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the pressure regulation method for a multi-branch self-pressurized irrigation network as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the pressure regulation method for a multi-branch self-pressurized irrigation network as described in any one of claims 1 to 7.

Images