Self-adaptive pipe network dynamic hydraulic balance online optimization method

Abstract

The invention relates to the technical field of self-adaptive pipe network dynamic hydraulic balance online optimization, and discloses a self-adaptive pipe network dynamic hydraulic balance online optimization method. By constructing a node-pipeline topology model and defining a control variable and a threshold value, data input standardization is realized; the opening degree and flow of the valve are collected in real time, a hydraulic impedance and spectrum matrix is constructed, and the dynamic working condition is accurately reflected; decomposing the spectrum matrix, extracting global and local oscillation mode features, and endowing differentiated weights for adjustment; a residual error is calculated based on the water demand and the flow, a tolerance threshold and projection verification are set, and subsequent operation is triggered only during real unbalance; the opening adjustment direction is determined according to spectral analysis and sensitivity evaluation, and blind movement is avoided; the optimal step length is analyzed and solved, amplitude limiting is carried out, and convergence speed and impact suppression are both considered; the valve opening is updated according to branches and issued in a closed loop mode, continuous online optimization and automatic control are achieved, and the real-time performance, the stability and the intelligent level of a pipe network are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of xxxxx, and in particular to an adaptive online optimization method for dynamic hydraulic balance of a pipe network. Background Art

[0002] With the acceleration of urbanization, the scale of urban water distribution networks continues to expand, the number of nodes and pipeline lengths have doubled, and the network's operating state has exhibited strong time-varying, nonlinear, and distributed characteristics. Traditional network hydraulic balance optimization relies on static design models or empirical rules, including pump station scheduling based on empirical curves, network energy-saving optimization based on linear programming, and offline optimization based on heuristic algorithms (such as genetic algorithms and ant colony algorithms). These methods typically treat the network as a static flow system, assuming relatively gentle fluctuations in demand and pressure. This ignores the dynamic characteristics of the water supply system, such as rapid changes in node demand, water hammer effects, and lags in valve opening adjustment.

[0003] The commonly used EPANET software in existing technologies provides hydraulic simulation and pollutant tracking capabilities, but its solver primarily targets network equations under steady-state or quasi-steady-state conditions, making it difficult to meet the requirements of online optimization in the millisecond to second range. Tightly coupling simulation and optimization for real-time control can be computationally intensive and, limited by the accuracy of the network model and boundary conditions, can lead to convergence difficulties or distorted solutions due to initial condition deviations. Some studies have attempted to apply model predictive control (MPC) to water distribution networks, but the high-dimensional variables and complex constraints make online solutions difficult to complete in a short timeframe, and the system lacks adaptability to system disturbances and measurement noise. Furthermore, existing dynamic balancing methods often rely on empirical fitting of flow-pressure relationships or consider only a single objective (such as minimizing energy consumption or equalizing pressure), ignoring the multimodal oscillation characteristics of the network and the coupled effects of valve openings. Consequently, they cannot respond quickly to sudden changes in demand or water quality, leading to risks such as node overpressure and pipeline siltation. At the same time, traditional adjustment methods lack a unified online monitoring and graded alarm mechanism, and cannot perform effective residual tolerance testing before valve adjustment. Once the adjustment direction is wrong, it may oscillate repeatedly and fail to converge to the desired balance.

[0004] To this end, this project aims to propose an adaptive online optimization method for dynamic hydraulic balance in pipeline networks. This method digitally models the network nodes and pipeline structure, constructs a hydraulic impedance matrix and a node coupling matrix using real-time valve opening and flow rate data, and performs spectral decomposition on these matrices to reveal the system's oscillation modes and energy distribution characteristics. By combining node-level residual calculation with a projection verification strategy, the method accurately determines the degree of supply-demand imbalance and modal importance, then allocates the direction and amplitude of flow adjustments according to the weights of different modes. Through a linear step-size analytical solution and amplitude limiting constraints, the method achieves precise and stable valve opening updates, ultimately establishing a closed-loop online control mechanism. Summary of the Invention

[0005] The present invention provides an adaptive online optimization method for dynamic hydraulic balance of a pipe network, which helps solve the problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: an adaptive online optimization method for dynamic hydraulic balance of a pipe network, comprising: S1. Establish a pipe network topology structure including nodes and pipelines, define pipe network variables and related thresholds, and form a basic data model that can be used for dynamic analysis; S2, real-time collection of operating parameters of each pipeline in the pipe network, and construction of hydraulic impedance matrix and node spectrum matrix based on the collected data; S3, perform spectral decomposition on the node spectrum matrix and extract eigenvalues ​​and corresponding eigenvectors; S4. Calculate the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment; S5. Set a tolerance threshold for the calculated residual, perform tolerance determination and projection verification. If the residual exceeds the tolerance range, perform projection analysis and verify the system response under each mode. If it fails, issue an alarm and perform eigendecomposition analysis again. S6. Determine the direction of valve opening adjustment based on the spectrum analysis results and the driving flow adjustment process, analyze the sensitivity of pipeline flow to valve opening, and calculate the adjustment amount and direction; S7. By analyzing and calculating the step length and setting the limit, the optimal solution and range constraint of the adjustment step length are obtained. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned to recalculate; if the step length is within the preset range, the opening adjustment is performed; S8. Based on the adjustment results, the valve opening is updated according to the branch conditions, and the latest control quantity is sent to the corresponding valve.

[0007] Optionally, establishing a pipe network topology structure including nodes and pipelines, defining pipe network variables and related thresholds, and forming a basic data model that can be used for dynamic analysis specifically includes: Set the discrete control period index to ;in, is a set of non-negative integers; Get the total number of pipe network nodes, recorded as ; Set the pipe network node collection to ; Get the total number of valves in the pipeline, record it as ; Set the pipeline with valve collection to ;in, is a set of positive integers; Construct a directed incidence matrix: ; in, , used to describe the connectivity relationship between nodes and pipelines; is the node index, and its value range is ; The pipeline contains the valve index; Get pipeline The nominal hydraulic resistance constant is expressed as ; Set the minimum allowable valve opening to ; Initialize valve opening vector ;in, For the moment pipeline The valve opening ratio, .

[0008] Optionally, the real-time collection of operating parameters of each pipeline in the pipe network and the construction of a hydraulic impedance matrix and a node spectrum matrix based on the collected data specifically include: Read valve opening , forming the vector ; Computational pipeline At the moment Instant hydraulic resistance ; Constructing the impedance diagonal matrix ;in, Place the elements in the brackets on the diagonal in order and set the rest to zero; for OK A real matrix of columns; Construct the node spectrum matrix: ;in, for OK A real matrix of columns; is the Laplace matrix of the node coupling.

[0009] Optionally, performing spectral decomposition on the node spectral matrix to extract eigenvalues ​​and corresponding eigenvectors specifically includes: Find spectral decomposition: , ;in, For the modal eigenvalues; is the corresponding unit eigenvector; Sort the eigenvalues ​​in ascending order: .

[0010] Optionally, the calculation of the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment specifically includes: The moment node Water demand is recorded as ,time pipeline The flow rate is recorded as ; Compute nodes At the moment The flow residual ; All residuals are combined into a vector .

[0011] Optionally, a tolerance threshold is set for the calculated residual, and tolerance determination and projection verification are performed. If the residual exceeds the tolerance range, projection analysis is performed and the system response under each mode is verified. If it fails, an alarm is issued and the characteristic decomposition analysis is performed again, specifically including: Assume the residual tolerance threshold is ,and ; like , then jump to step S8; otherwise, calculate the residual in the Projection coefficient under mode ; Assume that the projection reconstruction tolerance is ,and ; like , then alarm and return to step S4; otherwise, go to step S6.

[0012] Optionally, the method of determining the opening adjustment direction based on the spectrum analysis results and the driving flow adjustment process, analyzing the sensitivity of the pipeline flow to the valve opening, and calculating the adjustment amount and adjustment direction specifically includes: Calculate the normalization constant ; like , then the alarm spectrum matrix is ​​all zero, there is no valid pattern, and the alarm is terminated; otherwise, the following calculations are performed: Computational pipeline Original traffic shaping : ; Let the minimum sensitivity threshold be ,and ; Computational pipeline Flow rate to opening sensitivity ; Calculate the original opening adjustment ; Calculate the maximum absolute opening adjustment ; Calculate the normalized opening adjustment direction : ;like , alarm that there is no valid direction, and return to step S4.

[0013] Optionally, the step length is calculated analytically and the limit is set to optimize the adjustment step length and set the range constraint. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned for recalculation; if the step length is within the preset range, the opening adjustment is performed, specifically including: Calculate residuals by direction The projection increment ; Set the objective function to ;in, is the step size factor; Find the optimal step size ; Let the maximum step length be ,and ; like , then alarm and return to step S4; like , then take ;in, The step size factor for opening adjustment.

[0014] Optionally, the valve opening is updated according to the branch conditions based on the adjustment result, and the latest control quantity is sent to the corresponding valve, specifically including: like , then let ; Otherwise, let ; Will Send to valve for execution; make , return to step S2 and continue online optimization.

[0015] The present invention has the following beneficial effects: 1. By accurately depicting the relationship between the nodes and pipelines of the water supply and drainage network in the form of a directed association matrix, and simultaneously defining the various variables and thresholds required for the operation of the network, an innovative unified modeling of topological information and control parameters is achieved. Unlike the traditional reliance on manual drawing or simple table maintenance, this matrix method stores the network scale, number of nodes, and valve pipeline information in a structured manner, avoiding parameter omissions and format inconsistencies, and providing clear and unified basic data for subsequent matrix operations and spectral decomposition. On the one hand, standardized data models reduce deployment and maintenance costs and improve applicability to large-scale pipeline networks; on the other hand, precise network mapping ensures the correctness and stability of subsequent operations, solving the adjustment errors caused by topology inconsistencies at the source and providing a solid data foundation for the entire online optimization.

[0016] 2. In traditional pipe network analysis, hydraulic calculations mostly rely on static models or periodic offline analysis, lacking real-time response to dynamic changes in valve opening. The solution innovatively combines the real-time collected valve opening data with the pipeline resistance constant, and constructs the hydraulic impedance diagonal matrix and the node coupling spectrum matrix through online calculation. It not only accurately reflects the dynamic characteristics of each pipeline resistance as the opening changes, but also fully describes the flow-pressure coupling relationship between nodes. Real-time matrix construction realizes online hydraulic model updates, which can accurately reflect the on-site working conditions without offline correction; at the same time, the spectrum matrix provides complete input for subsequent spectral decomposition, ensuring the accuracy of feature extraction. Compared with existing static or semi-dynamic models, this link improves the timeliness and precision of pipe network hydraulic analysis, and solves the problems of system response lag and model error accumulation.

[0017] 3. Traditional pipe network optimization often only adjusts the entire network as a whole, making it difficult to identify flow and pressure fluctuation patterns in different regions or frequency bands. The solution innovatively performs spectral decomposition on the node coupling spectrum matrix, and achieves a dual characterization of the system's low-frequency, large-scale fluctuations and high-frequency local disturbances by extracting the characteristic intensities and corresponding vectors of multiple oscillation modes. On the one hand, the residuals can be projected and distributed according to different modes, so that subsequent adjustments can give greater weight to important modes; on the other hand, it can distinguish between global and local effects, avoiding local congestion or inability to make targeted adjustments due to ignoring high-frequency disturbances, thereby improving the accuracy and stability of the adjustment strategy. Compared with existing methods that rely solely on empirical rules or single indicators, this link achieves an in-depth analysis of the dynamic behavior of the pipe network, truly achieving regulation based on local conditions and models.

[0018] 4. The solution proposes to calculate the flow surplus of each node based on the real-time node water demand and pipeline flow data, and form all the surpluses into a residual vector to quantitatively characterize the supply and demand balance deviation of each node in the entire network. Unlike the traditional method of only counting the overall water volume difference or judging based on a single node, this method achieves node-level precision and can clearly identify the specific location and degree of hydraulic imbalance. It provides accurate input data for subsequent tolerance judgment and modal projection; at the same time, it helps operators quickly locate problem areas and helps to formulate more targeted adjustment strategies; compared with the existing methods that rely on pipeline network simulation or experience judgment, it improves the early warning efficiency and fault location speed, and solves the problem of difficulty in timely detection and quantification of node imbalance in the past.

[0019] 5. To avoid frequent adjustments due to measurement noise or slight fluctuations, this solution innovatively sets a tolerance threshold for node residuals and verifies them through modal projection reconstruction when the limit is exceeded to determine whether to enter the adjustment process. This can not only filter out false alarms caused by daily minor fluctuations, reducing unnecessary computational burden and equipment wear on the system; it also judges reconstruction errors through modal projection, ensuring that adjustments are only triggered when there is a real imbalance and the numerical quality is reliable, thereby improving the robustness of the algorithm. Compared with existing methods that generally use fixed thresholds or no verification mechanism, this link takes into account both sensitivity and stability, and solves the problem of frequent adjustments or missed adjustments.

[0020] 6. The solution combines the residual projection results of the modal components and the corresponding modal weights to calculate the original flow adjustment amount of each pipeline, and determines the adjustment direction based on the sensitivity evaluation of the flow to the valve opening. This move breaks through the traditional extensive adjustment method of single flow increase or decrease and realizes targeted flow correction. Sensitivity analysis eliminates invalid or excessive adjustments and ensures the scientific nature of the opening adjustment; at the same time, different adjustment weights are assigned to different modes to balance the global and local optimization goals. Compared with existing technologies that rely on manual experience fitting sensitivity or simple proportional adjustment, this link improves the adjustment effect and system stability, and solves the problem of difficulty in taking into account the needs of various regions in complex pipeline networks.

[0021] 7. To ensure both rapid and stable adjustments, the solution introduces an analytical solution for the first time. This closed-form optimization of the objective function between the residual projection increment and the step size factor yields the optimal step size and sets a maximum opening change limit. This analytical solution reduces the iterative search computational effort, enabling efficient online computation; the limit constraint prevents pressure fluctuations or network shocks caused by single opening changes. Compared to existing methods that often rely on empirical step sizes or simple linear increments, this approach balances convergence speed and safety, resolving the issues of overshoot oscillation and slow convergence that can often occur in online network optimization.

[0022] 8. Based on the aforementioned adjustment results and branch conditions, the valve openings are updated in real time, and control instructions are sent to the actuators, forming a complete online closed-loop control mechanism. Through continuous iteration and real-time feedback, the pipeline network is able to automatically adapt to load changes and maintain hydraulic balance. At the same time, the branch logic is organically combined with threshold judgment, and strategies can be switched under special conditions, further improving the flexibility and safety of the system. Compared with traditional periodic manual scheduling or single offline optimization, this link realizes true all-weather, full-network automated online control, solving the pain points of manual scheduling delays and poor response time. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] Example, see Figure 1 , an adaptive online optimization method for dynamic hydraulic balance of a pipe network, comprising: S1. Establish a pipe network topology structure including nodes and pipelines, define pipe network variables and related thresholds, and form a basic data model that can be used for dynamic analysis; S2, real-time collection of operating parameters of each pipeline in the pipe network, and construction of hydraulic impedance matrix and node spectrum matrix based on the collected data; S3, perform spectral decomposition on the node spectrum matrix and extract eigenvalues ​​and corresponding eigenvectors; S4. Calculate the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment; S5. Set a tolerance threshold for the calculated residual, perform tolerance determination and projection verification. If the residual exceeds the tolerance range, perform projection analysis and verify the system response under each mode. If it fails, issue an alarm and perform eigendecomposition analysis again. S6. Determine the direction of valve opening adjustment based on the spectrum analysis results and the driving flow adjustment process, analyze the sensitivity of pipeline flow to valve opening, and calculate the adjustment amount and direction; S7. By analyzing and calculating the step length and setting the limit, the optimal solution and range constraint of the adjustment step length are obtained. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned to recalculate; if the step length is within the preset range, the opening adjustment is performed; S8. Based on the adjustment results, the valve opening is updated according to the branch conditions, and the latest control quantity is sent to the corresponding valve.

[0026] By establishing a network topology structure including nodes and pipelines and defining control variables and thresholds, the problem that traditional network online optimization is difficult to unify the basic data model and leads to inconsistent algorithm input is solved; by collecting operating parameters such as valve opening and pipeline flow in real time and constructing hydraulic impedance matrix and node coupling spectrum matrix, the deficiency that static model cannot reflect the changes in on-site working conditions is compensated, and the accurate description of the dynamic characteristics of the network is achieved; by spectral decomposition of the spectrum matrix and extracting modal eigenvalues ​​and eigenvectors, the problem that previous optimization methods cannot distinguish between global large-scale fluctuations and local small-scale disturbances is solved, so that adjustment measures can give differentiated weights to different oscillation modes; by The steps of calculating the residual of flow data and setting tolerance thresholds and projection verification for the residuals filter out measurement noise and slight fluctuations, avoid the risk of over-adjustment or underreporting, and ensure that subsequent processes are triggered only when there is a real imbalance; by using the spectral decomposition results to drive flow adjustment and opening direction determination, scientific direction guidance and sensitivity analysis are achieved, avoiding system oscillations caused by blindly increasing or decreasing flow; by analytically solving the step size and setting the limit constraint, the optimal opening increment is efficiently obtained, which not only speeds up the convergence speed but also prevents the impact of large-scale adjustments on the pipeline network; finally, by updating the valve opening under branch conditions and issuing control instructions in a closed loop, continuous online optimization and automated closed-loop control are achieved. Through the seamless connection of the above links, the problems of delayed response, poor adjustment accuracy, and easy oscillation of traditional pipeline network optimization are solved, the real-time performance, stability, and intelligence level of the system are improved, which is beneficial to energy saving, loss reduction, and safe and reliable operation of urban pipeline networks.

[0027] The establishment of a pipe network topology structure including nodes and pipelines, definition of pipe network variables and related thresholds, and formation of a basic data model that can be used for dynamic analysis specifically include: Set the discrete control period index to ;in, is a set of non-negative integers; defines the time scale of discrete control for subsequent data collection and iteration; Get the total number of pipe network nodes, recorded as ; Set the pipe network node collection to ; Get the total number of valves in the pipeline, record it as ; Set the pipeline with valve collection to ;in, is a set of positive integers; determine the system scale, and subsequent matrix dimensions are based on and ; Construct a directed incidence matrix: ; in, , used to describe the connectivity relationship between nodes and pipelines; is the node index, and its value range is ; The pipeline contains the valve index; the connection direction between each pipeline and two nodes is accurately described in matrix form to facilitate subsequent matrix operations; Get pipeline The nominal hydraulic resistance constant is expressed as Quantify pipeline physical properties for calculating hydraulic impedance; Set the minimum allowable valve opening to ; Initialize valve opening vector ;in, For the moment pipeline The valve opening ratio, ; Ensure that there is still minimum flow when the valve is closed and initialize the system.

[0028] By setting the discrete control cycle index and time scale to unify the steps of each data collection and iteration sequence, the control confusion and data mismatch problems caused by unclear division of the control cycle are solved; by obtaining and defining the total number of pipeline nodes and the total number of valves in the pipeline, and constructing the corresponding node set and pipeline set, the system scale and matrix dimension are accurately determined, avoiding the calculation crash caused by the error in model scale estimation; by describing the connection relationship between nodes and pipelines in the form of a directed association matrix, the current situation of scattered and difficult automated processing of traditional topological information is overcome, providing standardized and verifiable topological structure data for subsequent matrix operations; by obtaining The steps of determining the nominal hydraulic resistance constant of the pipeline and quantifying the physical characteristics of the pipeline solve the calculation deviation caused by the lack of resistance parameters or assumption errors, and achieve objective and accurate resistance estimation; by setting the minimum allowable valve opening and initializing the valve opening vector, the hidden dangers of the valve being completely closed, resulting in no flow in the pipeline network or control failure, are avoided, while ensuring that the system has basic flow capacity when it is first started; through the above steps, discrete timing management, topology structure modeling, physical parameter quantification and valve initialization are organically combined to form a systematic and repeatable basic data model, which provides solid and reliable data support for the entire online optimization process and improves the robustness and applicability of the algorithm.

[0029] The real-time collection of operating parameters of each pipeline in the pipe network and the construction of a hydraulic impedance matrix and a node spectrum matrix based on the collected data specifically include: Read valve opening , forming the vector ; Get the current pipeline opening for subsequent impedance calculation; Computational pipeline At the moment Instant hydraulic resistance ;Describe the real-time impact of valve adjustment on pipeline impedance; Constructing the impedance diagonal matrix ;in, Place the elements in the brackets on the diagonal in order and set the rest to zero; for OK The impedance of each pipeline is integrated into a diagonal matrix to facilitate unified calculation; Construct the node spectrum matrix: ;in, for OK A real matrix of columns; is the Laplace type matrix of the coupling between nodes; forming the Laplace type coupling matrix between nodes, preparing for the subsequent spectral decomposition.

[0030] By reading valve openings and forming a unified vector, centralized management of the real-time opening status of each pipeline is achieved, solving the problem of subsequent calculation confusion caused by decentralized data storage. By calculating the pipeline's instantaneous hydraulic resistance, the impact of valve adjustment on resistance changes is clearly quantified, overcoming the shortcomings of traditional models in which resistance estimation often relies on experience or simplified assumptions. By constructing a diagonal resistance matrix to efficiently integrate the impedance data of each pipeline, a structured and directly operable input is provided for subsequent matrix operations, avoiding the performance waste of multiple loop reading and matching. By constructing a node spectrum matrix to describe the coupling relationship between nodes, the hydraulic impedance and node coupling information are innovatively integrated to form an efficient matrix suitable for spectral decomposition, solving the limitation that constructing a coupling matrix or impedance matrix alone cannot obtain overall dynamic information. The above steps enable real-time extraction and efficient integration of pipeline network operating parameters, providing an accurate and complete input matrix for subsequent spectral decomposition and feature extraction, and improving the response speed and analytical accuracy of the dynamic hydraulic model to changing operating conditions.

[0031] The spectral decomposition of the node spectrum matrix and the extraction of eigenvalues ​​and corresponding eigenvectors specifically include: Find spectral decomposition: , ;in, For the Modal eigenvalue, also called modal strength or oscillation frequency index, physically quantifies the The strength / rigidity of a network oscillation mode or fluid-pressure coupling mode, small The mode corresponds to low-frequency, large-scale flow-pressure fluctuations; Corresponding to high-frequency, small-scale local fluctuations; used in the scheme to assign different adjustment weights to the residual projection, Make the high-frequency mode step size smaller to ensure system stability; is the corresponding unit eigenvector; extract the system's inherent oscillation mode and its intensity; Sort the eigenvalues ​​in ascending order: ; Organize modal priorities from low to high by frequency or energy.

[0032] By performing spectral decomposition on the node spectral matrix and extracting modal eigenvalues ​​and corresponding eigenvectors, the network oscillation mode and intensity are directly linked to the intrinsic dynamic behavior of the system for the first time, solving the defect of traditional optimization that only focuses on a single indicator and has difficulty in distinguishing different fluctuation modes; by distinguishing between low-frequency large-scale fluctuations and high-frequency local disturbances, it is effectively ensured that the adjustment strategy can deal with network-wide and local imbalances respectively in a targeted manner, avoiding local congestion or network-wide oscillation caused by single-frequency band control; by sorting the modal eigenvalues ​​from low to high according to energy or frequency and organizing their priorities, flexible calling of each oscillation mode in the adjustment weight distribution is achieved, thereby improving the regulation efficiency and stability; through the above steps, the physical information contained in the spectral matrix is ​​deeply mined, providing a scientific basis for subsequent residual projection and adjustment decisions, improving the accuracy and reliability of pipeline network optimization, and is an important breakthrough in intelligent pipeline network control.

[0033] The calculation of the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment specifically includes: The moment node Water demand is recorded as ,time pipeline The flow rate is recorded as ; Compute nodes At the moment The flow residual ; All residuals are combined into a vector ; Quantify the current supply and demand imbalance of the node and provide a basis for adjustment.

[0034] By calculating the node flow residuals based on the real-time data of each node's water demand and pipeline flow, a precise quantification of the supply and demand deviation of the entire network is achieved, solving the problem that traditional methods can only roughly estimate the overall difference but cannot locate the imbalanced nodes; by unifying the calculated residuals into residual vectors, structured data input is provided for subsequent tolerance judgment and modal projection, avoiding the efficiency loss of multiple scattered calculations and data reorganization; through the above steps, the field measurement data is closely integrated with the model calculation, which not only can timely detect the supply and demand imbalance of each node, but also provide an accurate basis for subsequent adjustment strategies, improve the problem location speed and data processing efficiency, and is conducive to achieving efficient online closed-loop optimization.

[0035] The tolerance threshold is set for the calculated residual, and tolerance determination and projection verification are performed. If the residual exceeds the tolerance range, projection analysis is performed and the system response under each mode is verified. If it fails, an alarm is issued and the characteristic decomposition analysis is performed again, which specifically includes: Assume the residual tolerance threshold is ,and ; Effect: If , it is considered that the node supply and demand balance is within an acceptable range, and the opening adjustment of this cycle is skipped. Value impact: too small, frequent triggering of adjustments, system oscillation, and large amount of calculation; too large: obvious imbalances may be ignored, affecting water supply quality. Value basis: based on the scale of the pipe network and the range of daily flow fluctuations. Value recommendation: take the daily maximum flow of a typical node For example, the maximum demand of a node , optional .

[0036] like , then jump to step S8, if the residual is small enough, skip the adjustment to save resources; otherwise, calculate the residual in step S8. Projection coefficient under mode ; Decompose the residual into each mode and quantify the contribution of each mode; Assume that the projection reconstruction tolerance is ,and ; Function: Test spectral projection and the true residual If the error exceeds If the value is too small, it will be strictly checked and prone to frequent recalculation due to numerical noise; too large, the pattern reconstruction error may be ignored, resulting in wrong direction adjustment. Value selection basis: based on the numerical calculation accuracy and residual magnitude. Value recommendation: take the typical value of the residual like average ,but .

[0037] like , an alarm is issued and the process returns to step S4; otherwise, the process goes to step S6; ensuring the numerical accuracy and preventing abnormal data from causing misadjustment.

[0038] By setting a reasonable tolerance threshold for the residual vector and performing tolerance judgment, false alarms caused by measurement errors and short-term fluctuations are effectively filtered out, and the problems of system oscillation and waste of computing resources caused by frequent ineffective adjustments are solved; by performing modal projection reconstruction and verifying the projection error when the residual exceeds the limit, it is ensured that subsequent optimization is triggered only when the modal information is reliable and the imbalance is real, avoiding erroneous adjustments caused by abnormal points or noise data; through the dual control mechanism of tolerance judgment and projection verification, strict control of adjustment triggering conditions is achieved, the robustness and stability of the algorithm are enhanced, the risk of misadjustment and missed adjustment is reduced, and reliable quality assurance is provided for online optimization of the pipeline network.

[0039] The method of determining the opening adjustment direction based on the spectrum analysis results and the driving flow adjustment process, analyzing the sensitivity of the pipeline flow to the valve opening, and calculating the adjustment amount and adjustment direction specifically includes: Calculate the normalization constant ; Normalize the damping of each mode to make it dimensionless; like , then the alarm spectrum matrix is ​​all zero, there is no valid mode, and it is terminated to avoid division by zero and invalid adjustment; otherwise, calculate in sequence: Computational pipeline Original traffic shaping : ; Calculate the flow correction according to the weight of each mode and the residual projection; Let the minimum sensitivity threshold be ,and ; Function: When the flow rate is extremely small or the opening is close to zero, It will not be zero, avoid division by zero or stagnation of direction. Impact of value: If it is too small, the original formula will be kept for calculation, which may result in division by zero; if it is too large, the slight flow change will be artificially magnified, which may lead to misadjustment. Basis for value selection: Based on the minimum measurable flow accuracy on site. Value recommendation: Slightly higher than the minimum resolution of the flow sensor, such as resolution ,but .

[0040] Computational pipeline Flow rate to opening sensitivity ; Characterize the linear approximate gain of flow opening to prevent zero flow failure; Calculate the original opening adjustment ;Convert flow adjustment into opening adjustment to ensure consistent dimensions; Calculate the maximum absolute opening adjustment ; Calculate the normalized opening adjustment direction : ;like , alarm no valid direction, and return to step S4; extract direction information, limit to , ensuring that the step size is applied consistently.

[0041] By driving the flow adjustment direction based on the spectral decomposition results and combining the steps of analyzing the sensitivity of pipeline flow to valve opening, targeted and scientific adjustment guidance is achieved, solving the problems of blind adjustment and easy oscillation in traditional experience or single proportional regulation; by calculating the original flow adjustment amount and converting it into the opening adjustment amount, the flow target and valve control are effectively connected, avoiding execution errors caused by numerical mismatch; through the above steps, the adjustment needs of each pipeline are accurately grasped and appropriate adjustment efforts are given, which not only ensures the promotion effect of hydraulic balance of the entire network, but also prevents pressure shock caused by local excessive adjustment, thereby improving the stability and reliability of system operation.

[0042] The step length is calculated analytically and the limit is set to optimize the adjustment step length and set the range constraint. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned to recalculate; if the step length is within the preset range, the opening adjustment is performed, specifically including: Calculate residuals by direction The projection increment ;Evaluate the linear effect of unit opening adjustment on residuals; Set the objective function to ;in, is the step size factor; define the curve of the residual sum of squares changing with the step size to facilitate analytical minimization; Find the optimal step size ;beg The minimum value point of , obtains the optimal adjustment range; Let the maximum step length be ,and ; Function: Limit the maximum ratio of the opening update per cycle to prevent large steps from causing system overshoot or oscillation. Value impact: If it is too small, it can respond quickly but easily cause pressure fluctuations; if it is too large, the convergence is slow and the system response is delayed. Value basis: Based on the maximum safe opening range of the valve, for example, the valve manufacturer recommends that a single adjustment should not exceed Then take . Value suggestion: first set to , and then make slight adjustments based on the actual response speed and pressure stability.

[0043] like , then alarm and return to step S4; like , then take ;in, The step size factor for the opening adjustment controls the overall opening update amplitude in this cycle to prevent it from being too large (causing oscillation) or too small (ineffective adjustment); avoid system instability or ineffectiveness caused by too large or too small a step size.

[0044] By constructing the objective function of the residual projection increment and the step size factor and solving the analytical optimal step size, closed-form optimization of the adjustment amplitude is achieved, solving the problems of large computational complexity and slow convergence of traditional iterative search methods; by setting the maximum opening change limit, pipeline pressure fluctuations and hydraulic shocks caused by one-time large adjustments are avoided; through the above steps, while ensuring adjustment efficiency and taking safety into account, online optimization can not only quickly eliminate imbalances but also maintain system pressure stability, providing efficient and safe guarantees for automated pipeline network control.

[0045] According to the adjustment result, the valve opening is updated according to the branch condition, and the latest control quantity is sent to the corresponding valve, specifically including: like , then let ; Otherwise, let ; Combine step size and direction to adjust or maintain valve opening within a bounded range; Will Send to valve for execution; make , return to step S2 and continue online optimization; Complete a closed loop and enter the next cycle.

[0046] By differentially updating the opening adjustment results according to branch conditions and issuing the steps of the valve actuator in real time, closed-loop control of the adjustment process is achieved, solving the problem of being unable to continuously iterate and correct in real time online; by looping back for subsequent optimization steps, it is ensured that the pipeline network can continuously adapt to changes in water use, avoiding the disadvantages of losing monitoring and control after a one-time adjustment; through a complete online closed-loop scheduling mechanism, the system can operate around the clock and automatically, minimizing human intervention while ensuring water supply quality, thereby improving the intelligence level and operating efficiency of the pipeline network.

[0047] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0048] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. An adaptive online optimization method for dynamic hydraulic balance of a pipe network, characterized in that: include: S1. Establish a pipe network topology structure including nodes and pipelines, define pipe network variables and related thresholds, and form a basic data model that can be used for dynamic analysis; S2, real-time collection of operating parameters of each pipeline in the pipe network, and construction of hydraulic impedance matrix and node spectrum matrix based on the collected data; S3, perform spectral decomposition on the node spectrum matrix and extract eigenvalues ​​and corresponding eigenvectors; S4. Calculate the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment; S5. Set a tolerance threshold for the calculated residual, perform tolerance determination and projection verification. If the residual exceeds the tolerance range, perform projection analysis and verify the system response under each mode. If it fails, issue an alarm and perform eigendecomposition analysis again. S6. Determine the direction of valve opening adjustment based on the spectrum analysis results and the driving flow adjustment process, analyze the sensitivity of pipeline flow to valve opening, and calculate the adjustment amount and direction; S7. By analyzing and calculating the step length and setting the limit, the optimal solution and range constraint of the adjustment step length are obtained. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned to recalculate; if the step length is within the preset range, the opening adjustment is performed; S8. Based on the adjustment results, the valve opening is updated according to the branch conditions, and the latest control quantity is sent to the corresponding valve.

2. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 1, characterized in that: The establishment of a pipe network topology structure including nodes and pipelines, definition of pipe network variables and related thresholds, and formation of a basic data model that can be used for dynamic analysis specifically include: Set the discrete control period index to ;in, is a set of non-negative integers; Get the total number of pipe network nodes, recorded as ; Set the pipe network node collection to ; Get the total number of valves in the pipeline, record it as ; Set the pipeline with valve collection to ;in, is a set of positive integers; Construct a directed incidence matrix: ； in, , used to describe the connectivity relationship between nodes and pipelines; is the node index, and its value range is ; The pipeline contains the valve index; Get pipeline The nominal hydraulic resistance constant is expressed as ; Set the minimum allowable valve opening to ; Initialize valve opening vector ;in, For the moment pipeline The valve opening ratio, .

3. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 2, characterized in that: The real-time collection of operating parameters of each pipeline in the pipe network and the construction of a hydraulic impedance matrix and a node spectrum matrix based on the collected data specifically include: Read valve opening , forming the vector ; Computational pipeline At the moment Instant hydraulic resistance ; Constructing the impedance diagonal matrix ;in, Place the elements in the brackets on the diagonal in order and set the rest to zero; for OK A real matrix of columns; Construct the node spectrum matrix: ;in, for OK A real matrix of columns; is the Laplace matrix of the node coupling.

4. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 3, characterized in that: The spectral decomposition of the node spectrum matrix and the extraction of eigenvalues ​​and corresponding eigenvectors specifically include: Find spectral decomposition: , ;in, For the modal eigenvalues; is the corresponding unit eigenvector; Sort the eigenvalues ​​in ascending order: .

5. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 4, characterized in that: The calculation of the hydraulic flow residual of each node based on the water demand and pipeline flow data of each node at the current moment specifically includes: The moment node Water demand is recorded as ,time pipeline The flow rate is recorded as ; Compute nodes At the moment The flow residual ; All residuals are combined into a vector .

6. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 5, characterized in that: The tolerance threshold is set for the calculated residual, and tolerance determination and projection verification are performed. If the residual exceeds the tolerance range, projection analysis is performed and the system response under each mode is verified. If it fails, an alarm is issued and the characteristic decomposition analysis is performed again, which specifically includes: Assume the residual tolerance threshold is ,and ; like , then jump to step S8; otherwise, calculate the residual in the Projection coefficient under mode ; Assume that the projection reconstruction tolerance is ,and ; like , then alarm and return to step S4; otherwise, go to step S6.

7. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 6, characterized in that: The method of determining the opening adjustment direction based on the spectrum analysis results and the driving flow adjustment process, analyzing the sensitivity of the pipeline flow to the valve opening, and calculating the adjustment amount and adjustment direction specifically includes: Calculate the normalization constant ; like , then the alarm spectrum matrix is ​​all zero, there is no valid pattern, and the alarm is terminated; otherwise, the following calculations are performed: Computational pipeline Original traffic shaping : ； Let the minimum sensitivity threshold be ,and ; Computational pipeline Flow rate to opening sensitivity ; Calculate the original opening adjustment ; Calculate the maximum absolute opening adjustment ; Calculate the normalized opening adjustment direction : ;like , alarm that there is no valid direction, and return to step S4.

8. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 7, characterized in that: The step length is calculated analytically and the limit is set to optimize the adjustment step length and set the range constraint. If the step length parameter is abnormal, an alarm message is issued and the calculation is returned to recalculate; if the step length is within the preset range, the opening adjustment is performed, specifically including: Calculate residuals by direction The projection increment ; Set the objective function to ;in, is the step size factor; Find the optimal step size ; Let the maximum step length be ,and ; like , then alarm and return to step S4; like , then take ;in, The step size factor for opening adjustment.

9. The method for online optimization of the adaptive dynamic hydraulic balance of a pipe network according to claim 8, characterized in that: According to the adjustment result, the valve opening is updated according to the branch condition, and the latest control quantity is sent to the corresponding valve, specifically including: like , then let ; Otherwise, let ; Will Send to valve for execution; make , return to step S2 and continue online optimization.

Images