Water supply network optimization design method and system based on digital twinborn technology

Abstract

The invention provides a water supply pipe network optimization design method and system based on a digital twinborn technology, and belongs to the technical field of water supply pipe network optimization. The method comprises the steps that a three-dimensional digital model of the water supply network, real-time operation data collected by a pressure sensor and a flowmeter are input into a digital twin model, and the current hydraulic state distribution condition of each pipe section is simulated and calculated through the digital twin model in combination with the real-time operation data, an objective function of water supply network optimization design and the three-dimensional digital model; performing hydraulic characteristic prediction through a preset prediction algorithm in combination with the current hydraulic state distribution condition and the historical operation data of each pipe section; and a hydraulic characteristic error value between the hydraulic characteristic predicted value and the hydraulic characteristic actual value of each pipe section is obtained, pipe network parameters of the water supply pipe network are adjusted according to the hydraulic characteristic error values, and the actual operation state of the water supply pipe network is optimized through a feedback mechanism. By means of the provided optimal design scheme, high efficiency, economical efficiency and reliability of the mountain city water supply pipe network design are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water supply network optimization, in particular to a water supply network optimization design method and system based on digital twin technology. BACKGROUND

[0002] With the acceleration of urbanization, the optimization design of water supply network in mountainous cities has become an important research direction to improve water resource utilization efficiency, reduce operation cost and ensure water supply safety. As a core component of urban infrastructure, the design and operation efficiency of water supply network directly affects the economy, reliability and stability of the water supply system. However, traditional water supply network optimization methods face many challenges in dealing with complex terrain conditions, especially in terms of network topology design, hydraulic property simulation and dynamic optimization adjustment. For example, in mountainous cities, complex terrain features put higher requirements on the layout of the network, while traditional methods often fail to fully consider the impact of terrain elevation on hydraulic properties, resulting in limited economic efficiency and operation efficiency of network design. In addition, existing technical solutions generally lack dynamic monitoring and feedback mechanisms for real-time operation state of the network, making it difficult to adapt to hydraulic property fluctuations caused by terrain changes in actual operation, thereby affecting the stability and reliability of the water supply system. SUMMARY

[0003] To solve the above technical problems, the present application provides a water supply network optimization design method and system based on digital twin technology, which builds a three-dimensional digital model of the water supply network, deploys pressure sensors and flow meters in the physical system, collects real-time operation data, inputs the data into the digital twin model for hydraulic property simulation and prediction, and finally optimizes the actual operation state through a feedback mechanism, thereby realizing the efficiency, economy and reliability of water supply network design.

[0004] In a first aspect, the present application provides a water supply network optimization design method based on digital twin technology, which comprises: building a three-dimensional digital model of the water supply network, the three-dimensional digital model containing terrain elevation information, pipeline layout information and pump station location information; deploying pressure sensors and flow meters in the physical system, and collecting real-time operation data of the water supply network through the pressure sensors and the flow meters; loading the three-dimensional digital model into a digital twin model, and initializing the network parameters of the digital twin model according to the pre-acquired key network data; The real-time operation data is input into the digital twin model, and each pipe section of the water supply network is subjected to hydraulic calculation by the digital twin model in combination with the real-time operation data, a target function of the water supply network optimization design, and the three-dimensional digital model to simulate a current hydraulic state distribution of each pipe section. The hydraulic characteristics of each pipe section are predicted by a preset prediction algorithm in combination with the current hydraulic state distribution and historical operation data of each pipe section to obtain a hydraulic characteristic prediction value of each pipe section. The hydraulic characteristic error value between the hydraulic characteristic prediction value and the actual value of each pipe section is obtained, and the pipe network parameters of the water supply network are adjusted according to the hydraulic characteristic error value, and the actual operation state of the water supply network is optimized through a feedback mechanism.

[0005] In an embodiment, the adjusting the pipe network parameters of the water supply network according to the hydraulic characteristic error value comprises: optimizing the pump station start-stop time and / or the valve opening degree according to the hydraulic characteristic error value.

[0006] In an embodiment, the pipe network key data comprises pipe network design data, historical operation data and / or pipe network reference configuration data, and the pipe network regulation and control parameters comprise a pipe friction coefficient and a water pump efficiency.

[0007] In an embodiment, the target function of the water supply network optimization design comprises the following formula (1): (1); In the formula, f represents an annual total cost minimum value calculation function, represents a pipe network construction cost sub-function, represents an operation cost sub-function, represents a dynamic regulation and control cost sub-function.

[0008] In an embodiment, the pipe network construction cost sub-function comprises the following formula (2): (2); In the formula, n represents the number of branch pipe sections in the water supply network, represents a material cost per unit length of pipe, represents the length of the kth branch pipe, represents an installation cost coefficient related to the pipe diameter, represents the inner diameter of the kth branch pipe.

[0009] In an embodiment, the operation cost sub-function comprises the following formula (3): (3); In the formula, ​represents an electricity price, represents the first k represents a water pump power of the branch pipeline, represents an annual operation time length; wherein the water pump power is calculated by the following formula (4): (4); wherein, represents the first k represents a flow rate of the branch pipeline, represents a head height, represents a water pump efficiency.

[0010] In an embodiment, the head height is calculated by the following formula (5): (5); wherein, represents a terrain height difference, represents an additional head caused by a water hammer effect, represents a frictional resistance loss; The frictional resistance loss is calculated by the following formula: (6); wherein, represents a pipe friction coefficient, represents a square of a flow rate, represents a gravitational acceleration.

[0011] In an embodiment, the dynamic regulation cost sub-function includes the following formula (7): (7); wherein, represents the first k represents a pressure sensor cost of the branch pipeline, represents a control device cost.

[0012] In an embodiment, the arrangement position of the pressure sensor is determined by the following formula (8) and formula (9): (8); constraint condition: (9); ; wherein, represents a binary variable, indicating whether to arrange a pressure sensor at node i, represents a pressure fluctuation amplitude of node i, represents a maximum allowed pressure fluctuation.

[0013] In a second aspect, the application provides a water supply network optimization design system based on digital twinning technology, comprising: a three-dimensional digital model construction module for constructing a three-dimensional digital model of the water supply network, the three-dimensional digital model containing topographic elevation information, pipeline layout information and pump station position information; deploying pressure sensors and flow meters in the physical system; a data acquisition module for acquiring real-time operation data of the water supply network through the pressure sensors and the flow meters; a loading module for loading the three-dimensional digital model into a digital twinning model, and initializing the pipeline parameters of the digital twinning model according to the pre-acquired pipeline key data; a hydraulic simulation module for inputting the real-time operation data into the digital twinning model, and performing hydraulic calculation on each pipe section of the water supply network through the digital twinning model in combination with the real-time operation data, the objective function of water supply network optimization design, and the three-dimensional digital model to simulate the current hydraulic state distribution of each pipe section; a prediction module for predicting the hydraulic characteristics in combination with the current hydraulic state distribution of each pipe section and historical operation data through a preset prediction algorithm to obtain the predicted value of the hydraulic characteristics of each pipe section; a feedback mechanism module for obtaining the hydraulic characteristic error value between the predicted value and the actual value of the hydraulic characteristics of each pipe section, adjusting the pipeline parameters of the water supply network according to the hydraulic characteristic error value, and optimizing the actual operation state of the water supply network through a feedback mechanism.

[0014] The water supply network optimization design method and system based on digital twinning technology provided by the application can realize the efficiency, economy and reliability of water supply network design by constructing a three-dimensional digital model of the water supply network, deploying pressure sensors and flow meters in the physical system, acquiring real-time operation data, inputting the data into a digital twinning model for hydraulic characteristic simulation and prediction, and finally optimizing the actual operation state through a feedback mechanism. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the application, and therefore should not be regarded as limiting the scope of protection of the application. In the various drawings, similar components are denoted by similar reference numerals.

[0016] Figure 1 Fig. 1 shows a flowchart of a water supply network optimization design method based on digital twinning technology provided by the application; Figure 2A schematic diagram of a generation process of a three-dimensional digital model provided by the present application is shown. Figure 3 A structural schematic diagram of a water supply network optimization design system based on digital twinning technology provided by the present application is shown. DETAILED DESCRIPTION

[0017] The technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments.

[0018] The components of the present application generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] Hereinafter, the terms "include", "have", and their conjugates used in various embodiments of the present application are only intended to denote a certain characteristic, number, step, operation, element, component, or combination of the foregoing, and should not be construed as excluding the presence or addition of one or more other characteristics, numbers, steps, operations, elements, components, or combinations of the foregoing.

[0020] In addition, the terms "first", "second", "third", and the like are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

[0021] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of the present application belong. The terms (such as those defined in commonly used dictionaries) will be interpreted as having a meaning that is the same as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized or overly formal meaning unless clearly defined in various embodiments of the present application.

[0022] A regional pipe network topology optimization method based on a Steiner tree type is proposed in the prior art. This method takes the annual total cost as the objective function, solves and compares different pipe network types through graph theory algorithms, and obtains a more optimal pipe network topology structure. However, this technical solution mainly aims at the optimization design of pipe networks in flat areas and does not fully consider the influence of complex terrain features of mountain cities on pipe network layout. At the same time, this method lacks a dynamic monitoring and feedback mechanism for the real-time operation state of the pipe network, and it is difficult to cope with the problem of changes in hydraulic characteristics of the mountain city water supply pipe network caused by terrain elevation difference in actual operation.

[0023] Another existing technology provides a scheme focusing on the optimal arrangement of pressure monitoring points of the water supply pipe network, and uses an improved multi-objective bee mating optimization algorithm to solve the problem of premature convergence and low convergence accuracy of traditional optimization algorithms. However, this technical solution only focuses on the arrangement optimization of pressure monitoring points and fails to consider the special needs of mountain city water supply pipe networks from the perspective of overall pipe network design, such as water hammer effect caused by terrain elevation difference, pump station configuration optimization, and real-time regulation of dynamic operation of the pipe network. In addition, this method does not realize dynamic optimization management of the entire life cycle of the pipe network in combination with digital twin technology, and it is difficult to meet the requirements of modern water supply systems for efficiency and reliability.

[0024] The above problems show that the existing water supply pipe network optimization technology still has significant deficiencies in terms of pipe network topology design, dynamic operation optimization, and real-time monitoring feedback when dealing with complex terrain conditions in mountain cities. Therefore, there is an urgent need for an optimization design method that can fully consider the terrain features, hydraulic characteristics, and dynamic operation requirements of water supply pipe networks in mountain cities. The present application provides a digital twin technology-based optimization design method for water supply pipe networks in mountain cities, which aims to combine the advantages of digital twin technology, realize the efficiency, economy, and reliability of pipe network design through real-time interaction between virtual models and physical systems, and solve the deficiencies in the prior art.

[0025] Embodiment 1 The present application provides a digital twin technology-based optimization design method for water supply pipe networks, which builds a three-dimensional digital model of the water supply pipe network, deploys pressure sensors and flow meters in the physical system, collects real-time operation data, inputs the real-time operation data into the digital twin model for hydraulic characteristic simulation and prediction, and finally optimizes the actual operation state through a feedback mechanism, thereby realizing the efficiency, economy, and reliability of the water supply pipe network design.

[0026] Referring to Figure 1 The digital twin technology-based optimization design method for water supply pipe networks provided by the present application includes steps S101-S106, which are described below.

[0027] Step S101, a three-dimensional digital model of the water supply network is constructed, which contains topographic elevation information, pipeline layout information and pump station position information.

[0028] In this embodiment, geographic information system technology is used to obtain topographic elevation information, and pipeline layout information is planned in combination with the actual demand of the water supply area and the position information of the pump station is determined. Further, the topographic elevation information, pipeline layout information and pump station position information need to be digitized. Referring to Figure 2 The three-dimensional digital model can be constructed according to the topographic elevation information, pipeline layout information and pump station position information through the three-dimensional digital model.

[0029] Taking the water supply system of a mountainous city as an example, it is assumed that the city has complex terrain, significant elevation changes, and uneven spatial and temporal distribution of water supply demand. In order to accurately reflect these characteristics, geographic information system (GIS) technology is used to obtain topographic elevation information, and pipeline layout information is planned in combination with the actual demand of the water supply area. For example, in this case, the water supply network covers an area of about 50 square kilometers, including 10 pump stations and 300 branch pipelines. The length, inner diameter and material of each branch pipeline need to be clearly marked as basic data for subsequent calculation.

[0030] Step S102, deploying pressure sensors and flow meters in the physical system, collecting real-time operation data of the water supply network through the pressure sensors and the flow meters.

[0031] It should be noted that the physical system is the operation system in which the water supply network is located. In addition to including the water supply network, pressure sensors, flow meters and other devices, the physical system also includes other related devices, which are not limited here. After the construction of the three-dimensional digital model, step S102 is entered, that is, deploying pressure sensors and flow meters in the physical system to collect real-time operation data. In this embodiment, the arrangement position of the pressure sensor can be optimized according to the formula , the optimization goal of the arrangement position of the pressure sensor is to minimize the sum of squares of node pressure fluctuation amplitude (formula takes the minimum value), while satisfying the constraint condition On this basis, the specific arrangement position of the sensor is determined by using the particle swarm optimization algorithm. For example, in the above water supply system, it is assumed that the maximum allowable pressure fluctuation is 50kPa, and the algorithm calculates that pressure sensors need to be arranged at 50 key nodes. The selection of these nodes not only considers the amplitude of pressure fluctuation, but also takes into account the cost of pressure sensors and control equipment, to ensure that the dynamic control cost is minimized. In addition, the arrangement of flow meters is concentrated at the pump station outlets and the junctions of main branch pipelines, for monitoring the flow changes of each pipeline.

[0032] For example, a particle swarm optimization algorithm is used to determine 50 key nodes to place pressure sensors. The implementation process includes: first, problem modeling. The decision variable is to use a binary variable (0 or 1) to represent whether to place a pressure sensor at each node in the pipe network. If the pipe network has a node, a binary vector of corresponding dimensions is used. The objective function of this process is to minimize the uncertainty of pressure monitoring or maximize the fault detection capability. For example, the objective function of this process can be where x is the pressure sensor placement scheme, represents the variance of the pressure estimate at node i under scheme x. The constraint conditions include the total number of pressure sensors limit, i.e. and the spatial distribution constraint, which requires the distance between adjacent pressure sensors to be no less than a threshold d min .

[0033] Second, the specific implementation of the particle swarm algorithm. When initializing the particle swarm, each particle represents a sensor placement scheme, and the position vector where m particles are randomly generated, and each particle is ensured to meet the pressure sensor number constraint. When performing fitness evaluation, the pressure distribution of each scheme is calculated using the hydraulic model, and the evaluation indicators include fault detectability, pressure estimation accuracy, etc.

[0034] In the speed and position updating stage, the speed updating formula is where w is the inertia weight, which maintains the original motion trend of the particle, balances global search and local search, c1, c2 are learning factors, r1, r2 are random numbers, is the individual cognitive term, which guides the particle to approach its own historical optimal position, is the social learning term, which guides the particle to approach the global optimal position. Position updating is to map the continuous speed to the binary space, i.e. .

[0035] where the sigmoid function is a commonly used activation function, and the expression of the sigmoid function is as follows: ; The value range of the sigmoid function is between (0, 1).

[0036] where the random function represents a random function for generating (0, 1) random numbers.

[0037] Constraint processing is also performed during the algorithm running process. If a particle violates the pressure sensor quantity constraint, a node is randomly selected to be adjusted to 0 or 1. If a space constraint is violated, that is, the distance between two pressure sensors is too close, one of the pressure sensors is removed. Finally, when the maximum number of iterations (e.g., 100 times) is reached or the fitness value converges, the algorithm terminates. The 50 node combinations that optimize the objective function in the process are output as the arrangement scheme of the pressure sensors. In practical applications, multi-objective optimization can also be performed, considering cost, coverage rate, and fault positioning accuracy, etc., and combined with the digital twin model, real-time data is fed back to the digital twin model to dynamically adjust the arrangement position of the pressure sensors.

[0038] In step S103, the three-dimensional digital model is loaded into the digital twin model, and the pipe network parameters of the digital twin model are initialized and set according to the pre-acquired pipe network key data.

[0039] In an embodiment, the pipe network key data includes pipe network design data, historical operation data, and / or pipe network reference configuration data, which can be determined according to empirical data. The pipe network control parameters include pipe friction coefficient and water pump efficiency.

[0040] In this embodiment, loading and initialization of the digital twin model need to be performed. Specifically, the three-dimensional digital model containing terrain elevation information, pipe layout information, and pump station position information is loaded into the digital twin model, and the pipe friction coefficient and water pump efficiency and other pipe network control parameters in the digital twin model are initialized and set according to the pipe network design data, historical operation data, and / or empirical values, to prepare for simulation and prediction.

[0041] In step S104, the real-time operation data is input into the digital twin model, and the digital twin model performs hydraulic calculation on each pipe section of the water supply network in combination with the real-time operation data, the objective function of the water supply network optimization design, and the three-dimensional digital model to simulate the current hydraulic state distribution of each pipe section.

[0042] It should be noted that the collected real-time operation data may contain noise, outliers, and missing values, so the collected real-time operation data needs to be preprocessed. Noise can be removed by filtering algorithm, outliers can be removed according to a reasonable threshold, and missing values can be filled by interpolation method, so as to obtain standardized data (i.e., preprocessed real-time operation data) that can accurately reflect the actual operation state of the pipe network. The preprocessed real-time operation data can provide a reliable basis for subsequent simulation and prediction.

[0043] In the embodiment, the pre-processed real-time operation data is input into the digital twin model, combined with related formulas of pipe network construction cost, operation cost and dynamic regulation cost, and the principle of computational fluid dynamics is used to fully consider factors such as terrain elevation information, water hammer effect and resistance loss along the way, and the current hydraulic state distribution of each pipe section is simulated by calculating the pressure, flow rate and flow of each pipe section.

[0044] In step S105, the hydraulic characteristic prediction value of each pipe section is obtained by predicting the hydraulic characteristics of each pipe section based on the current hydraulic state distribution of each pipe section and historical operation data through a preset prediction algorithm.

[0045] In the embodiment, the preset prediction algorithm can include time series analysis, machine learning and other prediction algorithms, which are not limited herein. Based on the simulated current hydraulic state distribution of each pipe section, the time series analysis, machine learning and other prediction algorithms are used to predict the future hydraulic characteristics such as pressure fluctuation, flow change and pump station operation state change of each pipe network by combining historical operation data and trends, so as to obtain the hydraulic characteristic prediction value of each pipe section and provide a reference for adjusting the operation parameters of the pipe network.

[0046] In step S106, the hydraulic characteristic error value between the hydraulic characteristic prediction value and the actual value of each pipe section is obtained, and the pipe network parameters of the water supply pipe network are adjusted according to the hydraulic characteristic error value, and the actual operation state of the water supply pipe network is optimized through a feedback mechanism.

[0047] In the embodiment, the hydraulic characteristic error value can be evaluated for accuracy by using error analysis indicators such as root mean square error and mean absolute error. If the error exceeds the acceptable range, the cause is analyzed and the model is adjusted, such as recalibrating parameters, improving structure or supplementing data, until the result meets the requirements.

[0048] It should be further explained that the water supply pipe network optimization design method based on the digital twin technology can be applied to a water supply pipe network optimization design system based on the digital twin technology. The system can include a data preprocessing module that receives real-time operation data, and preprocesses the real-time operation data by denoising, removing outliers and filling missing values to obtain pre-processed real-time operation data. The system can also include a three-dimensional digital model construction module that receives terrain elevation information, pipe layout information, pump station location information and other basic information, and constructs a three-dimensional digital model according to the terrain elevation information, pipe layout information and pump station location information. The system can also include a hydraulic simulation module that receives the three-dimensional digital model and the pre-processed real-time operation data, and outputs the pressure, flow rate, flow distribution of each pipe section of the pipe network, as well as the current hydraulic state distribution such as additional head caused by water hammer effect and resistance loss along the way.

[0049] The system can further comprise a prediction module receiving the current hydraulic state distribution output by the hydraulic simulation module and historical operation data, and outputting predicted future pipe network pressure fluctuation, flow change, pump station operation state change and other hydraulic characteristic prediction values. The predicted data can be used to predict the pipe network operation condition, and then propose optimization and improvement of the pipe network. The system can further comprise a feedback mechanism module receiving the hydraulic characteristic prediction values sent by the prediction module and the actual hydraulic characteristic values actually collected subsequently, performing error analysis according to the hydraulic characteristic prediction values and the actual hydraulic characteristic values, outputting the error analysis result, and outputting model adjustment suggestions according to the error analysis result, so as to optimize the model parameters or the water supply pipe network structure. The water supply pipe network operation data can be continuously monitored, and the above process can be repeated to form a closed-loop feedback system, thereby ensuring the stability and economy of the water supply pipe network operation.

[0050] In the embodiment, a virtual model (i.e., a digital twin model) corresponding to the physical system is constructed through a data twin technology, and real-time data interaction of the physical system and the digital twin model is realized. On one hand, real-time operation data of the physical system is input into the digital twin model for simulation and prediction, thereby providing a basis for optimization; on the other hand, optimization results output by the digital twin model are fed back to the physical system to regulate the actual pipe network. At the same time, the digital twin model is continuously updated and optimized according to new real-time data, thereby forming a dynamic and continuously improved process, which is used for efficient management and optimization of the whole life cycle of the water supply pipe network.

[0051] In an embodiment, the adjusting the pipe network parameters of the water supply pipe network according to the hydraulic characteristic error value comprises: optimizing pump station start-stop time and / or valve opening degree according to the hydraulic characteristic error value.

[0052] It should be noted that after the pump station start-stop time and / or valve opening degree is optimized according to the hydraulic characteristic error value, the actual operation state can also be optimized through a feedback mechanism.

[0053] In the present embodiment, the initial pipe network topology of the water supply network can be generated by using the improved Steiner tree algorithm. Specifically, first, the basic parameters and constraint conditions are determined. The basic parameters include the position, water demand, and other data of each water consumption node in the water supply area, and the pipe laying cost related parameters. The constraint conditions can include terrain restrictions, pipe length and pipe diameter range, and other constraint conditions. Second, a graph model is constructed. Specifically, the water consumption nodes are set as the vertices of the graph, the potential pipe connection paths are set as the edges, and the weights of the edges are determined according to the pipe laying cost, length, and terrain complexity. Then the improved Steiner tree algorithm is used for calculation. On the basis of the classical Steiner tree algorithm, the new nodes (Steiner points) are found and added to the tree structure. According to the edge weights and the existing topology, under the premise of meeting the water supply pressure and flow requirements, the appropriate nodes and edges are selected, and the influence of terrain elevation on hydraulic characteristics is considered to optimize the pipe layout. Finally, the initial pipe network topology that meets the water demand and takes into account the cost and terrain factors is generated, and the direction of each branch pipe segment and the connection nodes are determined.

[0054] When the improved Steiner tree algorithm generates the initial pipe network topology, the reduction of cost is an important target, but it is not simply pursuing the absolute optimal cost. The improved Steiner tree algorithm takes into account the pipe construction cost by setting the edge weights, and preferentially selects the connection path with lower cost. Further, the improved Steiner tree algorithm also considers factors such as terrain restrictions and hydraulic characteristics requirements. In mountainous cities, the influence of the path through complex terrain on construction, maintenance cost and hydraulic stability is considered, and the connection path is selected and optimized by considering the influence of terrain restrictions, hydraulic characteristics requirements, and the path through complex terrain on construction, maintenance cost and hydraulic stability. Therefore, the improved Steiner tree algorithm seeks a balance among cost, terrain, and hydraulic characteristics under multiple constraint conditions, and is not only for the purpose of optimal cost.

[0055] In the present embodiment, the genetic algorithm is used to optimize the pipe network parameters. Specifically, the objective function of the water supply pipe network optimization design includes the following formula (1): (1); In the formula, f represents the annual total cost minimum value calculation function, C represents the pipe network construction cost sub-function, Q represents the operation cost sub-function, and D represents the dynamic control cost sub-function.

[0056] ​​​​In this embodiment, the objective function in the genetic algorithm-based pipe network parameter optimization process is the above formula (1), i.e., minimizing the annual total cost min Z. In this optimization process, the improved Steiner tree algorithm is first used to generate an initial pipe network topology, and then the genetic algorithm is used to iteratively optimize the pipe network parameters. For example, assuming that the initial pipe network topology of the above water supply system contains 300 branch pipelines, after optimization by the genetic algorithm, some redundant pipelines are removed, and finally 250 branch pipelines are retained, thereby significantly reducing the construction cost and operation cost. In the process of pressure sensor arrangement optimization, the objective function is the following formula (8), i.e., minimizing the sum of squares of node pressure fluctuation amplitudes. Through the particle swarm optimization algorithm, the arrangement positions of pressure sensors for 50 key nodes are finally determined, ensuring the stability and economy of the pipe network operation.

[0057] It should be noted that in the above example, the genetic algorithm can finally retain 250 branch pipelines because in the optimization process, the genetic algorithm regards each pipeline in the pipe network as a binary variable, simulates the biological evolution process to find the optimal solution by constructing an objective function containing economy and reliability, and setting constraints such as all water consumption nodes must be connected, the node pressure must meet the minimum service water head, and the final number of retained pipelines must not exceed 250. In this process, the genetic algorithm uses operations such as selection, crossover, and mutation to search in the complex pipe network layout solution space. The selection operation allows pipe network layout schemes with high fitness to have a greater chance of being retained and inherited. The crossover operation allows the characteristics of different schemes to be integrated to produce new schemes that may be better. The mutation operation increases population diversity and avoids the algorithm from falling into local optima. Through continuous iteration, schemes that meet the constraints and achieve a good balance in cost and reliability are gradually selected, thereby achieving the goal of retaining 250 branch pipelines.

[0058] The computational process of the genetic algorithm begins with problem modeling. Each pipeline segment in the network is represented by a binary variable to indicate whether it should be retained. An objective function is constructed that includes the sum of construction and operating costs, as well as network reliability, with weighted coefficients balancing these two factors. Constraints are also set, such as ensuring all water-using nodes are connected, node pressure meets the minimum service head, and the final number of retained pipeline segments does not exceed 250. Next, the genetic algorithm is implemented. A population containing multiple binary-coded individuals (network layout schemes) is initialized, ensuring the initial scheme satisfies connectivity constraints. Then, the fitness value is calculated for each individual, and a roulette wheel or tournament selection method is used to select a parent individual. The selected parent individuals undergo single-point or multi-point crossover to generate offspring, and mutation is performed on them with a low probability. If offspring violate connectivity constraints or the number of pipeline segments exceeds 250, a repair algorithm is used, or the pipeline with the least contribution to reliability is removed, respectively. The selection, crossover, and mutation operations are repeated until the maximum number of iterations or fitness convergence is met, marking the termination condition. Finally, from the Pareto optimal frontier solution set obtained from multi-objective optimization, the scheme that satisfies the requirement of retaining 250 pipeline segments and has the best overall performance was selected. The scheme was then verified and adjusted through hydraulic simulation to ensure that the pressure distribution and flow rate meet the design standards.

[0059] In this embodiment, the pipeline construction cost sub-function includes the following formula (2): (2); In the formula, n represents the number of branch pipe segments in the water supply network. This represents the material cost per unit length of pipe. This represents the length of the k-th branch pipeline. This represents the installation cost coefficient related to the pipe diameter. This represents the inner diameter of the k-th branch pipeline.

[0060] For example, assuming a branch pipeline is 1000 meters long, has an inner diameter of 0.5 meters, a material cost of 200 yuan / meter, and an installation cost coefficient of yuan / meter^m, then the construction cost of this pipeline is 200×1000+×0.5=200000+250=200250 yuan.

[0061] In this embodiment, the operating cost sub-function includes the following formula (3): (3); In the formula, This indicates the unit price of electricity. Indicates the first k The pump power of the branch pipeline section Indicates the annual runtime; In the formula, the power of the water pump is calculated by the following formula (4): (4); In the formula, It is the first k Flow rate of branch pipeline It is the lift height. It refers to the efficiency of the water pump.

[0062] In this embodiment, the head height is calculated using the following formula (5): (5); In the formula, It is the difference in terrain elevation. The additional head is caused by the water hammer effect. It is the friction loss along the friction path; The friction loss along the route is calculated by the following formula: (6); in, It is the coefficient of friction of the pipeline. It's the flow rate. It is gravitational acceleration.

[0063] For example, assuming a branch pipeline has a flow rate of 0.2 cubic meters per second, a head of 50 meters, a pump efficiency of 0.8, an electricity price of 0.6 yuan per kilowatt-hour, and an annual operating time of 4000 hours, then the operating cost of this pipeline section is: Yuan.

[0064] In this embodiment, the dynamic control cost sub-function includes the following formula (7): (7); In the formula, It is the first k Cost of pressure sensors for branch pipeline sections. It's about controlling equipment costs.

[0065] For example, assuming the cost of each pressure sensor is 2000 yuan and the cost of each control device is 0 yuan, if there are 50 pressure sensors and 50 control devices, the total dynamic control cost is 50 × (2000 + 0) = 300 yuan.

[0066] In the embodiment, a particle swarm optimization algorithm is introduced to optimize the arrangement position of the pressure sensor. The arrangement position of the pressure sensor is determined by constructing a special optimization model and solving the model by using the particle swarm optimization algorithm, and is comprehensively determined in combination with an actual cost factor. First, an optimization model is constructed with the sum of squares of the node pressure fluctuation amplitudes being minimized as an objective function, and constraint conditions of the node pressure fluctuation amplitudes not exceeding an allowed maximum value and a binary variable representing arrangement or not are set. Then, the particle swarm optimization algorithm is used to initialize the particle swarm, each particle representing an arrangement scheme, and the particle position and velocity are randomly determined. The fitness value of each particle arrangement scheme is calculated according to the optimization model, and the particle updates the velocity and position according to the historical optimal position of the particle and the global optimal position of the particle swarm by using a specific formula, and it is ensured that the constraint conditions are met. Iteration is continued until the termination condition is met, and the scheme corresponding to the global optimal position at this time is the preliminary determined arrangement position of the pressure sensor. Finally, the dynamic control cost is calculated by comprehensively considering the cost of the pressure sensor and the control equipment, the cost is relatively optimized on the basis of meeting the pressure monitoring demand, and the final arrangement position of the pressure sensor is determined.

[0067] When the particle swarm optimization algorithm is used to optimize the arrangement position of the pressure sensor, the particle swarm is first initialized. A certain number of particles are determined, each particle representing a pressure sensor arrangement scheme, the arrangement position of the particle in the pipe network node is determined by a binary variable, and the velocity of each particle is randomly initialized. Then, the fitness value is calculated. According to the objective function (the sum of squares of the node pressure fluctuation amplitudes is minimized) and the constraint conditions (the node pressure fluctuation amplitude does not exceed the allowed maximum value, and whether to arrange is represented by a binary variable) of the pressure sensor arrangement position optimization model, the fitness value of each particle current arrangement scheme is calculated. The smaller the fitness value is, the better the scheme is. Then, the velocity and position of the particle are updated. The particle updates the velocity and position according to the historical optimal position of the particle and the global optimal position of the particle swarm by using a specific formula, and it is ensured that the position meets the constraint conditions when updating. If it does not meet the constraint conditions, it is corrected. Finally, the termination condition is checked. If the maximum number of iterations is reached, or the fitness value of the global optimal solution changes very little within a certain number of iterations, etc., the algorithm stops, and the sensor arrangement scheme corresponding to the global optimal position at this time is the optimization result (i.e., the preliminary determined arrangement position of the pressure sensor). If it does not meet the condition, the iteration optimization is continued. At the same time, the cost of the sensor and the control equipment, etc. are comprehensively considered to minimize the dynamic control cost, and the final arrangement position is determined.

[0068] In the embodiment, the arrangement position of the pressure sensor is determined by using the following formula (8) and formula (9): (8); Constraint condition: (9); ; wherein, is a binary variable, indicating whether a pressure sensor is arranged at node i, is the pressure fluctuation amplitude of node i, is the maximum pressure fluctuation allowed.

[0069] In this embodiment, by introducing the digital twin technology, combining the improved Steiner tree algorithm, genetic algorithm and particle swarm optimization algorithm, the efficiency, economy and reliability of the mountain city water supply network design are realized. The specific implementation fully embodies the technical advantages of the present application, and has a wide application prospect.

[0070] In order to verify the effectiveness of the method of the present application, the water supply system of a certain mountain city is selected as an example. The terrain of the water supply system is complex, and the water supply demand presents a significant spatial and temporal distribution uneven characteristic. By applying the method of the present application, a three-dimensional digital model of the water supply network is first constructed, and the terrain elevation information, pipeline layout information and pump station location information are determined. Subsequently, 50 pressure sensors and 20 flow meters are deployed in the physical system to collect real-time operation data. The collected data is input into the digital twin model for hydraulic characteristic simulation and prediction, and it is found that the resistance loss of some pipelines is large, resulting in high operating cost. By adjusting the pipeline parameters, such as optimizing the pump station start-stop time and valve opening, the stability and economy of the pipeline operation are finally realized. Specifically, the pipeline construction cost is reduced from 6 million yuan to 1 million yuan, the operating cost is reduced from 1.2 million yuan per year to 0.9 million yuan, the dynamic control cost is reduced from 0.5 million yuan per year to 0.35 million yuan, and the total annual cost is reduced from 7.7 million yuan to 6.25 million yuan, with a reduction of 18.8%.

[0071] In summary, the method of the present application realizes the efficiency, economy and reliability of the mountain city water supply network design by introducing the digital twin technology, combining the improved Steiner tree algorithm, genetic algorithm and particle swarm optimization algorithm. The specific embodiment fully embodies the technical advantages of the present application, and has a wide application prospect.

[0072] Example 2 In addition, the present application provides a water supply network optimization design system based on digital twin technology.

[0073] As Figure 3 shown, a water supply network optimization design system based on digital twin technology includes: A three-dimensional digital model construction module 301 is used to construct a three-dimensional digital model of the water supply network, which contains terrain elevation information, pipeline layout information and pump station location information; pressure sensors and flow meters are deployed in the physical system; The data acquisition module 302 is configured to acquire real-time operation data of the water supply network through the pressure sensor and the flow meter. The loading module 303 is configured to load the three-dimensional digital model into a digital twin model, and initialize and set parameters of the digital twin model according to pre-acquired key data of the water supply network. The hydraulic simulation module 304 is configured to input the real-time operation data into the digital twin model, and perform hydraulic calculation on each pipe section of the water supply network through the digital twin model in combination with the real-time operation data, a target function of the water supply network optimization design, and the three-dimensional digital model, to simulate a current hydraulic state distribution of each pipe section. The prediction module 305 is configured to perform hydraulic characteristic prediction through a preset prediction algorithm in combination with the current hydraulic state distribution of each pipe section and historical operation data, to obtain a hydraulic characteristic prediction value of each pipe section. The feedback mechanism module 306 is configured to acquire a hydraulic characteristic error value between the hydraulic characteristic prediction value and an actual hydraulic characteristic value of each pipe section, adjust parameters of the water supply network according to the hydraulic characteristic error value, and optimize an actual operation state of the water supply network through a feedback mechanism. The water supply network optimization design system based on the digital twin technology provided in this embodiment can implement the water supply network optimization design method based on the digital twin technology provided in Embodiment 1, and details are not repeated here to avoid repetition.

[0074] The water supply network optimization design system based on the digital twin technology provided in this embodiment introduces the digital twin technology, and combines the improved Steiner tree algorithm, the genetic algorithm and the particle swarm optimization algorithm, to realize the efficiency, economy and reliability of the water supply network design of the mountainous city.

[0075] It should be noted that, in this document, the terms “comprising” and “including” or any other variant thereof are intended to cover non-exclusive inclusions, so that processes, methods, articles or terminals including a series of elements not only include those elements, but also include other elements not explicitly listed, or include elements inherent to such processes, methods, articles or terminals. Without more limitations, the element defined by the statement “comprising a” does not exclude the presence of another identical element in the process, method, article or terminal including the element.

[0076] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and necessary general hardware platform, of course, also can be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a plurality of instructions to make a terminal execute the method described in each embodiment of the present application.

[0077] The embodiments of the present application are described above in combination with the drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, not restrictive. Those skilled in the art can make many forms without departing from the purpose of the present application and the scope protected by the claims under the inspiration of the present application, which all belong to the protection of the present application.

Claims

1. A water supply network optimization design method based on digital twin technology, characterized in that, The method comprises: constructing a three-dimensional digital model of a water supply network, the three-dimensional digital model containing topographic elevation information, pipeline layout information and pump station position information; deploying pressure sensors and flow meters in a physical system, and collecting real-time operation data of the water supply network through the pressure sensors and the flow meters; loading the three-dimensional digital model into a digital twin model, and initializing the pipeline parameters of the digital twin model according to pre-acquired pipeline key data; inputting the real-time operation data into the digital twin model, and performing hydraulic calculation on each pipe section of the water supply network through the digital twin model in combination with the real-time operation data, an objective function of the optimization design of the water supply network and the three-dimensional digital model, to simulate the current hydraulic state distribution of each pipe section; performing hydraulic characteristic prediction through a preset prediction algorithm in combination with the current hydraulic state distribution of each pipe section and historical operation data, to obtain a hydraulic characteristic prediction value of each pipe section; acquiring a hydraulic characteristic error value between the hydraulic characteristic prediction value and an actual hydraulic characteristic value of each pipe section, adjusting the pipeline parameters of the water supply network according to the hydraulic characteristic error value, and optimizing the actual operation state of the water supply network through a feedback mechanism.

2. The method of claim 1, wherein, The adjusting of the pipeline parameters of the water supply network according to the hydraulic characteristic error value comprises: optimizing pump station start-stop time and / or valve opening degree according to the hydraulic characteristic error value.

3. The method of claim 1, wherein, The pipeline key data comprises pipeline design data, historical operation data and / or pipeline reference configuration data, and the pipeline regulation and control parameters comprise a pipeline friction coefficient and a water pump efficiency.

4. The method of claim 1, wherein, The objective function of the optimization design of the water supply network comprises the following formula (1): (1) In the formula, represents the annual total cost minimum value calculation function, represents the pipe network construction cost sub-function, represents the operation cost sub-function, represents the dynamic regulation cost sub-function.

5. The method of claim 4, wherein, The pipeline construction cost sub-function comprises the following formula (2): (2) wherein n represents the number of branch pipe sections in the water supply network, represents the material cost per unit length of pipe, represents the length of the kth branch pipe section, represents the installation cost coefficient related to the pipe diameter, represents the internal diameter of the kth branch pipe section.

6. The method of claim 5, wherein, The operation cost sub-function comprises the following formula (3): (3) In the formula, represents the electricity price, represents the first k represents the water pump power of the branch pipeline of the section, represents the annual operation time; The water pump power is calculated by the following formula (4): （4） wherein is the first k segment branch line flow, is the head height, is the pump efficiency.

7. The method of claim 6, wherein, The lift height is calculated by the following formula (5): (5) wherein is the terrain elevation difference, is the additional head due to water hammer effect, is the frictional resistance loss; The along-path resistance loss is calculated by the following formula: （6） wherein, is the pipe friction factor, is the square of the flow velocity, is the gravitational acceleration.

8. The method of claim 7, wherein, The dynamic regulation and control cost sub-function comprises the following formula (7): （7） wherein is the cost of the pressure sensor of the branch line of the section k is the cost of the control device is the cost of the control device 9. The method of claim 1, wherein, The arrangement positions of the pressure sensors are determined by the following formula (8) and formula (9): （8） Constraint condition: （9） wherein is a binary variable indicating whether a pressure sensor is arranged at node i, is the pressure fluctuation amplitude at node i, is the maximum pressure fluctuation allowed.

10. A water supply network optimization design system based on digital twin technology, characterized in that, The system comprises: a three-dimensional digital model construction module, configured to construct a three-dimensional digital model of a water supply network, the three-dimensional digital model containing topographic elevation information, pipeline layout information and pump station position information, and deploy pressure sensors and flow meters in a physical system; a data collection module, configured to collect real-time operation data of the water supply network through the pressure sensors and the flow meters; a loading module, configured to load the three-dimensional digital model into a digital twin model, and initialize the pipeline parameters of the digital twin model according to pre-acquired pipeline key data; a hydraulic simulation module, configured to input the real-time operation data into the digital twin model, and perform hydraulic calculation on each pipe section of the water supply network through the digital twin model in combination with the real-time operation data, an objective function of the optimization design of the water supply network and the three-dimensional digital model, to simulate the current hydraulic state distribution of each pipe section; and a feedback optimization module, configured to acquire a hydraulic characteristic error value between a hydraulic characteristic prediction value and an actual hydraulic characteristic value of each pipe section, adjust the pipeline parameters of the water supply network according to the hydraulic characteristic error value, and optimize the actual operation state of the water supply network. a prediction module configured to predict hydraulic characteristics of each pipe section by combining preset prediction algorithms with current hydraulic state distribution and historical operation data of each pipe section to obtain predicted values of the hydraulic characteristics of each pipe section; a feedback mechanism module configured to obtain error values of the hydraulic characteristics between the predicted values of the hydraulic characteristics of each pipe section and actual values of the hydraulic characteristics of each pipe section, adjust pipe network parameters of the water supply pipe network according to the error values of the hydraulic characteristics, and optimize actual operation states of the water supply pipe network through a feedback mechanism.

Images