Heating system whole life cycle predictive maintenance method based on digital twinning

Abstract

The present application belongs to the technical field of heating pipe network operation and maintenance, and provides a heating system full life cycle predictive maintenance method based on digital twinning, comprising: constructing a digital twinning model and separately marking a pipe elbow simulation unit, collecting water body ion concentration parameters of the pipe network in real time, calculating a water body scaling saturation index in combination with a solubility product constant, realizing pre-judgment of initial precipitation risk, under the condition of existing initial precipitation risk, determining whether the elbow local part satisfies the substantial precipitation precipitation condition in combination with the scaling saturation index, constructing a water body flow difference-substantial precipitation amount coupling model by analyzing the relevance of water body flow difference and actual precipitation amount in historical operation samples, realizing prediction of the elbow local precipitation amount, finally inputting the real-time flow difference into the coupling model to obtain the current elbow substantial precipitation amount, and automatically generating a corresponding elbow position fixed-point cleaning maintenance instruction when the precipitation amount exceeds a safety threshold.

Description

Technical Field

[0001] This invention belongs to the field of heating network operation and maintenance technology, specifically a predictive maintenance method for the entire life cycle of a heating system based on digital twins. Background Technology

[0002] Existing predictive maintenance solutions for heating systems are mostly designed for inorganic scale in tap water replenishment systems. They rely on macroscopic parameters such as system-wide pressure difference, temperature difference, and total flow rate for monitoring. They are only suitable for the operation and maintenance of clean water quality pipe networks and generally suffer from the defects of single scenario and lagging monitoring.

[0003] Heating secondary systems that use greywater or reclaimed water for replenishment have unique operation and maintenance challenges: the secondary pipe network maintains an operating water temperature of 30-50℃ year-round, which is exactly the right range for the reproduction of corrosive microorganisms such as iron bacteria and sulfate-reducing bacteria; in addition, greywater is rich in nutrients such as organic matter, ferrous ions, and sulfates, which will encourage microorganisms to secrete a large amount of extracellular polymeric substances (EPS), which will carry water impurities and corrosion products to form biological slime on the pipe wall, thus causing problems such as under-deposit corrosion and reduced heating efficiency.

[0004] Pipe bends are high-risk areas for biofilm deposition. Fluid flow through bends generates eddies that separate from the boundary layer, causing localized velocity reduction and media stagnation and accumulation, resulting in a significantly higher deposition risk compared to straight pipe sections. In the early stages of biofilm formation, it only causes minor changes in the local flow field at the bend, without affecting global hydraulic and thermal parameters. Traditional monitoring methods cannot identify early, hidden problems. By the time global parameters trigger alarms, the biofilm has already hardened and compacted, causing irreversible damage through under-deposit corrosion, thus missing the optimal maintenance window.

[0005] Current digital twin heating operation and maintenance technologies have significant shortcomings. Most models employ homogeneous modeling of the entire pipe network, failing to perform detailed, independent simulations of easily deposited structures at elbows. This makes it impossible to obtain microscopic parameters such as local flow velocity and accumulated media retention at elbows. Furthermore, existing models do not consider the coupling effect of reclaimed water quality characteristics and the suitable temperature range for microorganisms, making it impossible to distinguish the deposition mechanisms of inorganic scale and biological slime. This hinders the accurate quantification of the evolution trends of local slime deposition and under-scale corrosion at elbows, preventing early prediction and refined predictive maintenance throughout the entire lifecycle, and making it difficult to adapt to the specific operation and maintenance needs of reclaimed water replenishment heating systems.

[0006] To this end, the present invention provides a predictive maintenance method for the entire life cycle of a heating system based on digital twins. Summary of the Invention

[0007] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0008] The technical solution adopted by this invention to solve its technical problem is:

[0009] One of the objectives of this invention is to provide a predictive maintenance method for the entire lifecycle of a heating system based on digital twins, comprising:

[0010] Step S10: Construct a digital twin model of the heating system, mark the position of the simulation unit of the pipe bend in the digital twin model, obtain the real-time water ion concentration parameters of the heating system, and calculate the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial precipitation risk.

[0011] Step S20: If there is an initial risk of precipitation, obtain the local flow field data inside the pipe elbow simulation unit in the digital twin model, calculate the water flow difference between the inside and outside of the pipe elbow simulation unit, and analyze whether the conditions for substantial precipitation are met under the current water flow difference by combining the water scaling saturation index.

[0012] Step S30: If the conditions for substantial precipitation are met, then for the simulation unit of the pipe bend, analyze the linear relationship between the water flow difference sample and the corresponding actual precipitation amount, and construct a coupled model of water flow difference-substantial precipitation amount.

[0013] Step S40: Input the currently calculated water flow difference into the water flow difference-actual sedimentation coupling model, calculate the actual sedimentation at the current pipeline bend simulation unit location, and when the actual sedimentation exceeds the safety threshold, determine that actual sedimentation has occurred and generate a cleaning and maintenance instruction for the current pipeline bend simulation unit location.

[0014] As a further improvement of the present invention, the construction process of the digital twin model is as follows:

[0015] Obtain data on the physical network topology of the heating system, pipe material parameters, and equipment operating parameters;

[0016] A digital twin model is constructed using 3D modeling software based on physical pipeline network topology data, pipeline material parameters, and equipment operating parameters.

[0017] As a further improvement of the present invention, the specific process of marking the position of the simulation unit of the pipe elbow in the digital twin model is as follows:

[0018] In the digital twin model of the heating system, all pipe segments where the fluid flow direction changes are identified, the pipe segments are marked as pipe elbow simulation units, and the three-dimensional spatial coordinates of each pipe elbow simulation unit are recorded.

[0019] As a further improvement of the present invention, the specific process for determining whether there is an initial precipitation risk is as follows:

[0020] Obtain the ion concentration parameters of the water in the heating system, including calcium ion concentration, magnesium ion concentration, carbonate ion concentration and sulfate ion concentration;

[0021] Obtain the solubility product constant of the target precipitate at the current water temperature. The target precipitate includes calcium carbonate, calcium sulfate, and magnesium carbonate.

[0022] Based on the ion concentration parameters of the water body, calculate the actual ion product of the corresponding ions of each target precipitate in the water body;

[0023] The ratio of each actual ion product to the corresponding solubility product constant is calculated to obtain the scaling saturation index of the water body corresponding to each target precipitate.

[0024] The maximum value of the scaling saturation index of the water body corresponding to each target precipitate is extracted as the comprehensive scaling saturation index of the water body;

[0025] The comprehensive water body scaling saturation index was compared with the initial sedimentation risk threshold;

[0026] If the overall water body scaling saturation index is greater than the initial sedimentation risk threshold, then an initial sedimentation risk is determined to exist.

[0027] If the overall water body scaling saturation index is less than or equal to the initial sedimentation risk threshold, then it is determined that there is no initial sedimentation risk.

[0028] As a further improvement of the present invention, the specific process of analyzing whether the conditions for substantial precipitation are met under the current water flow difference is as follows:

[0029] Obtain the current real-time total flow of the heating system, as well as the geometric parameters of the pipe bend simulation unit, including the pipe inner diameter and the bend curvature radius;

[0030] Calculate the average flow velocity of the fluid in the straight pipe section based on the real-time total flow rate and the pipe inner diameter;

[0031] The ratio of the pipe's inner diameter to the elbow's radius of curvature is calculated to obtain the elbow curvature ratio.

[0032] Obtain the flow field offset coefficient, and calculate the equivalent average flow velocity of the inner half-section and the equivalent average flow velocity of the outer half-section of the bend based on the average flow velocity, the bend curvature ratio and the flow field offset coefficient.

[0033] Multiply the equivalent average velocity of the inner half-section of the elbow by the half-section area of ​​the pipe to obtain the water flow rate inside the elbow; multiply the equivalent average velocity of the outer half-section of the elbow by the half-section area of ​​the pipe to obtain the water flow rate outside the elbow.

[0034] The absolute difference between the water flow rate on the outside of the elbow and the water flow rate on the inside of the elbow is calculated to obtain the water flow rate difference between the inside and outside of the pipe elbow simulation unit; and the ratio of the water flow rate difference to the real-time total flow rate is calculated to obtain the dimensionless flow rate difference coefficient.

[0035] The precipitation judgment value is obtained by multiplying the comprehensive water body scaling saturation index with the flow difference coefficient.

[0036] If the precipitation judgment value is greater than the actual precipitation threshold, then the actual precipitation condition is met under the current water flow difference.

[0037] If the precipitation determination value is less than or equal to the actual precipitation threshold, then the actual precipitation condition is not met.

[0038] As a further improvement of the present invention, the specific process of analyzing the linear correlation between the water flow difference sample and the corresponding actual sedimentation amount is as follows:

[0039] Obtain multiple historical operating time periods of the pipe bend simulation unit within the historical operating cycle of the heating system;

[0040] Extract the historical water flow difference calculated within each historical operating period as a water flow difference sample;

[0041] Retrieve the actual sedimentation amount corresponding to each historical operating period as a sample of actual sedimentation amount;

[0042] Multiple initial data sample pairs are constructed by matching the water flow difference samples within the same historical operating period with the actual sedimentation samples one by one.

[0043] Outlier detection and removal are performed on multiple initial data sample pairs to obtain an effective sample set;

[0044] Based on the effective sample set, the Pearson correlation coefficient between the water flow difference sample and the actual sedimentation sample was calculated.

[0045] If the absolute value of the Pearson correlation coefficient is greater than or equal to the linear correlation threshold, then it is determined that there is a significant linear correlation between the water flow difference sample and the actual sedimentation amount.

[0046] If the absolute value of the Pearson correlation coefficient is less than the linear correlation threshold, it is determined that there is no significant linear correlation between the water flow difference sample and the actual sedimentation amount.

[0047] As a further improvement of the present invention, the specific process of constructing the water flow difference-actual sedimentation amount coupling model is as follows:

[0048] In the presence of a significant linear correlation, the least squares method was used to perform a univariate linear regression fitting on the effective sample set with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable, and the linear regression coefficient and intercept were calculated.

[0049] Based on the linear regression coefficients and the intercept, a coupled model of water flow difference and actual sedimentation is constructed. The expression of the coupled model is: Actual sedimentation = Linear regression coefficient × Water flow difference + Intercept.

[0050] As a further improvement of the present invention, the specific process of constructing the water flow difference-actual sedimentation amount coupling model also includes:

[0051] In the absence of a significant linear correlation, a nonlinear coupling model of water flow difference and actual sedimentation is constructed by using a multinomial nonlinear fitting method to regress the effective sample set with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable.

[0052] As a further improvement of the present invention, the specific process for calculating the actual sedimentation amount at the current pipeline elbow simulation unit position is as follows:

[0053] The water flow difference calculated in real time is obtained, and the water flow difference is substituted into the water flow difference-actual sedimentation amount coupling model to obtain the predicted actual sedimentation amount at the current pipeline bend simulation unit position.

[0054] As a further improvement of the present invention, the specific process of generating cleaning and maintenance instructions for the current pipeline bend simulation unit position is as follows:

[0055] If the predicted amount of actual sedimentation is greater than the safety threshold, then it is determined that actual sedimentation has occurred at the current pipeline elbow simulation unit location.

[0056] If substantial sedimentation is determined to have occurred, the three-dimensional spatial coordinates and unique equipment identification code of the current pipe bend simulation unit are extracted from the digital twin model of the heating system.

[0057] Based on three-dimensional spatial coordinates, unique equipment identification code, and predicted actual sedimentation amount, a cleaning and maintenance instruction is generated for the current simulated unit location of the pipe bend.

[0058] The cleaning and maintenance instructions include three-dimensional spatial coordinates, the equipment's unique identification code, and the predicted amount of actual sediment.

[0059] The second objective of this invention is to provide a predictive maintenance system for the entire lifecycle of a heating system based on digital twins, comprising the following modules:

[0060] Modeling Risk Assessment Module: Construct a digital twin model of the heating system, mark the position of the simulation unit of pipe bend in the digital twin model, obtain the real-time water ion concentration parameters of the heating system, and calculate the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial precipitation risk.

[0061] Elbow flow field discrimination module: If there is an initial sedimentation risk, the local flow field data inside the pipe elbow simulation unit is obtained in the digital twin model, the water flow difference between the inside and outside of the pipe elbow simulation unit is calculated, and the water scaling saturation index is combined to analyze whether the conditions for substantial sedimentation are met under the current water flow difference.

[0062] Coupled model training module: If the condition for substantial precipitation is met, then for the simulation unit of the pipe bend, analyze the linear relationship between the water flow difference sample and the corresponding actual precipitation amount, and construct a coupled model of water flow difference-substantial precipitation amount.

[0063] Predictive maintenance instruction module: Input the currently calculated water flow difference into the water flow difference-actual sedimentation amount coupling model, calculate the actual sedimentation amount at the current pipeline bend simulation unit location, and when the actual sedimentation amount exceeds the safety threshold, determine that actual sedimentation has occurred and generate cleaning and maintenance instructions for the current pipeline bend simulation unit location.

[0064] The beneficial effects of this invention are:

[0065] First, this invention enables early and accurate prediction of hidden scaling in pipe bends, overcoming the shortcomings of traditional monitoring methods that are often delayed. Traditional heating scaling monitoring relies on macroscopic parameters such as global pressure and temperature differences, which can only identify visible faults after severe scaling in the pipe network, failing to capture early hidden scaling caused by local flow field distortions at bends. This invention combines water ion concentration and solubility product constant to calculate the scaling saturation index, predicting the initial precipitation risk from a thermodynamic perspective. Simultaneously, it uses a digital twin model to finely solve the flow difference between the inside and outside of the bend, characterizing the amplification effect of the local flow field on scaling deposition. This allows for early identification of potential scaling hazards at bends even when global pipe network parameters are normal, effectively advancing the fault prediction window.

[0066] Secondly, it enables refined and targeted operation and maintenance of the pipeline network, effectively reducing costs and increasing efficiency. Traditional heating pipeline networks often adopt a crude operation and maintenance model of uniform cleaning across the entire area, which is poorly targeted and results in serious resource waste. This invention relies on a digital twin elbow independent simulation unit, which can accurately locate the local scaling situation of a single elbow and quantitatively predict the amount of sediment. It generates targeted cleaning instructions only for elbows with excessive scaling, realizing precise operation and maintenance on demand. This avoids the waste of manpower and resources caused by blindly cleaning the entire area, and significantly improves the refinement and efficiency of pipeline network operation and maintenance.

[0067] Third, it adapts to various scaling scenarios, improving the accuracy and comprehensiveness of scaling prediction. This invention covers mainstream scaling substances in pipe networks such as calcium carbonate, calcium sulfate, and magnesium carbonate. It comprehensively determines scaling risk through multiple types of scale saturation indices, adapting to fluctuating water quality conditions and avoiding the problems of missed or false diagnoses caused by single scale detection. Simultaneously, it combines thermodynamic precipitation conditions and local flow field deposition characteristics for dual judgment, closely aligning with the actual evolution of scaling in elbows, significantly improving prediction accuracy, ensuring the stability of pipe network flow and heat exchange, and extending equipment lifespan. Attached Figure Description

[0068] The invention will now be further described with reference to the accompanying drawings.

[0069] Figure 1 This is a flowchart of the steps of the predictive maintenance method for the entire life cycle of a heating system based on digital twins, as described in this invention.

[0070] Figure 2 This is a system module diagram of the predictive maintenance system for the entire life cycle of a heating system based on digital twins, as proposed in this invention. Detailed Implementation

[0071] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0072] Example 1

[0073] like Figure 1 As shown in the embodiment of the present invention, the predictive maintenance method for the entire life cycle of a heating system based on digital twin technology is based on digital twin technology. It performs initial risk assessment by analyzing water ion concentration and solubility product constant, then quantifies the deposition amplification effect of local flow field distortion through the flow difference between the inside and outside of elbows, coupling dual conditions to determine the substantial precipitation condition. Simultaneously, it constructs an adaptive flow difference-precipitation coupling model based on historical sample data to achieve quantitative prediction of local scaling at elbows, and accurately triggers targeted maintenance commands based on precipitation thresholds. Specifically, it includes the following steps:

[0074] Step S10: Construct a digital twin model of the heating system, mark the position of the simulation unit of the pipe bend in the digital twin model, obtain the real-time water ion concentration parameters of the heating system, and calculate the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial precipitation risk.

[0075] In some embodiments, the specific process of constructing a digital twin model of the heating system in step S10 is as follows:

[0076] Obtain data on the physical network topology of the heating system, pipe material parameters, and equipment operating parameters;

[0077] Based on the physical pipeline topology data, pipeline material parameters, and equipment operating parameters, a digital twin model of the heating system that maps to the physical heating system is constructed using 3D modeling software (such as SolidWorks, Revit, and ANSYS DesignModeler).

[0078] In order to meet the requirements for accurate simulation of the local flow field at the pipe bend, the curvature radius, pipe diameter change and inner wall roughness at the pipe bend need to be marked when constructing the digital twin model of the heating system.

[0079] In some embodiments, the specific process of marking the position of the pipe elbow simulation unit in the digital twin model in step S10 is as follows:

[0080] In the digital twin model of the heating system, all pipe segments where the fluid flow direction changes are identified, the pipe segments are marked as pipe bend simulation units, and the three-dimensional spatial coordinates of each pipe bend simulation unit are recorded.

[0081] In some embodiments, the specific process of obtaining real-time water ion concentration parameters of the heating system in step S10, and calculating the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial sedimentation risk, is as follows:

[0082] Online water quality monitoring instruments deployed at the bends of the physical pipe network of the heating system are used to obtain the water ion concentration parameters of the heating system in real time. The water ion concentration parameters include at least the concentrations of calcium ions, magnesium ions, carbonate ions, and sulfate ions.

[0083] Obtain the solubility product constant of the target precipitate at the current water temperature, wherein the target precipitate includes calcium carbonate, calcium sulfate, and magnesium carbonate;

[0084] Based on the water body ion concentration parameters, calculate the actual ion product of the corresponding ions of each target precipitate in the water body;

[0085] For example, based on the water body ion concentration parameters, the actual ion product of the corresponding ions of each target precipitate in the water body is calculated, and the specific calculation formula is as follows: ,in, , , These are the actual ion products of calcium carbonate, calcium sulfate, and magnesium carbonate, respectively. , , , These are the real-time concentrations of calcium ions, magnesium ions, carbonate ions, and sulfate ions, respectively.

[0086] The ratio of each actual ion product to the corresponding solubility product constant is calculated to obtain the water body scaling saturation index corresponding to each target precipitate.

[0087] For example, the ratio of each actual ion product to the corresponding solubility product constant is calculated to obtain the scaling saturation index of the water body corresponding to each target precipitate. The specific calculation formula is as follows: , , ,in, , , These are the scaling saturation indices for calcium carbonate, calcium sulfate, and magnesium carbonate, respectively. , , These are the solubility product constants of calcium carbonate, calcium sulfate, and magnesium carbonate at the current water temperature, respectively.

[0088] The maximum value of the scaling saturation index of the water body corresponding to each target precipitate is extracted as the comprehensive scaling saturation index of the water body;

[0089] The comprehensive water body scaling saturation index is compared with the preset initial sedimentation risk threshold;

[0090] If the overall water body scaling saturation index is greater than the initial sedimentation risk threshold, then an initial sedimentation risk is determined to exist;

[0091] If the comprehensive water body scaling saturation index is less than or equal to the initial sedimentation risk threshold, then it is determined that there is no initial sedimentation risk.

[0092] It should be noted that the initial precipitation risk threshold is usually set to 1.0. When the scaling saturation index of the water body is greater than 1.0, it indicates that the scale-forming ions in the water body are in a supersaturated state and have a tendency to precipitate, that is, there is an initial precipitation risk. When the scaling saturation index of the water body is equal to 1.0, the water body is in a state of dissolution equilibrium. When the scaling saturation index of the water body is less than 1.0, the water body is in an unsaturated state, and the original precipitate may even dissolve.

[0093] Step S20: If there is an initial risk of precipitation, obtain the local flow field data inside the pipe elbow simulation unit in the digital twin model, calculate the water flow difference between the inside and outside of the pipe elbow simulation unit, and analyze whether the conditions for substantial precipitation are met under the current water flow difference by combining the water scaling saturation index.

[0094] In some embodiments, in step S20, if there is an initial risk of precipitation, the specific process of obtaining local flow field data inside the pipe elbow simulation unit in the digital twin model, calculating the water flow difference between the inside and outside of the pipe elbow simulation unit, and analyzing whether the conditions for substantial precipitation are met under the current water flow difference, in conjunction with the water scaling saturation index, is as follows:

[0095] Obtain the current real-time total flow of the heating system, as well as the geometric parameters of the pipe elbow simulation unit, including the pipe inner diameter and the elbow curvature radius;

[0096] Based on the real-time total flow rate and the pipe inner diameter, calculate the average flow velocity of the fluid in the straight pipe section; for example, the specific calculation formula is as follows: ,in, The average flow velocity, This represents the total real-time traffic. This refers to the inner diameter of the pipe.

[0097] The ratio of the pipe's inner diameter to the elbow's radius of curvature is calculated to obtain the elbow curvature ratio; for example, the specific calculation formula is as follows: ,in, The curvature ratio of the elbow. The radius of curvature of the elbow;

[0098] Obtain a preset flow field offset coefficient. Based on the average flow velocity, the bend curvature ratio, and the flow field offset coefficient, calculate the equivalent average flow velocity of the inner half-section and the outer half-section of the bend using empirical formulas for flow velocity distribution. For example, the specific calculation formula is as follows: ,in, The equivalent average flow velocity is the cross-sectional area on the inner side of the elbow. The equivalent average flow velocity is the velocity of the outer half-section of the elbow. This is the flow field offset coefficient;

[0099] Multiplying the equivalent average flow velocity of the inner half-section of the elbow by the half-section area of ​​the pipe yields the water flow rate inside the elbow; and multiplying the equivalent average flow velocity of the outer half-section of the elbow by the half-section area of ​​the pipe yields the water flow rate outside the elbow; for example, the specific calculation formula is as follows: ,in, This refers to the water flow rate inside the bend. The water flow rate on the outside of the bend. This represents half the cross-sectional area of ​​the pipe.

[0100] Calculate the absolute difference between the water flow rate on the outside and the water flow rate on the inside of the elbow to obtain the water flow rate difference between the inside and outside of the pipe elbow simulation unit; then calculate the ratio of the water flow rate difference to the real-time total flow rate to obtain a dimensionless flow rate difference coefficient; for example, the specific calculation formula is as follows: , ,in, Due to the difference in water flow, This is the flow difference coefficient;

[0101] The precipitation determination value is obtained by multiplying the comprehensive water body scaling saturation index with the flow difference coefficient; for example, the specific calculation formula is as follows: ,in, This is the precipitate determination value. The comprehensive water body scaling saturation index calculated in step S10;

[0102] The precipitation determination value is compared with the preset actual precipitation threshold.

[0103] If the precipitation determination value is greater than the actual precipitation threshold, then it is determined that the actual precipitation condition is met under the current water flow difference.

[0104] If the precipitation determination value is less than or equal to the actual precipitation threshold, then the actual precipitation condition is not met.

[0105] It should be noted that the flow field offset coefficient in the formula It is a dimensionless empirical constant used to characterize the degree to which the flow velocity center shifts outward due to centrifugal force when fluid flows through a bend; the flow field shift coefficient is usually experimentally calibrated based on the Reynolds number of the pipe and the absolute roughness of the inner wall, and generally ranges from 0.5 to 1.5; the actual precipitation threshold is a comprehensive judgment limit (e.g., set at 0.15); when the precipitation judgment value... When this threshold is exceeded, it means that the water body is not only thermodynamically supersaturated ( ), and the flow field distortion in the local area of ​​the bend ( The retention effect provided is strong enough that tiny crystal particles will overcome the shear suspension force of the fluid and substantially adhere and deposit in the low-velocity zone inside the bend.

[0106] Those skilled in the art should understand that the actual precipitation threshold is a comprehensive judgment limit (e.g., set to 0.15). When the precipitation judgment value (i.e., the product of the water body scaling saturation index and the flow difference coefficient) exceeds this threshold, it means that the water body is not only thermodynamically supersaturated, but also that the retention effect provided by the local flow field distortion (flow difference) of the bend is strong enough that the tiny crystal particles will overcome the suspension force and substantially adhere and deposit in the low flow velocity zone inside the bend.

[0107] Step S30: If the conditions for substantial precipitation are met, then for the simulation unit of the pipe bend, analyze the linear relationship between the water flow difference sample and the corresponding actual precipitation amount, and construct a coupled model of water flow difference-substantial precipitation amount.

[0108] In some embodiments, in step S30, if the conditions for substantial precipitation are met, the specific process for analyzing the linear correlation between the water flow difference sample and the corresponding actual precipitation amount for the pipe bend simulation unit is as follows:

[0109] Obtain multiple historical operating time periods of the pipe elbow simulation unit within the historical operating cycle of the heating system;

[0110] Extract the historical water flow difference calculated within each of the historical operating time periods as a water flow difference sample;

[0111] Retrieve historical maintenance and cleaning records or historical pipe wall thickness measurement data corresponding to each historical operating time period, extract or convert the corresponding actual sedimentation amount, and use it as an actual sedimentation amount sample.

[0112] By matching the water flow difference samples within the same historical operating period with the actual sedimentation samples, multiple initial data sample pairs are constructed.

[0113] Outlier detection and removal are performed on the multiple initial data sample pairs to obtain an effective sample set;

[0114] Based on the effective sample set, calculate the Pearson correlation coefficient between the water flow difference sample and the actual sedimentation sample;

[0115] The absolute value of the Pearson correlation coefficient is compared with a preset linear correlation threshold.

[0116] If the absolute value of the Pearson correlation coefficient is greater than or equal to the linear correlation threshold, then it is determined that there is a significant linear correlation between the water flow difference sample and the actual sedimentation amount.

[0117] If the absolute value of the Pearson correlation coefficient is less than the linear correlation threshold, it is determined that there is no significant linear correlation between the water flow difference sample and the actual sedimentation amount.

[0118] In some embodiments, the specific process of analyzing the linear correlation between the water flow difference sample and the corresponding actual sedimentation amount in step S30, and constructing the water flow difference-actual sedimentation amount coupling model, is as follows:

[0119] In the presence of a significant linear correlation, the effective sample set is fitted with a univariate linear regression using the least squares method, with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable, and the linear regression coefficient and intercept are calculated.

[0120] Based on the linear regression coefficients and the intercept, a coupled model of water flow difference-actual sedimentation is constructed. The expression of the coupled model is: Actual sedimentation = Linear regression coefficient × Water flow difference + Intercept;

[0121] In the absence of a significant linear correlation, the effective sample set is fitted with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable using a multinomial nonlinear fitting method to obtain the nonlinear fitting coefficients and constant term. Based on the nonlinear fitting coefficients, a nonlinear coupling model of water flow difference and actual sedimentation is constructed to establish a high-order mapping relationship between actual sedimentation and water flow difference, thereby achieving accurate fitting of the non-monotonic and nonlinear sedimentation evolution law.

[0122] It should be noted that the actual sediment volume sample can be obtained by weighing the scale removed during historical downtime maintenance, or by multiplying the scale thickness measured by an ultrasonic thickness gauge by the sediment density and the inner surface area of ​​the elbow.

[0123] Those skilled in the art should understand that the Pearson correlation coefficient is used to measure the strength of the linear correlation between two variables, and its value ranges from -1 to 1; the linear correlation threshold is usually set to 0.75; when the absolute value of the Pearson correlation coefficient is greater than 0.75, it is statistically considered that the two have a strong linear correlation, that is, the greater the difference in water flow, the stronger the local flow field retention effect, resulting in a significant linear increase in the actual sedimentation amount;

[0124] The linear coupling model constructed using the least squares method can quickly and directly establish the mathematical mapping relationship between macroscopic fluid parameters (water flow difference) and microscopic scaling results (actual sedimentation amount) without relying on complex multiphysics field coupling simulation, thus greatly reducing the real-time computing power consumption of the digital twin system.

[0125] The nonlinear coupling model constructed by the nonlinear fitting method can accurately characterize the nonlinear mapping relationship between macroscopic fluid parameters (water flow difference) and microscopic scaling results (actual sedimentation amount) in scenarios where fluid scaling exhibits non-monotonic and complex evolution patterns. It effectively makes up for the shortcomings of linear models in terms of insufficient fitting accuracy and limited adaptability to a single scenario, and significantly improves the accuracy and generalization ability of predicting local sedimentation at pipe bends.

[0126] Step S40: Input the currently calculated water flow difference into the water flow difference-actual sedimentation coupling model, calculate the actual sedimentation at the current pipeline bend simulation unit location, and when the actual sedimentation exceeds the safety threshold, determine that actual sedimentation has occurred and generate a cleaning and maintenance instruction for the current pipeline bend simulation unit location.

[0127] In some embodiments, the specific process of inputting the currently calculated water flow difference into the water flow difference-actual sedimentation amount coupling model in step S40 to calculate the actual sedimentation amount at the current pipeline bend simulation unit location is as follows:

[0128] Obtain the water flow difference calculated in real time within the current operating cycle;

[0129] Substitute the water flow difference into the water flow difference-actual sedimentation amount coupling model, and calculate the predicted actual sedimentation amount at the current location of the pipe bend simulation unit using the linear regression coefficient and the intercept.

[0130] In some embodiments, the specific process of determining that substantial sedimentation has occurred and generating a cleaning and maintenance instruction for the current pipeline bend simulation unit location when the actual sedimentation amount exceeds the safety threshold in step S40 is as follows:

[0131] Obtain a pre-set safety threshold, which represents the maximum allowable amount of sediment adhesion at the pipe bend without affecting the normal hydraulic operation of the heating system;

[0132] The predicted amount of actual sedimentation is compared numerically with the safety threshold.

[0133] If the predicted actual sedimentation amount is greater than the safety threshold, it is determined that actual sedimentation has occurred at the current location of the pipeline bend simulation unit, and the degree of sedimentation has constituted a local operational hazard.

[0134] If substantial sedimentation is determined to have occurred, the three-dimensional spatial coordinates and unique equipment identification code of the current pipe elbow simulation unit are extracted from the digital twin model of the heating system.

[0135] Based on the three-dimensional spatial coordinates, the unique identification code of the equipment, and the predicted actual sedimentation amount, a cleaning and maintenance instruction is generated for the current location of the pipeline bend simulation unit.

[0136] The cleaning and maintenance instructions include at least three-dimensional spatial coordinates, the unique identifier of the equipment, and the predicted actual amount of sediment.

[0137] The cleaning and maintenance instructions are sent to the operation and maintenance management terminal of the heating system to prompt the operation and maintenance personnel to perform targeted cleaning.

[0138] It should be noted that the safety threshold is set comprehensively based on the pipe diameter of the pipe bend and the design flow rate requirements of the heating system. For example, the safety threshold can be set as the mass or thickness of the sediment that causes a 5% reduction in the local flow cross-sectional area of ​​the bend. When the predicted actual sediment amount exceeds this safety threshold, although the global differential pressure and total flow of the heating system have not yet triggered a macro alarm, the flow resistance in the local area of ​​the bend has begun to increase significantly, posing a risk of under-deposit corrosion or further rapid scaling.

[0139] By generating cleaning and maintenance instructions containing specific three-dimensional spatial coordinates and unique equipment identification codes, it can guide maintenance personnel to accurately locate specific bends with potential hazards before global parameters deteriorate, and perform local pulse flushing or targeted chemical cleaning. This avoids blind large-scale shutdown cleaning in traditional maintenance and achieves true early prediction and refined predictive maintenance.

[0140] Example 2

[0141] like Figure 2 As shown, based on the specific implementation process of Embodiment 1, the present invention provides a predictive maintenance system for the entire life cycle of a heating system based on digital twins, including the following modules:

[0142] Modeling Risk Assessment Module: Construct a digital twin model of the heating system, mark the position of the simulation unit of pipe bend in the digital twin model, obtain the real-time water ion concentration parameters of the heating system, and calculate the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial precipitation risk.

[0143] Elbow flow field discrimination module: If there is an initial sedimentation risk, the local flow field data inside the pipe elbow simulation unit is obtained in the digital twin model, the water flow difference between the inside and outside of the pipe elbow simulation unit is calculated, and the water scaling saturation index is combined to analyze whether the conditions for substantial sedimentation are met under the current water flow difference.

[0144] Coupled model training module: If the condition for substantial precipitation is met, then for the simulation unit of the pipe bend, analyze the linear relationship between the water flow difference sample and the corresponding actual precipitation amount, and construct a coupled model of water flow difference-substantial precipitation amount.

[0145] Predictive maintenance instruction module: Input the currently calculated water flow difference into the water flow difference-actual sedimentation amount coupling model, calculate the actual sedimentation amount at the current pipeline bend simulation unit location, and when the actual sedimentation amount exceeds the safety threshold, determine that actual sedimentation has occurred and generate cleaning and maintenance instructions for the current pipeline bend simulation unit location.

[0146] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A predictive maintenance method for the entire lifecycle of a heating system based on digital twins, characterized by: include: Step S10: Construct a digital twin model of the heating system, mark the position of the simulation unit of the pipe bend in the digital twin model, obtain the real-time water ion concentration parameters of the heating system, and calculate the water scaling saturation index based on the water ion concentration parameters and solubility product constant to determine whether there is an initial precipitation risk. Step S20: If there is an initial risk of precipitation, in the digital twin model, obtain the local flow field data inside the pipe elbow simulation unit, calculate the water flow difference between the inside and outside of the pipe elbow simulation unit, and combine the water scaling saturation index to analyze whether the conditions for substantial precipitation are met under the current water flow difference. Step S30: If the conditions for substantial precipitation are met, then for the simulation unit of the pipe bend, analyze the linear relationship between the water flow difference sample and the corresponding actual precipitation amount, and construct a coupled model of water flow difference-substantial precipitation amount. Step S40: Input the currently calculated water flow difference into the water flow difference-actual sedimentation coupling model, calculate the actual sedimentation at the current pipeline bend simulation unit location, and when the actual sedimentation exceeds the safety threshold, determine that actual sedimentation has occurred and generate a cleaning and maintenance instruction for the current pipeline bend simulation unit location.

2. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The construction process of the digital twin model is as follows: Obtain data on the physical network topology of the heating system, pipe material parameters, and equipment operating parameters; A digital twin model is constructed using 3D modeling software based on physical pipeline network topology data, pipeline material parameters, and equipment operating parameters.

3. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process of marking the location of the simulation unit of the pipe bend in the digital twin model is as follows: In the digital twin model of the heating system, all pipe segments where the fluid flow direction changes are identified, the pipe segments are marked as pipe elbow simulation units, and the three-dimensional spatial coordinates of each pipe elbow simulation unit are recorded.

4. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for determining whether there is an initial risk of sedimentation is as follows: Obtain the ion concentration parameters of the water in the heating system, including calcium ion concentration, magnesium ion concentration, carbonate ion concentration and sulfate ion concentration; Obtain the solubility product constant of the target precipitate at the current water temperature. The target precipitate includes calcium carbonate, calcium sulfate, and magnesium carbonate. Based on the ion concentration parameters of the water body, calculate the actual ion product of the corresponding ions of each target precipitate in the water body; The ratio of each actual ion product to the corresponding solubility product constant is calculated to obtain the scaling saturation index of the water body corresponding to each target precipitate. The maximum value of the scaling saturation index of the water body corresponding to each target precipitate is extracted as the comprehensive scaling saturation index of the water body; The comprehensive water body scaling saturation index was compared with the initial sedimentation risk threshold; If the overall water body scaling saturation index is greater than the initial sedimentation risk threshold, then an initial sedimentation risk is determined to exist. If the overall water body scaling saturation index is less than or equal to the initial sedimentation risk threshold, then it is determined that there is no initial sedimentation risk.

5. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for analyzing whether the conditions for substantial precipitation are met under the current water flow difference is as follows: Obtain the current real-time total flow of the heating system, as well as the geometric parameters of the pipe bend simulation unit, including the pipe inner diameter and the bend curvature radius; Calculate the average flow velocity of the fluid in the straight pipe section based on the real-time total flow rate and the pipe inner diameter; The ratio of the pipe's inner diameter to the elbow's radius of curvature is calculated to obtain the elbow curvature ratio. Obtain the flow field offset coefficient, and calculate the equivalent average flow velocity of the inner half-section and the equivalent average flow velocity of the outer half-section of the bend based on the average flow velocity, the bend curvature ratio and the flow field offset coefficient. Multiply the equivalent average velocity of the inner half-section of the elbow by the half-section area of ​​the pipe to obtain the water flow rate inside the elbow; multiply the equivalent average velocity of the outer half-section of the elbow by the half-section area of ​​the pipe to obtain the water flow rate outside the elbow. The absolute difference between the water flow rate on the outside of the elbow and the water flow rate on the inside of the elbow is calculated to obtain the water flow rate difference between the inside and outside of the pipe elbow simulation unit; and the ratio of the water flow rate difference to the real-time total flow rate is calculated to obtain the dimensionless flow rate difference coefficient. The precipitation judgment value is obtained by multiplying the comprehensive water body scaling saturation index with the flow difference coefficient. If the precipitation judgment value is greater than the actual precipitation threshold, then the actual precipitation condition is met under the current water flow difference. If the precipitation determination value is less than or equal to the actual precipitation threshold, then the actual precipitation condition is not met.

6. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for analyzing the linear correlation between the water flow difference sample and the corresponding actual sedimentation amount is as follows: Obtain multiple historical operating time periods of the pipe bend simulation unit within the historical operating cycle of the heating system; Extract the historical water flow difference calculated within each historical operating period as a water flow difference sample; Retrieve the actual sedimentation amount corresponding to each historical operating period as a sample of actual sedimentation amount; Multiple initial data sample pairs are constructed by matching the water flow difference samples within the same historical operating period with the actual sedimentation samples one by one. Outlier detection and removal are performed on multiple initial data sample pairs to obtain an effective sample set; Based on the effective sample set, the Pearson correlation coefficient between the water flow difference sample and the actual sedimentation sample was calculated. If the absolute value of the Pearson correlation coefficient is greater than or equal to the linear correlation threshold, then it is determined that there is a significant linear correlation between the water flow difference sample and the actual sedimentation amount. If the absolute value of the Pearson correlation coefficient is less than the linear correlation threshold, it is determined that there is no significant linear correlation between the water flow difference sample and the actual sedimentation amount.

7. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for constructing the coupled model of water body flow difference and actual sedimentation amount is as follows: In the presence of a significant linear correlation, the least squares method was used to perform a univariate linear regression fitting on the effective sample set with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable, and the linear regression coefficient and intercept were calculated. Based on the linear regression coefficients and the intercept, a coupled model of water flow difference and actual sedimentation is constructed. The expression of the coupled model is: Actual sedimentation = Linear regression coefficient × Water flow difference + Intercept.

8. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process of constructing the coupled model of water body flow difference and actual sedimentation amount also includes: In the absence of a significant linear correlation, a nonlinear coupling model of water flow difference and actual sedimentation is constructed by using a multinomial nonlinear fitting method to regress the effective sample set with the water flow difference sample as the independent variable and the actual sedimentation sample as the dependent variable.

9. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for calculating the actual sedimentation amount at the current pipeline bend simulation unit location is as follows: The water flow difference calculated in real time is obtained, and the water flow difference is substituted into the water flow difference-actual sedimentation amount coupling model to obtain the predicted actual sedimentation amount at the current pipeline bend simulation unit position.

10. The predictive maintenance method for the entire lifecycle of a heating system based on digital twins according to claim 1, characterized in that, The specific process for generating cleaning and maintenance instructions for the current pipeline bend simulation unit location is as follows: If the predicted actual sedimentation amount is greater than the safety threshold, then it is determined that actual sedimentation has occurred at the current pipeline elbow simulation unit location; If substantial sedimentation is determined to have occurred, the three-dimensional spatial coordinates and unique equipment identification code of the current pipe bend simulation unit are extracted from the digital twin model of the heating system. Based on three-dimensional spatial coordinates, unique equipment identification code, and predicted actual sedimentation amount, a cleaning and maintenance instruction is generated for the current simulated unit location of the pipe bend. The cleaning and maintenance instructions include three-dimensional spatial coordinates, the equipment's unique identification code, and the predicted amount of actual sediment.

Images