Customized hemodialysis water supply pipe network system based on digital twinning and 3D printing

Abstract

The present application relates to the technical field of medical water treatment, in particular to a customized hemodialysis water supply pipe network system based on digital twinning and 3D printing, which comprises a sensor data acquisition module, a digital twinning hydraulic simulation module, a structure collaborative optimization module, a pipe manufacturing instruction generation module and a 3D printing pipe forming module. In the present application, by constructing a hydraulic and water flow mapping relationship synchronized with the actual running state, the pipe geometry parameters and flow distribution are dynamically deduced based on the bed terminal water characteristics, so that the branches maintain a coordinated water supply state under different load conditions, and the optimized structure parameters are directly converted into executable manufacturing paths, realizing the directional forming of pipe size and internal structure, avoiding the accumulation of artificial processing errors, supporting differentiated branch configuration, reducing the influence of later adjustment on the overall pipe network, and improving water supply consistency, structure adaptability and operation controllability.

Description

Technical Field

[0001] This invention relates to the field of medical water treatment technology, and in particular to a customized hemodialysis water supply network system based on digital twins and 3D printing. Background Technology

[0002] The field of medical water treatment technology refers to a specialized technical system centered around the preparation, transportation, and quality assurance of water used in medical institutions for diagnostic and treatment support. Core aspects include raw water pretreatment, reverse osmosis purification, circulation disinfection, pipeline transportation, and end-use water control. In particular, hemodialysis water needs to meet the requirements of low conductivity, low endotoxin, and low microbial load. This field involves the structural design, material selection, connection methods, cleaning and disinfection paths of the water supply network, and interface matching with hemodialysis equipment. Its overall goal is to build a stable, controllable, and medically compliant dedicated water treatment and distribution system.

[0003] Among them, the traditional customized hemodialysis water supply network system refers to a system solution that is specially designed and arranged according to the spatial layout and number of beds of the dialysis center. The technical issues it addresses are the continuous delivery of purified water for hemodialysis under multi-terminal conditions and pipeline adaptation. It involves manually measuring the on-site dimensions and drawing two-dimensional or three-dimensional pipeline drawings based on experience, selecting standard specification pipes such as stainless steel or medical plastic pipes, cutting and welding or hot-melting them to form a fixed-direction ring or dendritic network. At the same time, the pipe diameter, length, elbow position and number of interfaces are determined in advance during the design stage, and the processing and installation are completed accordingly to form a complete hemodialysis water supply network system.

[0004] Traditional customized hemodialysis water supply networks are based on static drawings and fixed pipe diameter configurations. The pipeline structure and branch parameters are locked in the design stage. During operation, they lack the ability to continuously sense changes in water pressure, water velocity, and water circuit operation status. Due to differences in equipment models and usage intensity, hydraulic distribution imbalances are prone to occur in different beds. Insufficient pressure or flow fluctuations in local branches are difficult to identify. Abnormal operating conditions are mainly detected by manual inspection, resulting in a delayed response. Pipe fittings are manufactured in a standardized manner and assembled on-site. Structural adjustments require overall modification, increasing maintenance costs and downtime risks, which restricts the long-term stable operation of the system. Summary of the Invention

[0005] To address the challenges of traditional customized hemodialysis water supply networks, which rely on static drawings and fixed pipe diameters, locking in pipeline structure and branch parameters during the design phase, lacking continuous monitoring of changes in water pressure, velocity, and operating status during operation, and prone to hydraulic imbalances due to equipment model differences and varying usage intensity at different beds, making it difficult to identify insufficient pressure or flow fluctuations in local branches, relying primarily on manual inspections for abnormal operating conditions with delayed response, and requiring standardized processing and on-site assembly of pipe fittings, structural adjustments necessitate overall modifications, increasing maintenance costs and downtime risks, and hindering long-term stable system operation, this invention provides a customized hemodialysis water supply network system based on digital twins and 3D printing.

[0006] On the one hand, a customized hemodialysis water supply network system based on digital twins and 3D printing was provided, which includes: The sensor data acquisition module collects electrical signals of water flow velocity, water pressure and water circuit status at key nodes in the hemodialysis water supply system, performs analog-to-digital conversion to form digital signals, obtains the equipment model and usage frequency of the bed, calculates the target water pressure range and flow requirements, and constructs a node status dataset. The digital twin hydraulic simulation module, based on the node status dataset, uses computational fluid dynamics algorithms to analyze the water flow velocity distribution and water pressure conditions, identify the hydraulic supply deviation value of the bed terminal, calculate the dynamic hydraulic and water flow distribution characteristics and the bed hydraulic matching degree, and output the digital twin hydraulic simulation results. The structural collaborative optimization module, based on the digital twin hydraulic simulation results, uses a genetic algorithm to iteratively adjust the pipe diameter, connection angle, and layout parameters, implements differentiated diameter design and flow allocation for the bed terminal branch pipe sections, calculates the bed pressure compliance rate, and constructs a set of pipe network optimization parameters; The pipeline manufacturing instruction generation module analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, calculates the 3D printing path, determines the printing jet rate, material flow rate and temperature control, and generates a 3D printing manufacturing instruction sequence.

[0007] As a further aspect of the present invention, the node status dataset includes bed water consumption level, node water supply load level, water circuit operation status identifier, and stable operation status; the digital twin hydraulic simulation results include node water pressure distribution, flow distribution, water flow state distribution, and bed hydraulic matching results; the pipeline network optimization parameter set includes pipe diameter specification configuration, branch flow distribution ratio, system pressure balance index, and bed water supply compliance index; and the 3D printing manufacturing instruction sequence includes forming path data, printing speed parameters, layer thickness parameters, and printing temperature parameters.

[0008] As a further aspect of the present invention, the sensor data acquisition module includes: The signal acquisition submodule acquires water flow velocity signals, water pressure signals, and water circuit status electrical signals of key nodes in the hemodialysis water supply system. It performs synchronous alignment based on channel amplitude changes and sampling period, quantizes and sorts discrete sampled values ​​and binds them with time stamps, and generates a set of node digital signal sequences. The demand calculation submodule, based on the set of node digital signal sequences, obtains the bed equipment model code and usage frequency record, reads the model-related water pressure reference range and flow reference range, performs range mapping and comparison calculation based on the node water flow velocity value and water pressure value, and generates a set of bed water supply demand parameters. The state construction submodule calls the set of bed water supply demand parameters and the set of node digital signal sequences, performs consistency comparison on the node water flow velocity value, water pressure value, and water circuit state electrical signal, and performs node state matching, item integration and structured arrangement based on the numerical correspondence to establish a node state dataset.

[0009] As a further aspect of the present invention, the water pressure reference range and the flow rate reference range are generated from the original usage data, equipment parameters and actual operating conditions related to the bed equipment model. Specifically, by collecting the operating data of the bed equipment, and based on the typical water pressure and flow rate range of different equipment models, combined with the equipment usage frequency and common operating environment, the minimum and maximum values ​​of water pressure and flow rate of the equipment under preset rated operating conditions are determined as the upper and lower limits of the water pressure reference range and the flow rate reference range.

[0010] As a further aspect of the present invention, the digital twin hydraulic simulation module includes: The node status construction submodule collects bed terminal flow records, water pressure sampling values ​​and timestamp sequences based on the node status dataset. It performs time alignment on the flow sequence and water pressure sequence according to the node identifier, performs interpolation compensation and missing label processing, rearranges the node data items, and generates a node hydraulic status vector set. The flow pressure deviation identification submodule calls the node hydraulic state vector set, calculates the simulated flow velocity and simulated water pressure of each node according to the node topology association and pipeline geometry, performs alignment comparison with the measured values ​​and extracts the flow velocity difference and water pressure difference sequence, performs scale normalization and direction calibration, calculates the supply offset according to the bed position identifier, and obtains the bed hydraulic supply deviation value. The twin matching output submodule performs joint mapping on the velocity and pressure distribution matrix based on the bed hydraulic supply deviation value and the node hydraulic state vector set, synchronously integrates the waterway operation status signal, calculates the matching degree based on the bed number, and establishes digital twin hydraulic simulation results.

[0011] As a further aspect of the present invention, the structural collaborative optimization module includes: The pipe parameter correction submodule obtains pipe diameter, connection angle and layout parameter data based on the digital twin hydraulic simulation results. It performs residual alignment and trend comparison between the continuous flow observation sequence and the simulation output. It uses a genetic algorithm to iteratively optimize the pipe parameter state variables based on the fitness function to generate pipe segment state correction coefficients. The branch configuration submodule calls the topology information of the branch pipe segment of the bed terminal according to the pipe segment status correction coefficient, performs matching judgment on the node flow demand input and the conveying capacity of adjacent pipe segments, and performs combination screening based on the pipe diameter candidate set and the flow difference range to generate the branch diameter configuration value. The pressure assessment submodule calls the branch diameter configuration value and connects the pressure acquisition value of the bed node. It performs interval determination on the pressure acquisition value and the set bed pressure benchmark interval, counts the ratio of the number of qualified nodes to the total number of nodes, calculates the bed pressure compliance rate, and constructs a pipeline optimization parameter set.

[0012] As a further aspect of the present invention, the pipeline manufacturing instruction generation module includes: The geometric structure analysis submodule analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, obtains the pipeline segment axis coordinate sequence and performs vector difference operation, obtains the cross-sectional contour point set and performs closure judgment, collects connection topology identifiers and verifies the relationship between adjacent pipeline segments, and obtains the pipeline three-dimensional geometric description matrix. The path planning calculation submodule, based on the three-dimensional geometric description matrix of the pipeline, obtains the height sequence between printing layers and performs layer-by-layer mapping. For the multi-layer cross-sectional contour point set, it calculates the path point candidate sequence, compares the path point spacing between adjacent layers and adjusts the direction order to generate a set of printing path trajectories. The manufacturing instruction arrangement submodule obtains the trajectory segment length change rate and performs interval division based on the printing path trajectory set, calculates the printing injection rate, material flow rate and temperature control interval based on the interval results, and performs timing combination according to the trajectory segment arrangement order to generate a 3D printing manufacturing instruction sequence.

[0013] As a further aspect of the present invention, the system further includes: The 3D printing pipe forming module, based on the 3D printing manufacturing instruction sequence, executes nozzle dynamic trajectory control, adjusts the printing nozzle speed and material extrusion amount, and implements real-time monitoring and feedback adjustment of printing layer thickness and forming temperature to form hemodialysis water supply pipes; The hemodialysis water supply pipe fittings include the pipe fitting geometry, internal flow channel structure, and interface connection type.

[0014] As a further aspect of the present invention, the 3D printed tube forming module includes: The trajectory control submodule, based on the 3D printing manufacturing instruction sequence, obtains the nozzle spatial displacement instruction, collects the displacement change corresponding to the time axis, compares the displacement change point by point according to the equipment stroke parameters, performs coordinate recursion calculation according to the continuous interpolation rule, and generates the nozzle trajectory coordinate sequence. The extrusion adjustment submodule collects the nozzle running speed sampling and material supply rate sampling based on the nozzle trajectory coordinate sequence, aligns the sampling in time, calculates the material distribution per unit path according to the relationship between path length and speed, judges the deviation of the distribution in intervals, and generates an extrusion flow rate calibration value. The forming integration submodule calls the extrusion flow calibration value, monitors the printing layer thickness sampling amount and the forming area temperature sampling amount, calculates the difference between the layer thickness sampling amount and the layer thickness benchmark, and makes a range judgment on the temperature sampling amount. It then performs solid deposition according to the jet control sequence to generate the hemodialysis water supply pipe.

[0015] As a further aspect of the present invention, the layer thickness reference is a target layer thickness value determined based on the printer's set printing parameters, material type, and printing process requirements, serving as a reference standard for the actual thickness during the printing process.

[0016] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: By constructing a hydraulic and flow mapping relationship synchronized with the actual operating status, the geometric parameters and flow distribution of the pipeline are dynamically extrapolated based on the water usage characteristics of the bed terminal. This enables the branches to maintain a coordinated water supply state under different load conditions. Furthermore, the optimized structural parameters are directly converted into executable manufacturing paths, achieving directional forming of pipe fitting dimensions and internal structures. This avoids the accumulation of errors from manual processing and supports differentiated branch configurations, reducing the impact of later adjustments on the overall pipeline network and improving water supply consistency, structural adaptability, and operational controllability. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the accompanying drawings without creative effort.

[0018] Figure 1 This is a system schematic diagram of the present invention; Figure 2 This is a schematic diagram of the system framework of the present invention; Figure 3 This is a flowchart of the sensor data acquisition module in this invention; Figure 4 This is a flowchart of the digital twin hydraulic simulation module in this invention; Figure 5 This is a flowchart of the structural collaborative optimization module in this invention; Figure 6 This is a flowchart of the pipeline manufacturing instruction generation module in this invention; Figure 7 This is a flowchart of the 3D printing pipe forming module in this invention. Detailed Implementation

[0019] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0020] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0021] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0022] In this embodiment of the invention, sometimes the subscript such as W1 is written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0023] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0024] This invention provides a customized hemodialysis water supply network system based on digital twins and 3D printing, such as... Figure 1-2 The diagram shown illustrates a customized hemodialysis water supply network system based on digital twins and 3D printing. The system includes: The sensor data acquisition module collects electrical signals of water flow velocity, water pressure and water circuit status at key nodes in the hemodialysis water supply system, performs analog-to-digital conversion to form digital signals, obtains the equipment model and usage frequency of the bed, calculates the target water pressure range and flow requirements, and constructs a node status dataset. The digital twin hydraulic simulation module, based on the node status dataset, uses computational fluid dynamics algorithms to analyze the water flow velocity distribution and water pressure conditions, identify the hydraulic supply deviation value of the bed terminal, calculate the dynamic hydraulic and water flow distribution characteristics and the bed hydraulic matching degree, and output the digital twin hydraulic simulation results. The structural collaborative optimization module, based on the results of digital twin hydraulic simulation, uses a genetic algorithm to iteratively adjust the pipe diameter, connection angle and layout parameters, implements differentiated diameter design and flow distribution for the branch pipe sections at the bed terminals, calculates the bed pressure compliance rate, and constructs a set of pipe network optimization parameters. The pipeline manufacturing instruction generation module analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, calculates the 3D printing path, determines the printing jet rate, material flow rate and temperature control, and generates a 3D printing manufacturing instruction sequence. The 3D printing pipe forming module, based on the 3D printing manufacturing instruction sequence, executes nozzle dynamic trajectory control, adjusts the printing nozzle speed and material extrusion amount, and implements real-time monitoring and feedback adjustment of printing layer thickness and forming temperature to form hemodialysis water supply pipes; The node status dataset includes bed water consumption level, node water supply load level, water circuit operation status identifier, and stable operation status. The digital twin hydraulic simulation results include node water pressure distribution, flow distribution, water flow state distribution, and bed hydraulic matching results. The pipeline network optimization parameter set includes pipe diameter specification configuration, branch flow distribution ratio, system pressure balance index, and bed water supply compliance index. The 3D printing manufacturing instruction sequence includes forming path data, printing speed parameters, layer thickness parameters, and printing temperature parameters. The hemodialysis water supply pipe fittings include pipe fitting geometry, internal flow channel structure, and interface connection form.

[0025] Specifically, such as Figure 2 , 3 As shown, the sensor data acquisition module includes: The signal acquisition submodule acquires water flow velocity signals, water pressure signals, and water circuit status electrical signals of key nodes in the hemodialysis water supply system. It performs synchronous alignment based on channel amplitude changes and sampling period, quantizes and sorts discrete sampled values ​​and binds them with time stamps, and generates a set of node digital signal sequences. The signal acquisition submodule is equipped with a high-frequency data acquisition unit and a multi-channel signal synchronization unit. This module deploys electromagnetic flowmeters, piezoresistive pressure sensors, and water circuit status monitoring probes at key pipeline nodes in the hemodialysis water supply system to capture real-time water flow velocity signals, water pressure signals, and electrical signals indicating the water circuit's operating status. The data acquisition unit is set to a sampling frequency of 1000 Hz to ensure accurate capture of transient fluid changes. Based on the signal amplitude variation characteristics of the physical channels, the synchronization unit selects the first significant peak as the synchronization trigger point and uses hardware phase-locked loop technology to synchronize the sampling periods of the flow, pressure, and water circuit status signals, eliminating time phase differences caused by sensor response delays. Subsequently, the quantization processing unit within the module performs analog-to-digital conversion and sorting of the discrete analog sampled values, binding a globally unified time (UTC) time stamp accurate to the second to each quantized sampling point, generating a set of node digital signal sequences containing timestamps, flow values, pressure values, and water circuit status values. As shown in Table 1, this module collects and integrates the raw physical quantity data of different monitoring nodes within a specific time slice, providing a foundation for subsequent calculations.

[0026] Table 1: Multidimensional Signal Acquisition Sequence Table for Water Supply Network Nodes

[0027] As shown in Table 1, the flow rate, water pressure and waterway status data of the nodes were recorded at the millisecond-level timestamp, and the extremely low synchronization delay was monitored, which verified the high precision and high synchronization of the signal acquisition.

[0028] The demand calculation submodule, based on the node digital signal sequence set, obtains the bed equipment model code and usage frequency record, reads the model-related water pressure reference range and flow reference range, performs range mapping and comparison calculation based on the node water flow velocity value and water pressure value, and generates a set of bed water supply demand parameters. The demand calculation submodule integrates an equipment feature database and an interval mapping calculation unit. First, based on the previously generated set of node digital signal sequences, this module parses the implicit bed identification information and retrieves the model code of the hemodialysis machine configured for the corresponding bed (e.g., Fresenius4008S or NikkisoDBB-27) and the original usage frequency record of the equipment from the equipment feature database. The mapping calculation unit reads the rated water pressure reference range (e.g., 250 kPa to 350 kPa) and flow rate reference range (e.g., 500 ml to 800 ml per minute) required for the normal operation of this type of equipment based on the model code. Next, the module performs a comparison calculation, converting the real-time collected water flow velocity values ​​from the nodes into volumetric flow rate values ​​and performing a difference calculation with the flow rate reference range. Simultaneously, it performs an inclusion check on the collected water pressure values ​​and the water pressure reference range. If the real-time value is lower than the lower limit of the range, the module calculates the deficit; if it is higher than the upper limit, it calculates the overpressure. Through this interval mapping and difference quantification, the module generates a set of bed water supply demand parameters containing specific water supply surplus and deficit data for each bed at the current moment. For example, for bed numbered Bed_05, if the real-time water pressure is 240 kPa and the lower limit of the reference is 250 kPa, then a pressure demand parameter of -10 kPa is generated, which clarifies the specific adjustment requirements.

[0029] The state construction submodule calls the bed water supply demand parameter set and the node digital signal sequence set, performs consistency comparison on the node water flow velocity value, water pressure value, and water circuit state electrical signal, and performs node state matching, item integration and structured arrangement based on the numerical correspondence to establish a node state dataset. The state construction submodule comprises a multi-dimensional data fusion unit and a state structured orchestration unit. This module calls upon the bed-level water supply demand parameter set and the node digital signal sequence set to initiate a consistency comparison program. The fusion unit first checks whether the flow and pressure change trends of the same node at the same time conform to the general laws of fluid mechanics (such as the negative correlation between velocity and pressure in Bernoulli's principle), and performs comprehensive verification by combining this with the stability of the water circuit state electrical signals, eliminating outliers caused by sensor malfunctions. Subsequently, the module performs state matching based on numerical correspondences, mapping the verified physical quantity data to system operating state descriptions, such as specific state labels like "normal water supply," "low pressure and low flow," and "local water circuit anomaly." The orchestration unit further integrates the scattered labels, establishing an index according to spatial topological order and temporal sequence, reorganizing single-point-time data into logically related structured entries, ultimately establishing a node state dataset that reflects the real-time operation of the entire pipeline network. This process ensures that the data upon which subsequent analysis is based is not only a collection of numerical values, but also a state description with physical meaning and logical connections.

[0030] Specifically, such as Figure 2 ,4 As shown, the digital twin hydraulic simulation module includes: The node status construction submodule collects bed terminal flow records, water pressure sampling values ​​and timestamp sequences based on the node status dataset. It performs time alignment on the flow sequence and water pressure sequence according to the node identifier, performs interpolation compensation and missing label processing, rearranges the node data items, and generates a node hydraulic status vector set. The node status construction submodule operates with a time-series alignment algorithm and interpolation compensation logic. Based on the node status dataset, this module extracts flow records, water pressure sampling values, and corresponding timestamp sequences from the bed terminals. Due to minor fluctuations in network transmission or acquisition frequency, data from different sources may exhibit time-series misalignment. This module performs strict time-series alignment of the flow and water pressure sequences based on the node's unique identifier (ID) and a high-precision master node timestamp. For time-series breaks or missing data segments discovered during alignment, the module employs a cubic spline interpolation algorithm. It constructs a smooth curve using three valid data points before and after the missing point, calculates the value at the missing time, and marks it with an "interpolated" label for differentiation. After completion, the module rearranges the node data items in chronological order, combining scalar data into a multi-dimensional vector to generate a node hydraulic state vector set containing flow velocity, pressure, time, and their derivative characteristics. For example, by using an interpolation algorithm, the number of sampling points for a certain node within 10 seconds was successfully increased from 98 to the standard 100, ensuring the continuity of the time series and providing a complete data foundation for subsequent trend analysis.

[0031] The flow pressure deviation identification submodule calls the node hydraulic state vector set, and calculates the simulated flow velocity and simulated water pressure of each node using computational fluid dynamics algorithms based on the node topology association and pipeline geometry. It performs alignment comparison with the measured values ​​and extracts the flow velocity difference and water pressure difference sequence. It performs scale normalization and direction calibration, calculates the supply offset based on the bed position identifier, and obtains the bed hydraulic supply deviation value. The flow-pressure deviation identification submodule incorporates a computational fluid dynamics (CFD) simulation engine and a deviation analysis unit. This module calls upon the node hydraulic state vector set and, based on the physical topology and geometric parameters of the pipeline network, uses a CFD algorithm to calculate the simulated flow velocity and simulated water pressure values ​​at the nodes. The deviation analysis unit performs node alignment comparison between the simulated and measured values, extracting the flow velocity deviation sequence and the water pressure deviation sequence. Subsequently, Z-score standardization is performed on the deviation sequences, normalizing the data scale to a distribution range with a mean of 0 and a standard deviation of 1, and directionally calibrating the deviation sign according to the flow direction. Based on the bed location identifier, the module extracts the standardized deviation value of the corresponding bed location terminal node, calculates the weighted Euclidean distance between the flow velocity deviation and the water pressure deviation, i.e., the supply offset, thereby obtaining the accurate bed location hydraulic supply deviation value. For example, the measured flow velocity at a certain bed terminal is 0.45 m / s and the measured water pressure is 0.28 MPa. CFD simulation calculations show that the theoretical flow velocity at this node is 0.50 m / s and the theoretical water pressure is 0.30 MPa. After standardization, the characteristic value of the flow velocity deviation is -0.8 and the characteristic value of the water pressure deviation is -0.6. The comprehensive calculation shows that the supply offset of this bed is a decrease of 0.05 m / s and a decrease of 0.02 MPa in pressure.

[0032] The twin matching output submodule performs joint mapping on the velocity and water pressure distribution matrix based on the bed hydraulic supply deviation value and the node hydraulic state vector set, synchronously integrates the water circuit operation status signal, calculates the matching degree based on the bed number, and establishes digital twin hydraulic simulation results. The twin matching output submodule integrates a multiphysics mapping engine and a similarity evaluation unit. This module first uses the flow velocity, water pressure, and waterway operation status data from the node state dataset to jointly map the distribution matrices of water flow velocity and water pressure onto the grid nodes of the virtual pipe network model via the mapping engine. It then synchronously superimposes the waterway operation status signals as discrete state fields onto the virtual pipe network model, thereby completing the collaborative integration of multidimensional hydraulic operation information and constructing a digital twin hydraulic simulation model. Subsequently, the simulation model is driven to perform fluid dynamics calculations, outputting predicted hydraulic states of the virtual pipe network. The evaluation unit compares the virtual state output by the simulation with the actual state collected by field sensors based on the bed number, calculating the cosine similarity between the two as a matching index. When the matching score exceeds 0.95, it confirms that the digital twin model has accurately reproduced the physical entity state; if it is lower than this value, it triggers model parameter fine-tuning and a re-simulation. After model validation, by comparing simulation results with actual monitoring data, the hydraulic supply deviation values ​​of each bed terminal were identified, and the dynamic hydraulic and flow distribution characteristics and bed hydraulic matching degree were calculated to form digital twin hydraulic simulation results. For example, the calculated flow field distribution matching degree between the simulation model and the actual monitoring data was 0.98, verifying the accuracy of the twin model. The deviation distribution characteristics output by the model provide a reliable data benchmark for subsequent structural optimization.

[0033] Specifically, such as Figure 2 , 5 As shown, the structural collaborative optimization module includes: The pipe parameter correction submodule obtains pipe diameter, connection angle and layout parameter data based on digital twin hydraulic simulation results. It performs residual alignment and trend comparison between the continuous flow observation sequence and simulation output. It uses a genetic algorithm to iteratively optimize the pipe parameter state variables based on the fitness function to generate pipe segment state correction coefficients. The pipe parameter correction submodule consists of a residual analysis unit and a parameter adaptive update unit. Based on digital twin hydraulic simulation results, this module reads initial geometric data such as pipe diameter, connection angle, and layout parameters. The module continuously acquires flow observation sequences over a period of time, compares them with the output sequence of the simulation model, and calculates the residual value (i.e., observed value minus simulated value) at each moment. The update unit uses a genetic algorithm to iteratively optimize pipe parameter state variables (mainly pipe roughness coefficient and effective inner diameter) based on the trend of residual changes over time (e.g., the drift of the residual mean). The optimization logic is set as follows: The module first calculates the moving average of the residuals as a fitness evaluation index, uses the pipe roughness coefficient and effective inner diameter as chromosome-encoded gene loci, sets the population size to 50, the crossover probability to 0.8, and the mutation probability to 0.1. If the residual mean exceeds a preset dead zone range (e.g., ±0.01), genetic iteration is initiated: through selection, crossover, and mutation operations, the module searches for parameter combinations that minimize the residuals, converging to obtain the optimal pipe segment state correction coefficient after 20 iterations. For example, if the observed flow rate is consistently lower than the simulated flow rate and the residual mean is negative 0.2, the genetic algorithm will automatically search for the optimal adjustment amount to increase the roughness coefficient through population evolution, simulate the increase in resistance caused by scaling on the inner wall of the pipe, and thus generate a correction coefficient that can reflect the aging or scaling state of the pipe, ensuring that the simulation model evolves synchronously with the physical entity.

[0034] The branch configuration submodule calls the topology information of the branch pipe segment of the bed terminal according to the pipe segment status correction coefficient, performs matching judgment on the node flow demand input and the conveying capacity of adjacent pipe segments, and performs combination filtering based on the pipe diameter candidate set and the flow difference range to generate the branch diameter configuration value. The branch configuration submodule includes a fluid resistance matching unit and pipe diameter optimization logic. This module calls upon the specific branch pipe topology information of the bed terminal based on the pipe segment status correction coefficient. For each bed, the module inputs the node flow requirement and, combined with the corrected delivery capacity of adjacent pipe segments (i.e., the maximum flow rate that the upstream pipe segment can provide under the current pressure), performs a matching judgment. If the delivery capacity is insufficient, the module initiates a pipe diameter selection procedure. This procedure calculates the head loss along the pipe diameter at a specific flow rate based on a candidate pipe diameter set (e.g., standard or non-standard specifications such as 8 mm, 10 mm, and 12 mm inner diameters), and compares the calculation results with the flow rate difference range. The module selects the pipe diameter combination that meets the flow rate requirement while minimizing pressure loss, generating the final branch diameter configuration value. For example, for a bed with a required flow rate of 600 ml / min, the module, after calculation and comparison, eliminates the 8 mm pipe diameter (causing excessive pressure drop) and the 12 mm pipe diameter (causing excessively low flow rate), ultimately selecting and outputting a non-standard customized branch diameter configuration value of 10.5 mm to achieve optimal hydraulic balance.

[0035] The pressure assessment submodule calls the branch diameter configuration value and connects the pressure acquisition value of the bed node. It performs interval judgment on the pressure acquisition value and the set bed pressure benchmark interval, counts the ratio of the number of qualified nodes to the total number of nodes, calculates the bed pressure compliance rate, and constructs a set of pipeline optimization parameters. The pressure assessment submodule is equipped with interval determination logic and a statistical analysis unit. This module calls the branch diameter configuration value and receives the pressure data collected from the bed nodes in real time. For each collected pressure value, the module performs a strict interval determination against a pre-set bed pressure benchmark interval (e.g., 280 kPa to 320 kPa). If the pressure value falls within the interval, it is marked as "compliant"; otherwise, it is marked as "non-compliant". The statistical analysis unit then traverses the bed nodes, counts the number of nodes marked as "compliant", divides it by the total number of nodes, and calculates the bed pressure compliance rate. Based on this compliance rate and the specific distribution of non-compliant nodes, the module adjusts the pipeline parameters in reverse, constructing a pipeline optimization parameter set that includes suggested pipe diameter adjustments, booster pump power settings, etc. For example, in a dialysis center with 50 beds, 48 ​​nodes are assessed as compliant, resulting in a compliance rate of 96%. For the remaining 2 non-compliant nodes, the module generates optimization parameters for local pipe diameter fine-tuning, aiming to increase the compliance rate to 100%.

[0036] Specifically, such as Figure 2 , 6 As shown, the pipeline manufacturing instruction generation module includes: The geometric structure analysis submodule analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, obtains the pipeline segment axis coordinate sequence and performs vector difference operation, obtains the cross-sectional contour point set and performs closure judgment, collects connection topology identifiers and verifies the relationship between adjacent pipeline segments, and obtains the three-dimensional geometric description matrix of the pipeline. The geometric structure analysis submodule embeds a 3D spatial analysis algorithm and a topology verification unit. This module is responsible for analyzing the geometric dimensions and structural characteristics data in the pipeline network optimization parameter set. The analysis algorithm first extracts the start and end coordinate sequences of the pipe segment's axis and performs vector difference operations to determine the spatial orientation vector and length of the pipe segment. Next, the module acquires the cross-sectional contour point set along the axis normal plane and performs a closure check to ensure that the contour points are connected end-to-end to form a closed shape, thus guaranteeing the watertightness of the pipeline entity. The topology verification unit simultaneously collects the topology identifiers at the connection points, verifies whether adjacent pipe segments interfere or break in space, and confirms whether the connection angle meets the requirements for smooth transition in fluid dynamics. After the above processing, the module outputs a 3D geometric description matrix of the pipeline containing the centerline coordinates of the pipe segments, changes in cross-sectional radius, and connection relationships. For example, the module analyzes that the centerline of a certain elbow pipe segment is an arc with a radius of 50 mm and a cross-section of a ring with an inner diameter of 10 mm, maintaining G1 continuity with the upstream and downstream straight pipe segments, together forming a precise 3D manufacturing blueprint.

[0037] The path planning calculation submodule, based on the pipe's three-dimensional geometric description matrix, obtains the height sequence between printing layers and performs layer-by-layer mapping. It calculates the candidate sequence of path points for the multi-layer cross-sectional contour point set, compares the spacing between path points in adjacent layers and adjusts the direction order to generate a set of printing path trajectories. The path planning calculation submodule operates a layered slicing algorithm and path optimization logic. Based on the pipe's 3D geometric description matrix, this module sets the inter-layer height sequence (e.g., setting the layer height to 0.2 mm). The module performs layer-by-layer mapping of the 3D model along the Z-axis, obtaining a set of multi-layer cross-sectional contour points on each layer's height plane. For each layer contour, the module calculates candidate sequence of path points for the filling path and the outer wall contour path. The optimization logic adjusts the path point order by comparing the Euclidean distance between path points in adjacent layers to reduce nozzle idle movement and avoid stringing across layers. Finally, the module generates a set of printing path trajectories containing a series of 3D coordinate points, clearly defining the nozzle's specific movement path in each layer. For example, for a pipe with a height of 100 mm, the module divides it into 500 layers and plans a continuous spiral printing trajectory from the inside out, ensuring the smoothness of the pipe's inner wall and minimizing fluid resistance.

[0038] The manufacturing instruction orchestration submodule obtains the trajectory segment length change rate and performs interval division based on the printing path trajectory set. It calculates the printing injection rate, material flow rate and temperature control interval based on the interval results, and performs timing combination according to the trajectory segment arrangement order to generate a 3D printing manufacturing instruction sequence. The manufacturing instruction orchestration submodule integrates a G-code generator and a process parameter matching library. Based on the set of printing path trajectories, this module first calculates the rate of change of the trajectory segment length, i.e., the path curvature, and divides the trajectory into intervals (e.g., straight segment intervals, large curvature turning intervals). For different interval results, the module calculates suitable printing parameters: increasing the printing jet rate to 60 mm / s in straight segments and decreasing it to 30 mm / s in turning intervals to ensure accuracy. Simultaneously, the module calculates the material flow rate (E-axis extrusion rate) and temperature control intervals (e.g., setting the nozzle temperature to 240 degrees Celsius and the heated bed temperature to 80 degrees Celsius). Finally, based on the physical arrangement of the trajectory segments, the module combines the execution timing of position, speed, temperature, and extrusion instructions to generate a 3D printing manufacturing instruction sequence (G-code) conforming to the 3D printer communication protocol. As shown in Table 2, this instruction sequence precisely controls multiple physical parameters in the manufacturing process.

[0039] Table 2: Comparison Table of 3D Printing Manufacturing Instructions and Process Parameters

[0040] As shown in Table 2, the module generates a sequence of instructions that includes precise coordinate positioning and dynamic process parameter adjustment. In particular, the extrusion rate is finely adjusted along the path to ensure the uniformity of material deposition, which is crucial for ensuring the pressure resistance of the hemodialysis tubing.

[0041] Specifically, such as Figure 2 , 7 As shown, the 3D printed pipe forming module includes: The trajectory control submodule, based on the 3D printing manufacturing instruction sequence, obtains the nozzle spatial displacement instruction, collects the displacement change corresponding to the time axis, compares the displacement change point by point according to the equipment stroke parameters, performs coordinate recursion calculation according to the continuous interpolation rule, and generates the nozzle trajectory coordinate sequence. The trajectory control submodule includes a motion interpolation controller and a position feedback closed loop. Based on the 3D printing manufacturing instruction sequence, this module extracts the nozzle spatial displacement instructions. It collects the corresponding displacement change on the time axis, i.e., the difference between the target position and the current position, and compares the displacement change point-by-point according to the equipment's travel parameters (such as the stepper motor's step angle and the lead screw pitch). The module performs coordinate recursion calculations according to continuous interpolation rules (such as linear interpolation or circular interpolation), decomposing long-distance displacement instructions into micron-level stepping pulse sequences to generate a high-density nozzle trajectory coordinate sequence. This process ensures that the nozzle can strictly follow the planned path with extremely high trajectory accuracy (error controlled within 0.05 mm). For example, when printing a 90-degree pipe bend, the module generates thousands of dense intermediate coordinate points using a circular interpolation algorithm, driving the motor to smoothly coordinate, thus avoiding the sharp-edged step effect and ensuring the streamlined structure of the pipe's inner wall.

[0042] The extrusion adjustment submodule collects the nozzle running speed and material supply rate samples based on the nozzle trajectory coordinate sequence, aligns the samples in time, calculates the material distribution per unit path according to the relationship between path length and speed, judges the distribution deviation within an interval, and generates the extrusion flow calibration value. The extrusion regulation submodule is equipped with a dynamic flow balancing algorithm. This module collects real-time samples of the nozzle running speed and material supply rate (i.e., the extrusion motor speed) based on the nozzle trajectory coordinate sequence. The module first performs strict time alignment on these two samples, and then calculates the theoretical material distribution required per unit path based on the physical relationship between path length (L), nozzle moving speed (V), layer height (H), and line width (W). The calculation logic is: theoretical extrusion volume equals nozzle speed multiplied by layer height multiplied by line width multiplied by the time derivative. The module reads the extruder's filament diameter parameters (1.75mm or 2.85mm), the effective radius parameters of the feed roller (5-8mm), and the step angle parameters of the extruder's stepper motor (e.g., 1.8°, corresponding to 200 steps per revolution). Based on the sampling of the extrusion motor speed, the module first calculates the number of motor rotations (speed × time derivative), then calculates the filament feeding length (number of rotations × 2π × filament feeding wheel radius), and finally converts the motor speed into the actual extrusion volume using the formula: actual extrusion volume = filament cross-sectional area × filament feeding length = π × (filament diameter / 2)² × filament feeding length. The module compares this theoretical extrusion volume with the converted actual extrusion volume to calculate the distribution deviation. For this deviation, a range judgment is performed. If the deviation exceeds ±2%, the module immediately generates a compensation signal and adjusts the extrusion flow rate calibration value. For example, when the actual extrusion volume is detected to be low when the nozzle is making high-speed bends, the module calculates that the extrusion rate needs to be increased by 5% and sends the corrected flow rate calibration value to the extrusion motor controller to prevent tube wall porosity defects caused by under-extrusion.

[0043] The forming integration submodule calls the extrusion flow calibration value, monitors the printing layer thickness sampling amount and the forming area temperature sampling amount, calculates the difference between the layer thickness sampling amount and the layer thickness benchmark, and makes a range judgment on the temperature sampling amount. It then performs solid deposition according to the jetting control sequence to generate hemodialysis water supply pipes. The forming integration submodule has an environmental monitoring and fused deposition execution unit. This module calls the extrusion flow rate calibration value and continuously monitors the sampled thickness of the printed layer (acquired via a laser rangefinder) and the sampled ambient temperature of the forming area. The module calculates the difference between the sampled layer thickness and the preset layer thickness benchmark (0.2 mm). If the cumulative layer thickness deviation exceeds 0.05 mm, it automatically adjusts the Z-axis height for compensation. Simultaneously, the module performs range judgment on the temperature sampled volume to ensure that the ambient temperature is maintained within the optimal window for material crystallization (e.g., 60°C to 70°C) to prevent warping and deformation. After the parameters are confirmed to be correct, the module drives the nozzle to deposit molten medical-grade polymer material layer by layer according to the jet control sequence. After each layer is deposited, the material rapidly cools and solidifies, firmly bonding with the previous layer through molecular chain entanglement, ultimately generating a hemodialysis water supply pipe with a smooth inner wall, precise dimensions, and meeting biocompatibility requirements. The pipe has been tested, and its pressure resistance and fluid delivery efficiency have met the design standards, as shown in the experimental verification results in Table 3.

[0044] Table 3: Performance Test Results of Customized Hemodialysis Water Supply Fittings

[0045] As shown in Table 3, through the digital twin and precision control manufacturing of the entire process described above, the hemodialysis water supply pipes produced in this embodiment are significantly superior to products manufactured by traditional processes in terms of pressure resistance, surface quality and dimensional accuracy, proving the effectiveness and superiority of this customized system.

[0046] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of protection of the described technical solutions.

Claims

1. A customized hemodialysis water supply network system based on digital twin and 3D printing, characterized in that, The system includes: The sensor data acquisition module collects electrical signals of water flow velocity, water pressure and water circuit status at key nodes in the hemodialysis water supply system, performs analog-to-digital conversion to form digital signals, obtains the equipment model and usage frequency of the bed, calculates the target water pressure range and flow requirements, and constructs a node status dataset. The digital twin hydraulic simulation module, based on the node status dataset, uses computational fluid dynamics algorithms to analyze the water flow velocity distribution and water pressure conditions, identify the hydraulic supply deviation value of the bed terminal, calculate the dynamic hydraulic and water flow distribution characteristics and the bed hydraulic matching degree, and output the digital twin hydraulic simulation results. The structural collaborative optimization module, based on the digital twin hydraulic simulation results, uses a genetic algorithm to iteratively adjust the pipe diameter, connection angle, and layout parameters, implements differentiated diameter design and flow allocation for the bed terminal branch pipe sections, calculates the bed pressure compliance rate, and constructs a set of pipe network optimization parameters; The pipeline manufacturing instruction generation module analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, calculates the 3D printing path, determines the printing jet rate, material flow rate and temperature control, and generates a 3D printing manufacturing instruction sequence.

2. The digital twin and 3D printing based customized hemodialysis water supply network system according to claim 1, characterized in that, The node status dataset includes bed water consumption level, node water supply load level, water circuit operation status identifier, and stable operation status. The digital twin hydraulic simulation results include node water pressure distribution, flow distribution, water flow state distribution, and bed hydraulic matching results. The pipeline network optimization parameter set includes pipe diameter specification configuration, branch flow distribution ratio, system pressure balance index, and bed water supply compliance index. The 3D printing manufacturing instruction sequence includes forming path data, printing speed parameters, layer thickness parameters, and printing temperature parameters.

3. The digital twin and 3D printing based customized hemodialysis water supply network system according to claim 1, characterized in that, The sensor data acquisition module includes: The signal acquisition submodule acquires water flow velocity signals, water pressure signals, and water circuit status electrical signals of key nodes in the hemodialysis water supply system. It performs synchronous alignment based on channel amplitude changes and sampling period, quantizes and sorts discrete sampled values ​​and binds them with time stamps, and generates a set of node digital signal sequences. The demand calculation submodule, based on the set of node digital signal sequences, obtains the bed equipment model code and usage frequency record, reads the model-related water pressure reference range and flow reference range, performs range mapping and comparison calculation based on the node water flow velocity value and water pressure value, and generates a set of bed water supply demand parameters. The state construction submodule calls the set of bed water supply demand parameters and the set of node digital signal sequences, performs consistency comparison on the node water flow velocity value, water pressure value, and water circuit state electrical signal, and performs node state matching, item integration and structured arrangement based on the numerical correspondence to establish a node state dataset.

4. The digital twin and 3D printing based customized hemodialysis water supply network system according to claim 3, characterized in that, The water pressure reference range and flow rate reference range are generated from the original usage data, equipment parameters and actual operating conditions related to the bed equipment model. Specifically, by collecting the operating data of the bed equipment, based on the typical water pressure and flow rate range of different equipment models, combined with the equipment usage frequency and common operating environment, the minimum and maximum values ​​of water pressure and flow rate of the equipment under preset rated operating conditions are determined as the upper and lower limits of the water pressure reference range and flow rate reference range.

5. The digital twin and 3D printing based customized hemodialysis water supply network system according to claim 1, wherein, The digital twin hydraulic simulation module includes: The node status construction submodule collects bed terminal flow records, water pressure sampling values ​​and timestamp sequences based on the node status dataset. It performs time alignment on the flow sequence and water pressure sequence according to the node identifier, performs interpolation compensation and missing label processing, rearranges the node data items, and generates a node hydraulic status vector set. The flow pressure deviation identification submodule calls the node hydraulic state vector set, calculates the simulated flow velocity and simulated water pressure of each node according to the node topology association and pipeline geometry, performs alignment comparison with the measured values ​​and extracts the flow velocity difference and water pressure difference sequence, performs scale normalization and direction calibration, calculates the supply offset according to the bed position identifier, and obtains the bed hydraulic supply deviation value. The twin matching output submodule performs joint mapping on the velocity and pressure distribution matrix based on the bed hydraulic supply deviation value and the node hydraulic state vector set, synchronously integrates the waterway operation status signal, calculates the matching degree based on the bed number, and establishes digital twin hydraulic simulation results.

6. The digital twin and 3D printing based customized hemodialysis water supply network system, as claimed in claim 1 wherein, The structural collaborative optimization module includes: The pipe parameter correction submodule obtains pipe diameter, connection angle and layout parameter data based on the digital twin hydraulic simulation results. It performs residual alignment and trend comparison between the continuous flow observation sequence and the simulation output. It uses a genetic algorithm to iteratively optimize the pipe parameter state variables based on the fitness function to generate pipe segment state correction coefficients. The branch configuration submodule calls the topology information of the branch pipe segment of the bed terminal according to the pipe segment status correction coefficient, performs matching judgment on the node flow demand input and the conveying capacity of adjacent pipe segments, and performs combination screening based on the pipe diameter candidate set and the flow difference range to generate the branch diameter configuration value. The pressure assessment submodule calls the branch diameter configuration value and connects the pressure acquisition value of the bed node. It performs interval determination on the pressure acquisition value and the set bed pressure benchmark interval, counts the ratio of the number of qualified nodes to the total number of nodes, calculates the bed pressure compliance rate, and constructs a pipeline optimization parameter set.

7. The digital twin and 3D printing based customized hemodialysis water supply network system according to claim 1, wherein, The pipeline manufacturing instruction generation module includes: The geometric structure analysis submodule analyzes the geometric dimensions and structural characteristics of the pipeline network optimization parameter set, obtains the pipeline segment axis coordinate sequence and performs vector difference operation, obtains the cross-sectional contour point set and performs closure judgment, collects connection topology identifiers and verifies the relationship between adjacent pipeline segments, and obtains the pipeline three-dimensional geometric description matrix. The path planning calculation submodule, based on the three-dimensional geometric description matrix of the pipeline, obtains the height sequence between printing layers and performs layer-by-layer mapping. For the multi-layer cross-sectional contour point set, it calculates the path point candidate sequence, compares the path point spacing between adjacent layers and adjusts the direction order to generate a set of printing path trajectories. The manufacturing instruction arrangement submodule obtains the trajectory segment length change rate and performs interval division based on the printing path trajectory set, calculates the printing injection rate, material flow rate and temperature control interval based on the interval results, and performs timing combination according to the trajectory segment arrangement order to generate a 3D printing manufacturing instruction sequence.

8. The digital twin and 3D printing based customized hemodialysis water supply network system, as claimed in claim 1 wherein, The system also includes: The 3D printing pipe forming module, based on the 3D printing manufacturing instruction sequence, executes nozzle dynamic trajectory control, adjusts the printing nozzle speed and material extrusion amount, and implements real-time monitoring and feedback adjustment of printing layer thickness and forming temperature to form hemodialysis water supply pipes; The hemodialysis water supply pipe fittings include the pipe fitting geometry, internal flow channel structure, and interface connection type.

9. The digital twin and 3D printing based customized hemodialysis water supply network system, according to claim 8, wherein, The 3D printed tube forming module includes: The trajectory control submodule, based on the 3D printing manufacturing instruction sequence, obtains the nozzle spatial displacement instruction, collects the displacement change corresponding to the time axis, compares the displacement change point by point according to the equipment stroke parameters, performs coordinate recursion calculation according to the continuous interpolation rule, and generates the nozzle trajectory coordinate sequence. The extrusion adjustment submodule collects the nozzle running speed sampling and material supply rate sampling based on the nozzle trajectory coordinate sequence, aligns the sampling in time, calculates the material distribution per unit path according to the relationship between path length and speed, judges the deviation of the distribution in intervals, and generates an extrusion flow rate calibration value. The forming integration submodule calls the extrusion flow calibration value, monitors the printing layer thickness sampling amount and the forming area temperature sampling amount, calculates the difference between the layer thickness sampling amount and the layer thickness benchmark, and makes a range judgment on the temperature sampling amount. It then performs solid deposition according to the jet control sequence to generate the hemodialysis water supply pipe.

10. The digital twin and 3D printing based customized hemodialysis water supply network system, according to claim 9, wherein, The layer thickness reference is a target layer thickness value determined based on the printer's set printing parameters, material type, and printing process requirements, serving as a reference standard for the actual thickness during the printing process.

Images