A method for dynamic regulation and control of concrete curing process based on unsteady heat and moisture transfer

Abstract

The application discloses a kind of based on unsteady heat and humidity transport concrete curing process dynamic regulation and control method, belong to concrete intelligent curing technical field, this method includes based on the proportioning of concrete member, the thermal and humidity constitutive model parameter of concrete is determined;Based on the thermal and humidity constitutive model parameter of concrete establishes unsteady heat and humidity coupling transmission control model;Obtain the initial state when the concrete member to be cured starts curing;According to thermal and humidity constitutive model parameter and initial state, initialize the digital twin model of concrete member to be cured;Using improved ensemble Kalman filtering is inverted, and the digital twin model is updated according to the mean value of inversion result;Based on the state of updated digital twin model, construct MPC constraint predictive control problem to solve optimal curing action sequence.The application solves the problem that the traditional curing adopts fixed system of fixed time and fixed quantity, cannot respond to the space-time evolution of material internal heat and humidity state, leads to temperature peak out of control, internal and external temperature difference is too large, surface loses water too fast.

Description

Technical Field

[0001] This invention belongs to the field of intelligent concrete curing technology, and particularly relates to a dynamic control method for the concrete curing process based on unsteady heat and moisture transfer. Background Technology

[0002] After concrete is poured, its mechanical and durability properties develop highly dependent on early curing conditions. The core objective of curing is to maintain a suitable temperature and humidity environment to ensure sufficient hydration of the cementitious materials while preventing cracks caused by temperature stress or drying shrinkage. Especially in large-volume concrete structures (such as dams, foundation slabs, and bridge piers) and thin-walled structures (such as wall panels and shells), the quality of curing directly determines the structure's load-bearing capacity, impermeability, and service life.

[0003] Traditional curing methods employ fixed schedules and quantities, such as "covering with insulation blankets for 7 days after pouring, and watering 3 times a day." The fundamental flaw of this method is that the curing parameters (insulation thickness, watering frequency) are preset and completely fail to respond to actual changes in the internal temperature and humidity of the concrete. When encountering extreme weather (such as cold waves, hot and dry winds) or fluctuations in the concrete mix proportions, the fixed schedule cannot adaptively adjust, easily leading to uncontrolled temperature peaks, excessive internal and external temperature differences, or excessively rapid surface water loss. Furthermore, current control decisions heavily rely on the experience of on-site technicians. Determining "when to strengthen insulation" or "when to increase watering" lacks quantitative basis. This experience-based approach significantly reduces the reliability of decisions in scenarios involving large-volume concrete, high-strength concrete, and complex climatic conditions, resulting in large quality fluctuations.

[0004] Furthermore, traditional methods typically only take remedial measures after cracks appear or temperatures exceed limits, essentially a passive, reactive response. They lack the ability to predict future thermal and humidity conditions, making preventative control impossible before risks materialize. Moreover, existing curing techniques often treat temperature and humidity control separately, ignoring the strong physical coupling between the two: hydration exothermics drive the temperature field, temperature gradients drive moisture migration, and surface evaporation simultaneously removes heat and moisture—a complex interaction. Ignoring this coupling often leads to incomplete control strategies. Many curing schemes lack comprehensive determination of the thermal and humidity constitutive parameters for the concrete mix design; temperature stress calculations and crack risk assessments rely on empirical formulas, resulting in insufficient accuracy. Sensor monitoring, simulation analysis, and control decisions are often independent, failing to integrate data in real time to form a closed loop. Even when sensors are deployed, the data is only used for recording rather than driving active control. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this invention provides a dynamic control method for concrete curing based on unsteady heat and moisture transfer. This method solves the problems of traditional curing methods, which use a fixed time and quantity system, failing to respond to the spatiotemporal evolution of the internal heat and moisture state of the material, leading to uncontrolled temperature peaks, excessive internal and external temperature differences, and excessively rapid surface water loss.

[0006] To achieve the aforementioned objectives, the technical solution adopted by this invention is: a dynamic control method for the concrete curing process based on unsteady-state heat and moisture transfer, comprising: Determine the mix proportions of the concrete components to be cured, and based on the mix proportions, determine the thermal-wet constitutive model parameters of the concrete in the laboratory. An unsteady-state thermal-moisture coupling transport control model is established based on the parameters of the thermal-moisture constitutive model of concrete; the unsteady-state thermal-moisture coupling transport control model is determined based on the energy conservation equation, the mass conservation equation, the thermal boundary, and the moisture boundary; During the actual construction, temperature and humidity sensors are pre-embedded in the concrete components to be cured, and miniature weather stations are set up in the curing environment to collect environmental data; the environmental data includes wind speed, ambient temperature, ambient relative humidity and solar radiation intensity. The initial state of the concrete component to be cured is obtained based on temperature sensors, humidity sensors and a micro weather station; a digital twin model of the concrete component to be cured is initialized based on the parameters of the thermo-humid constitutive model and the initial state. At the current time step: real-time observation data is acquired based on temperature and humidity sensors, and an improved ensemble Kalman filter is used for inversion based on the real-time observation data. The digital twin model is then updated according to the mean of the inversion results of the ensemble members. The state vector of the improved ensemble Kalman filter includes the temperature, humidity, current isothermal moisture diffusivity, and thermal conductivity of each monitoring point. Specifically, the improved ensemble Kalman filter introduces a gain attenuation coefficient to control the gain during the posterior update. Based on the updated digital twin model of the state, we construct an MPC-constrained predictive control problem to find the optimal maintenance action sequence that minimizes the cost function over the next P control steps.

[0007] Furthermore, the thermal-wet constitutive model parameters of the concrete include hydration heat characteristic parameters, thermophysical parameters, moisture transport parameters, and boundary characteristic parameters; the determination of the thermal-wet constitutive model parameters of the concrete specifically involves: For cement paste samples with the same mix ratio, heat flow was continuously measured for several hours using an isothermal microcalorimeter under different temperature conditions, yielding a number of heat flow data. Based on the heat flow data under different temperature conditions, a kinetic equation was fitted using the global least squares method with the objective of minimizing the sum of squared errors in the degree of hydration prediction at each time point under each temperature condition, thus obtaining the hydration heat characteristic parameters. The kinetic equation incorporates a correction term to compensate for the nonlinear coupling effect of volume shrinkage and CSH densification in the later stages of hydration, based on existing reaction kinetic equations.

[0008]

[0009] in, The differential symbol; Degree of hydration; For time; It is the apparent activation energy; This is the universal gas constant; For reference temperature; The current temperature; It is the rate constant; The mechanism control index for the acceleration phase; The mechanism control index during the deceleration period; This is a correction factor for high hydration degree; This represents the theoretical heat release upon 100% hydration. This refers to the final degree of hydration, i.e., the actual maximum achievable degree of hydration. The heat release density of hydration; The density of concrete; Concrete specimens of different sizes but with the same mix proportions were cast, and thermophysical parameters were determined based on each specimen. The density of the concrete was measured using the drainage method. The thermal conductivity of each concrete specimen was measured using the transient plane heat source method, and the specific heat capacity of powder from concrete specimens of different ages at different temperatures was measured using differential scanning calorimetry. The moisture content, temperature, and degree of hydration at each measuring point were recorded. Based on the thermal conductivity, specific heat capacity, moisture content, temperature, and degree of hydration at each measuring point, thermal conductivity models and specific heat capacity models were fitted respectively.

[0010]

[0011] Thermal conductivity; The initial thermal conductivity; This represents the current moisture content. This represents the initial moisture content. The moisture content influence index; The hydration enhancement coefficient; Specific heat capacity; Based on specific heat capacity; This is a temperature correction factor; This is the hydration correction factor; Based on the moisture characteristic curves, isothermal moisture diffusion coefficient data, and thermal-moisture diffusion coefficient data of concrete specimens with the same mix proportions, moisture transport parameters were determined: Moisture characteristic curve models, isothermal moisture diffusion coefficient models, non-isothermal diffusion coefficient models, and chemically bound water terms were fitted based on the moisture characteristic curves, isothermal moisture diffusion coefficient data, and non-isothermal diffusion coefficient data.

[0012]

[0013]

[0014]

[0015] in, Effective saturation; Residual moisture content; This represents the saturated moisture content. These are fitting parameters related to the intake air value; For capillary pressure head; The fitting parameters for the pore size distribution width of the reaction; This represents the current isothermal moisture diffusion coefficient; The reference diffusion coefficient; The saturation effect index; The hydration resistance coefficient; This is the current non-isothermal diffusion coefficient; It is the thermal diffusion coupling factor; This refers to the amount of chemically bound water. The mass of bound water corresponding to a unit increase in hydration degree; In a wind tunnel, the relationship between the convective heat transfer coefficient and wind speed on the exposed concrete surface was determined, and the thickness and thermal conductivity of each curing covering material were obtained. Boundary characteristic parameters were measured: Based on the relationship between the convective heat transfer coefficient and wind speed, and the thickness and thermal conductivity of each curing covering material, the equivalent heat transfer coefficient with covering was calibrated; based on the relationship between the convective heat transfer coefficient and wind speed, the convective mass transfer coefficient was calibrated using the heat-mass analogy method; and based on the convective mass transfer coefficient, the surface evaporation flux was calibrated.

[0016]

[0017]

[0018]

[0019]

[0020] in, The convective heat transfer coefficient; The surface natural convection heat transfer coefficient; This refers to the wind speed coefficient; Wind speed; This refers to the wind speed coefficient; This is the equivalent heat transfer coefficient when there is a covering. For the first The thickness of the covering material; For the first Thermal conductivity of the covering material; The number of covering layers; For interlayer air thermal resistance; The convective mass transfer coefficient; air density; The specific heat capacity of air; For Lewis numbers; For surface evaporation flux; The relative humidity of the concrete surface; The density is the saturated water vapor density. The concrete surface temperature; The relative humidity of the environment; The ambient temperature; The molar mass of water; The density of liquid water; It is capillary pressure.

[0021] Furthermore, the expression for the energy conservation equation is:

[0022] in, The density of concrete; For temperature hydration degree is Specific heat capacity at time; The current temperature; For time; For gradient operators; The moisture content is hydration degree is Thermal conductivity at that time; For temperature gradient; For temperature hydration degree is The hydration heat release power density at that time.

[0023] Furthermore, the expression for the mass conservation equation is:

[0024] in, This represents the current moisture content. For time; For gradient operators; The moisture content is hydration degree is The current isothermal moisture diffusivity at that time; For moisture content gradient; For temperature The moisture content is The current non-isothermal diffusion coefficient at that time; For temperature gradient; This refers to the amount of chemically bound water.

[0025] Furthermore, the expressions for the thermal boundary and the wet boundary are:

[0026] in, Thermal conductivity; The current temperature; The direction of the unit outward normal to the concrete surface; This is the equivalent heat transfer coefficient when there is a covering. The concrete surface temperature; The ambient temperature; Solar radiation absorptivity; Solar radiation intensity; The latent heat of vaporization of water; For surface evaporation flux; For concrete surfaces; The expression for the wet boundary during normal evaporation is:

[0027] in, This represents the current isothermal moisture diffusion coefficient; This represents the current moisture content. When controlling the spray system, a constant boundary value is used for the thermal boundary:

[0028] in, This refers to the surface moisture content of the concrete. This represents the saturated moisture content.

[0029] Furthermore, the expression for the posterior update of the improved ensemble Kalman filter is:

[0030]

[0031]

[0032] in, For the first The posterior update result of the inversion of a set of perturbed members; The first time step generated based on the best state estimate of the previous time step The predicted state of a set member with perturbation at the current time step; This is the gain attenuation coefficient; The Kalman gain matrix; For real-time observation by sensors; The observation matrix; For the observation noise simulation term; It is a non-linearity index; This is the critical threshold for nonlinearity. This is an unsteady thermal-humidity coupled transport control model; The actual covariance of the set members after propagation through the unsteady thermal-humid coupling transmission control model; Covariance based on linearized approximation; It is the F-norm.

[0033] Furthermore, the cost function of the MPC-constrained predictive control problem is:

[0034] in, It is the cost function; For control step index; To control the total number of steps; for The temperature difference between inside and outside at any given moment; The target temperature difference; This is the weight matrix for the temperature difference term; for The relative humidity of the concrete surface at any given time; Minimum permissible surface relative humidity; This is the weight matrix for the humidity term; In order to be in The control vector applied at each time step; This is the weight matrix for the energy consumption term.

[0035] Furthermore, the control vector includes adjustable data of the concrete curing environment, including the thickness of the covering, the cooling water flow rate of the pre-embedded cooling water pipes, the cooling water temperature, the start and stop of the spray system, and the spray water temperature.

[0036] The beneficial effects of this invention are as follows: A correction term is introduced into the kinetic equation to compensate for the nonlinear coupling effect of volume shrinkage and CSH densification in the later stages of hydration. Traditional single-parameter models (such as fixed exponents) cannot accurately describe both acceleration and deceleration stages simultaneously. Two exponents are used to "decouple" the control mechanisms of different stages. Based on traditional ensemble Kalman filtering, a "nonlinearity index" is introduced to evaluate the degree of deviation from linearity in real time. When entering a strongly nonlinear region (such as near temperature peaks), the Kalman gain is automatically attenuated and the ensemble dispersion is expanded. During strongly nonlinear moments such as temperature peaks and sharp humidity drops, the algorithm avoids collapse due to overfitting to lagging observations. Thermal conductivity and moisture diffusion coefficient are used as state variables in the assimilation process, calibrating the model's internal parameters in real time and overcoming the limitation of traditional filtering that only updates the state. No manual trial and error of covariance parameters is required; the algorithm automatically adjusts the trust allocation based on the intensity of the physical process. Attached Figure Description

[0037] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0038] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0039] like Figure 1 As shown, in one embodiment of the present invention, a method for dynamic control of concrete curing process based on unsteady-state heat and moisture transfer includes: Determine the mix proportions of the concrete components to be cured, and based on the mix proportions, determine the thermal-wet constitutive model parameters of the concrete in the laboratory. An unsteady-state thermal-moisture coupling transport control model is established based on the parameters of the thermal-moisture constitutive model of concrete; the unsteady-state thermal-moisture coupling transport control model is determined based on the energy conservation equation, the mass conservation equation, the thermal boundary, and the moisture boundary; During the actual construction, temperature and humidity sensors are pre-embedded in the concrete components to be cured, and miniature weather stations are set up in the curing environment to collect environmental data; the environmental data includes wind speed, ambient temperature, ambient relative humidity and solar radiation intensity. The initial state of the concrete component to be cured is obtained based on temperature sensors, humidity sensors and a micro weather station; a digital twin model of the concrete component to be cured is initialized based on the parameters of the thermo-humid constitutive model and the initial state. At the current time step: real-time observation data is acquired based on temperature and humidity sensors, and an improved ensemble Kalman filter is used for inversion based on the real-time observation data. The digital twin model is then updated according to the mean of the inversion results of the ensemble members. The state vector of the improved ensemble Kalman filter includes the temperature, humidity, current isothermal moisture diffusivity, and thermal conductivity of each monitoring point. Specifically, the improved ensemble Kalman filter introduces a gain attenuation coefficient to control the gain during the posterior update. Based on the updated digital twin model of the state, we construct an MPC-constrained predictive control problem to find the optimal maintenance action sequence that minimizes the cost function over the next P control steps.

[0040] The thermal-wet constitutive model parameters of the concrete include hydration heat characteristic parameters, thermophysical parameters, moisture transport parameters, and boundary characteristic parameters; the determination of the thermal-wet constitutive model parameters of the concrete specifically includes: 1. Calibrate the hydration heat characteristic parameters.

[0041] Experimental Procedure: Cement paste samples were prepared according to the actual construction mix proportions. Using an isothermal micro calorimeter, heat flux was continuously measured at four temperature levels (approximately 23°C, 35°C, 45°C, and 55°C) using a TAM Air isothermal micro calorimeter. At least 168 hours, with no fewer than 3 parallel samples at each temperature.

[0042] Take samples of cement paste from the same batch, seal and cure until fully hydrated, then grind finely, and determine the final degree of hydration using the chemical binding water method. Dry at 105℃ until constant weight, then weigh (m). 105 Heat to constant weight at 950℃ and weigh (m). 950 Loss on ignition (LOI) = (m 105 m 950 ) / m 105 LOI=(m 105 m950) / m 105 Final degree of hydration =LOI / LOI 理论 LOI 理论 Calculated from the hydration reaction equations of each cementitious material component.

[0043] Data processing and parameter calibration: The heat flow data measured in each group were processed separately. Numerical integration was performed to obtain the cumulative heat release under different conditions. :

[0044] in, For temperature conditions Down, Real-time heat flow data; The theoretical maximum heat release is calculated based on the mass fraction of each cementitious material in the mix proportion and the heat of hydration. :

[0045] in, For the first Mass fraction of the gel material; For the first The heat release per unit mass of the gel material upon complete hydration (reference values: C3S≈517 J / g, C2S≈262 J / g, C3A≈1144 J / g, C4AF≈418 J / g).

[0046] Degree of hydration:

[0047] Using the global least squares method, the following kinetic equations were fitted with hydration degree data at four temperatures:

[0048] Obtain the apparent activation energy and dynamic parameters , , , .

[0049] in, The differential symbol; Degree of hydration; For time; The value represents temperature, which is a constant in isothermal experiments but a variable in field conditions. The apparent activation energy is expressed in J / mol. The universal gas constant is 8.314 J / mol·K. Reference temperature (296K); The mechanism control index for the acceleration phase; The mechanism control index during the deceleration period; The rate constant is 1 / s; This is a correction factor for high hydration degree; From the start of the reaction to At any moment, in temperature The cumulative heat release; This represents the theoretical heat release upon 100% hydration. This refers to the final degree of hydration, i.e., the actual maximum achievable degree of hydration. Double exponent ( , Independence: The weights of the "nucleation and growth" mechanisms during the accelerated phase are characterized. The weights of the "diffusion control" mechanism during the deceleration period are represented. The traditional single-parameter model is split into two independent indices to decouple the control mechanisms at different stages.

[0050] Correction item ( ): Compensates for the nonlinear coupling effect between "reaction front volume shrinkage" and "CSH densification" in the later stage of hydration, and solves the problem in the high hydration range ( The problem of predicted values ​​being too low (>0.7) Small amount (0.01~0.1).

[0051] Using the equivalent age framework The temperature effect is completely absorbed into the Arrhenius factor, parameter , , , It becomes a temperature-independent material constant, improving the ability to transfer models from the laboratory to the field.

[0052] Simultaneously, it outputs the hydration heat release power density, serving as an internal heat source for the energy conservation equation:

[0053] in, The heat release density of hydration; This refers to the density of concrete.

[0054] After parameter calibration, the variables in the curing process include descaling degree, temperature, time, and hydration heat release power density.

[0055] 2. Perform thermophysical parameter calibration.

[0056] Experimental Procedure: Concrete specimens of different sizes were poured and cured according to standard procedures. Specimens were removed at different curing ages (0.5 days, 1 day, 3 days, 7 days, and 28 days). Some specimens were tested immediately, while others were treated to different degrees of saturation (oven-dry, air-dry, and saturated) before testing. The density of the concrete was measured using the drainage method. Thermal conductivity was determined using the transient plane heat source method. Specific heat capacity was determined using differential scanning calorimetry. .

[0057] Data processing and parameter calibration:

[0058] in, Thermal conductivity, Given the current moisture content, The initial moisture content, The initial thermal conductivity is 1.8~2.5 W / (m·K). (0.3~0.6) is the moisture content influence index. (-0.05~0.15) represents the hydration enhancement coefficient.

[0059]

[0060] in, For specific heat capacity, Based on the specific heat capacity (850~1050 J / (kg·K)), This is a temperature correction factor (approximately 0.0005~0.001 K). -1 ), This is the hydration correction factor (approximately -0.05 to 0.1).

[0061] After parameter calibration, the variables during the curing process include specific heat capacity and thermal conductivity.

[0062] 3. Perform wet transmission parameter calibration.

[0063] Experimental procedures: A pressure plate apparatus (0~1.5MPa) was used for the high-pressure section, a pressure membrane apparatus (1.5~10MPa) for the medium-pressure section, and the saturated salt solution method (RH 75%~98%) for the low-saturation section to determine the equilibrium moisture content under different pressures and obtain moisture characteristic curves. Nuclear magnetic resonance (NMR) technology was used to monitor the internal moisture profile of the specimen during the one-dimensional drying process using CPMG pulse sequences, and the moisture diffusion coefficient was calculated. A constant temperature gradient was established in a sealed cylinder heated at one end and cooled at the other using a temperature gradient device. After stabilization, rapid slices were taken to measure the moisture content distribution, and the thermal and moisture diffusion coefficient was calculated under the condition of zero net moisture flux.

[0064] Data processing and parameter calibration:

[0065] in, Effective saturation; Residual moisture content; This represents the saturated moisture content. These are fitting parameters related to the intake air value; Capillary pressure head (m); The fitting parameters for the reaction pore size distribution width are denoted as .

[0066]

[0067] in, This represents the current isothermal moisture diffusion coefficient; The reference diffusion coefficient (m² / s) The saturation effect index, The hydration resistance coefficient is... This reflects the exponential enhancement effect of moisture content on the diffusion coefficient (m=3~8). This reflects the inhibitory effect of hydration-induced decrease in porosity and refinement of pore structure on water transport.

[0068]

[0069] in, This is the current non-isothermal diffusion coefficient; Thermal diffusion coupling factor (10) -3 ~10 -2 K -1 ), The Stefan-Boltzmann type response of water vapor saturation partial pressure at high temperature is modified.

[0070] Chemically bound water item:

[0071] in, This refers to the amount of chemically bound water. This represents the mass of bound water corresponding to a unit increase in hydration degree.

[0072] After parameter calibration, the variables during the curing process include moisture content, effective saturation, capillary head, capillary pressure, isothermal water diffusion coefficient, non-isothermal diffusion coefficient, and chemically bound water.

[0073] 4. Boundary characteristic parameter calibration.

[0074] Experimental procedure: Determine the convective heat transfer coefficient of exposed concrete surfaces by testing in a wind tunnel or referring to specifications. With wind speed The relationship between the materials and the curing covering (geotextile, insulation blanket, curing film) was determined. and thermal conductivity .

[0075] Data processing and parameter calibration:

[0076] Among them, the surface natural convection coefficient W / (m²·K), wind speed coefficient Wind speed coefficient , Wind speed (m / s).

[0077]

[0078] It is the equivalent heat transfer coefficient (W / (m²·K)). For the first The thickness of the covering layer (m), For the first The thermal conductivity of the covering material (W / (m·K)) The interlayer air thermal resistance is (m²·K / W).

[0079] Determining the convective mass transfer coefficient using heat and mass transfer analogy (Lewis relation) :

[0080] Among them, air density kg / m³, specific heat capacity of air J / (kg·K), Lewis number ≈0.85~1.25.

[0081] Calibrate surface evaporation flux based on convective mass transfer coefficient:

[0082]

[0083] in, The convective heat transfer coefficient; The surface natural convection heat transfer coefficient; This refers to the wind speed coefficient; Wind speed; This refers to the wind speed coefficient; This is the equivalent heat transfer coefficient when there is a covering. For the first The thickness of the covering material; For the first Thermal conductivity of the covering material; The number of covering layers; For interlayer air thermal resistance; The convective mass transfer coefficient; air density; The specific heat capacity of air; For Lewis numbers; For surface evaporation flux; The relative humidity of the concrete surface; The density is the saturated water vapor density. The concrete surface temperature; The relative humidity of the environment; The ambient temperature; The molar mass of water; It is watertight in liquid state; It is capillary pressure.

[0084] When events such as the laying of covering materials and the switching of formwork removal conditions occur on site, the equivalent heat transfer coefficient of the corresponding condition in the pre-built condition-parameter library is retrieved as the prior initial estimate for the improved ensemble Kalman filter. Subsequently, at each assimilation time step, the equivalent heat transfer coefficient is continuously corrected online through ensemble Kalman filtering, combined with on-site temperature, humidity, and meteorological observation data, to obtain a real-time equivalent heat transfer coefficient adapted to the current actual site environment. Among them, the benchmark parameters for the exposed concrete condition after formwork removal are taken from the wind tunnel calibration test results; the benchmark parameters for the condition with curing covering are taken from the experience dataset accumulated from historical projects of the same project or with the same mix proportion.

[0085] After parameter calibration, the variables during the curing process include ambient temperature, concrete surface temperature, wind speed, solar radiation intensity, ambient relative humidity, convective heat transfer coefficient, equivalent heat transfer coefficient, cover thickness, convective mass transfer coefficient, surface evaporation flux, surface relative humidity, saturated water vapor density, and saturated vapor pressure.

[0086] The expression for the energy conservation equation is:

[0087] in, The density of concrete; For temperature hydration degree is Specific heat capacity at time; The current temperature; For time; For gradient operators; The moisture content is hydration degree is Thermal conductivity at that time; For temperature gradient; For temperature hydration degree is The hydration heat release power density at that time.

[0088] The expression for the mass conservation equation is:

[0089] in, This represents the current moisture content. For time; For gradient operators; The moisture content is hydration degree is The current isothermal moisture diffusivity at that time; For moisture content gradient; For temperature The moisture content is The current non-isothermal diffusion coefficient at that time; For temperature gradient; This refers to the amount of chemically bound water.

[0090] The expressions for the thermal boundary and the wet boundary are:

[0091] in, Thermal conductivity; The current temperature; The direction of the unit outward normal to the concrete surface; This is the equivalent heat transfer coefficient when there is a covering. The concrete surface temperature; The ambient temperature; Solar radiation absorptivity; Solar radiation intensity; The latent heat of vaporization of water; For surface evaporation flux; For concrete surfaces; The expression for the wet boundary during normal evaporation is:

[0092] in, This represents the current isothermal moisture diffusion coefficient; The current moisture content is [value], and the volumetric moisture content is [value]. When controlling the spray system, a constant boundary value is used for the thermal boundary:

[0093] in, This refers to the surface moisture content of the concrete. This represents the saturated moisture content.

[0094] In this embodiment, the internal sensor deployment involves fixing digital temperature sensors and capacitive / fiber optic humidity sensors to the reinforcing steel frame or predetermined locations before concrete pouring. The sensor locations should include the structure center, different depths from the surface, variable cross-sections, and corners.

[0095] External weather station deployment: Install mini weather stations near the maintenance area to collect ambient temperature data in real time. relative humidity Wind speed and solar radiation intensity .

[0096] Data Acquisition and Transmission: Connect all sensors to the edge computing gateway, set them to collect data every 5-10 minutes, and upload the data to the cloud or local server via 4G / 5G / Wi-Fi.

[0097] Data output: Real-time temperature, humidity and environmental parameter sequences for each sensor location.

[0098] The expression for the posterior update of the improved ensemble Kalman filter is:

[0099]

[0100]

[0101] in, For the first The posterior update result of the inversion of a set of perturbed members; The first time step generated based on the best state estimate of the previous time step The predicted state of a set member with perturbation at the current time step; This is the gain attenuation coefficient; The Kalman gain matrix; For real-time observation by sensors; The observation matrix; For the observation noise simulation term; It is a non-linearity index; This is the critical threshold for nonlinearity. This is an unsteady thermal-humidity coupled transport control model; The actual covariance of the set members after propagation through the unsteady thermal-humid coupling transmission control model; Covariance based on linearized approximation; It is the F-norm.

[0102] In this embodiment, the state vector may also include moisture content, which can be obtained based on humidity data using the Kelvin equation and moisture characteristic curve.

[0103] The cost function of the MPC-constrained predictive control problem is:

[0104] in, It is the cost function; For control step index; To control the total number of steps; for The temperature difference between inside and outside at any given moment; The target temperature difference; This is the weight matrix for the temperature difference term; for The relative humidity of the concrete surface at any given time; Minimum permissible surface relative humidity; This is the weight matrix for the humidity term; In order to be in The control vector applied at each time step; This is the weight matrix for the energy consumption term.

[0105] Constraints for MPC-constrained predictive control problems may include preset upper limits for internal and external temperature difference, lower limits for surface humidity, upper and lower limits for cooling water temperature, maximum spray flow rate, upper and lower limits for cover material thickness, and extreme values ​​for equipment power.

[0106] The control vector includes adjustable data of the concrete curing environment, including the thickness of the covering, the flow rate of the pre-embedded cooling water pipe, the cooling water temperature, the start and stop of the spray and the spray water temperature (heating / cooling power and air supply temperature; equivalent thermal resistance of the insulation covering layer; atomized spray start and stop frequency, water volume, and water temperature; demolding time).

[0107] In this embodiment, after one prediction step is completed, the system waits for the next sensor sampling period (e.g., 10 minutes) and repeats the assimilation-prediction step. The new observation data is used for assimilation correction, eliminating the control bias and model prediction error from the previous round, thus forming continuous and stable closed-loop control.

Claims

1. A dynamic control method for the concrete curing process based on unsteady-state heat and moisture transfer, characterized in that, include: Determine the mix proportions of the concrete components to be cured, and based on the mix proportions, determine the thermal-wet constitutive model parameters of the concrete in the laboratory. An unsteady-state thermal-moisture coupling transport control model is established based on the parameters of the thermal-moisture constitutive model of concrete; the unsteady-state thermal-moisture coupling transport control model is determined based on the energy conservation equation, the mass conservation equation, the thermal boundary, and the moisture boundary; During the actual construction, temperature and humidity sensors are pre-embedded in the concrete components to be cured, and miniature weather stations are set up in the curing environment to collect environmental data; the environmental data includes wind speed, ambient temperature, ambient relative humidity and solar radiation intensity. The initial state of the concrete component to be cured is obtained based on temperature sensors, humidity sensors and a micro weather station; a digital twin model of the concrete component to be cured is initialized based on the parameters of the thermo-humid constitutive model and the initial state. At the current time step: real-time observation data is acquired based on temperature and humidity sensors, and an improved ensemble Kalman filter is used for inversion based on the real-time observation data. The digital twin model is then updated according to the mean of the inversion results of the ensemble members. The state vector of the improved ensemble Kalman filter includes the temperature, humidity, current isothermal moisture diffusivity, and thermal conductivity of each monitoring point. Specifically, the improved ensemble Kalman filter introduces a gain attenuation coefficient to control the gain during the posterior update. Based on the updated digital twin model of the state, we construct an MPC-constrained predictive control problem to find the optimal maintenance action sequence that minimizes the cost function over the next P control steps.

2. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The thermal-wet constitutive model parameters of the concrete include hydration heat characteristic parameters, thermophysical parameters, moisture transport parameters, and boundary characteristic parameters; the determination of the thermal-wet constitutive model parameters of the concrete specifically involves: For cement paste samples with the same mix ratio, heat flow was continuously measured for several hours using an isothermal microcalorimeter under different temperature conditions, yielding a number of heat flow data. Based on the heat flow data under different temperature conditions, a kinetic equation was fitted using the global least squares method with the objective of minimizing the sum of squared errors in the degree of hydration prediction at each time point under each temperature condition, thus obtaining the hydration heat characteristic parameters. The kinetic equation incorporates a correction term to compensate for the nonlinear coupling effect of volume shrinkage and CSH densification in the later stages of hydration, based on existing reaction kinetic equations. in, The differential symbol; Degree of hydration; For time; It is the apparent activation energy; This is the universal gas constant; For reference temperature; The current temperature; It is the rate constant; The mechanism control index for the acceleration phase; The mechanism control index during the deceleration period; This is a correction factor for high hydration degree; This represents the theoretical heat release upon 100% hydration. This refers to the final degree of hydration, i.e., the actual maximum achievable degree of hydration. The heat release density of hydration; The density of concrete; Concrete specimens of different sizes but with the same mix proportions were poured, and thermophysical parameters were determined based on each specimen. The density of the concrete was measured using the drainage method. The thermal conductivity of each concrete specimen was measured using the transient plane heat source method, and the specific heat capacity of powder from concrete specimens of different ages at different temperatures was measured using differential scanning calorimetry. The moisture content, temperature, and degree of hydration at each measuring point were recorded. Based on the thermal conductivity, specific heat capacity, moisture content, temperature, and degree of hydration at each measuring point, thermal conductivity models and specific heat capacity models were fitted respectively. Thermal conductivity; The initial thermal conductivity; This represents the current moisture content. This represents the initial moisture content. The moisture content influence index; The hydration enhancement coefficient; Specific heat capacity; Based on specific heat capacity; This is a temperature correction factor; This is the hydration correction factor; Based on the moisture characteristic curves, isothermal moisture diffusion coefficient data, and thermal-moisture diffusion coefficient data of concrete specimens with the same mix proportions, moisture transport parameters were determined: Moisture characteristic curve models, isothermal moisture diffusion coefficient models, non-isothermal diffusion coefficient models, and chemically bound water terms were fitted based on the moisture characteristic curves, isothermal moisture diffusion coefficient data, and non-isothermal diffusion coefficient data. in, Effective saturation; Residual moisture content; This represents the saturated moisture content. These are fitting parameters related to the intake air value; For capillary pressure head; The fitting parameters for the pore size distribution width of the reaction; This represents the current isothermal moisture diffusion coefficient; The reference diffusion coefficient; The saturation effect index; The hydration resistance coefficient; This is the current non-isothermal diffusion coefficient; It is the thermal diffusion coupling factor; This refers to the amount of chemically bound water. The mass of bound water corresponding to a unit increase in hydration degree; In a wind tunnel, the relationship between the convective heat transfer coefficient and wind speed on the exposed concrete surface was determined, and the thickness and thermal conductivity of each curing covering material were obtained. Boundary characteristic parameters were measured: Based on the relationship between the convective heat transfer coefficient and wind speed, and the thickness and thermal conductivity of each curing covering material, the equivalent heat transfer coefficient with covering was calibrated; based on the relationship between the convective heat transfer coefficient and wind speed, the convective mass transfer coefficient was calibrated using the heat-mass analogy method; and based on the convective mass transfer coefficient, the surface evaporation flux was calibrated. in, The convective heat transfer coefficient; The surface natural convection heat transfer coefficient; This refers to the wind speed coefficient; Wind speed; This refers to the wind speed coefficient; This is the equivalent heat transfer coefficient when there is a covering. For the first The thickness of the covering material; For the first Thermal conductivity of the covering material; The number of covering layers; For interlayer air thermal resistance; The convective mass transfer coefficient; air density; The specific heat capacity of air; For Lewis numbers; For surface evaporation flux; The relative humidity of the concrete surface; The density is the saturated water vapor density. The concrete surface temperature; The relative humidity of the environment; The ambient temperature; The molar mass of water; The density of liquid water; It is capillary pressure.

3. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The expression for the energy conservation equation is: in, The density of concrete; For temperature hydration degree is Specific heat capacity at time; The current temperature; For time; For gradient operators; The moisture content is hydration degree is Thermal conductivity at that time; For temperature gradient; For temperature hydration degree is The hydration heat release power density at that time.

4. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The expression for the mass conservation equation is: in, This represents the current moisture content. For time; For gradient operators; The moisture content is hydration degree is The current isothermal moisture diffusivity at that time; For moisture content gradient; For temperature The moisture content is The current non-isothermal diffusion coefficient at that time; For temperature gradient; This refers to the amount of chemically bound water.

5. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The expressions for the thermal boundary and the wet boundary are: in, Thermal conductivity; The current temperature; The direction of the unit outward normal to the concrete surface; This is the equivalent heat transfer coefficient when there is a covering. The concrete surface temperature; The ambient temperature; Solar radiation absorptivity; Solar radiation intensity; The latent heat of vaporization of water; For surface evaporation flux; For concrete surfaces; The expression for the wet boundary during normal evaporation is: in, This represents the current isothermal moisture diffusion coefficient; This represents the current moisture content. When controlling the spray system, a constant boundary value is used for the thermal boundary: in, This refers to the surface moisture content of the concrete. This represents the saturated moisture content.

6. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The expression for the posterior update of the improved ensemble Kalman filter is: in, For the first The posterior update result of the inversion of a set of perturbed members; The first time step generated based on the best state estimate of the previous time step The predicted state of a set member with perturbation at the current time step; This is the gain attenuation coefficient; The Kalman gain matrix; For real-time observation by sensors; The observation matrix; For the observation noise simulation term; It is a non-linearity index; This is the critical threshold for nonlinearity. This is an unsteady thermal-humidity coupled transport control model; The actual covariance of the set members after propagation through the unsteady thermal-humid coupling transmission control model; Covariance based on linearized approximation; It is the F-norm.

7. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 1, characterized in that, The cost function of the MPC-constrained predictive control problem is: in, It is the cost function; For control step index; To control the total number of steps; for The temperature difference between inside and outside at any given moment; The target temperature difference; This is the weight matrix for the temperature difference term; for The relative humidity of the concrete surface at any given time; Minimum permissible surface relative humidity; This is the weight matrix for the humidity term; In order to be in The control vector applied at each time step; This is the weight matrix for the energy consumption term.

8. The method for dynamic control of concrete curing process based on unsteady heat and moisture transfer according to claim 7, characterized in that, The control vector includes adjustable data of the concrete curing environment, including the thickness of the covering, the cooling water flow rate of the pre-embedded cooling water pipes, the cooling water temperature, the start and stop of the spray system, and the spray water temperature.

Images