A method and system for optimizing lost foam casting process parameters
By dividing the interface sub-regions and optimizing the negative pressure in segments during the lost foam casting process, and combining the filling-solidification coupling calculation, the problem of lack of physical uniqueness in parameter optimization in the existing technology is solved, and the global optimal solution is obtained, thus improving the scientificity and reliability of process optimization.
Patent Information
- Application Number
- CN202611200701.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-10
- Publication Date
- 2026-11-10
AI Technical Summary
In the optimization of existing lost foam casting process parameters, the constant time-averaged negative pressure cannot adapt to the differentiated requirements of each stage of filling, resulting in the disconnect between thermal boundary conditions and the molten metal filling process. Furthermore, the inverse calculation of the interface heat transfer coefficient is prone to getting trapped in local optima, lacks physical uniqueness, and cannot obtain the global optimal solution.
The interface sub-regions are divided along the filling direction. The heat transfer coefficient of the reference interface is calculated in segments based on the heat flow initiation time. The negative pressure is optimized in segments through the filling process. Combined with the filling-solidification coupling calculation, the response relationship between the optimization parameters and the porosity defect index is established, and the global optimal parameter combination is solved.
It ensures the physical uniqueness of the interfacial heat transfer boundary, optimizes parameters and precisely matches physical processes, reflects the evolution of multiphysics fields through coupled calculations, realizes the mathematical reliability and engineering reproducibility of the globally optimal negative pressure parameters, and improves the scientific decision-making process for process optimization.
Smart Images

Figure CN123069652A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of parameter optimization technology, specifically to a method and system for optimizing and analyzing process parameters in lost foam casting. Background Technology
[0002] In the optimization of lost foam casting process parameters, existing methods typically employ a constant time-averaged negative pressure for filling control, and the inverse calculation of the interfacial heat transfer coefficient heavily relies on initial guesses, converging to a certain value through iterative calculations. This approach has the following drawbacks: First, a constant time-averaged negative pressure cannot adapt to the differentiated requirements for permeability and air gap thickness in the initial, middle, and final stages of filling, leading to a disconnect between the thermal boundary conditions at each stage and the actual filling process and solidification evolution of the molten metal. This results in the optimized parameters lacking a true response to the physical time series. Second, when inversely calculating the interfacial heat transfer coefficient, gradient descent or similar algorithms starting from a single guess are prone to getting stuck in local optima. Different guesses may converge to multiple heat transfer coefficients with different values. Although these coefficients can match the measured temperature history to some extent, they lack physical uniqueness and cannot uniquely correspond to the true interfacial heat transfer capacity. Consequently, the optimal combination of negative pressure parameters obtained is only a mathematical local optimum under a specific guess, rather than the global true optimal solution of the physical system.
[0003] Therefore, it is necessary to adaptively divide the interface sub-regions along the filling direction and the wall thickness distribution, and to back-calculate the reference interface heat transfer coefficient with the minimum deviation based on the measured heat flow signal in stages. At the same time, the filling process is divided into three independent optimization negative pressure segments according to the cavity filling ratio. Then, through filling-solidification coupling calculation and successive refinement optimization strategy, the final combination of process parameters with physical uniqueness, reproducibility and global optimum can be obtained. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a method and system for optimizing process parameters in lost foam casting. This method solves the problem that existing lost foam casting process parameter optimization methods use a constant time-averaged negative pressure, and the interface heat transfer coefficient obtained by reverse calculation is prone to convergence to a local optimum due to the reliance on guessed values. As a result, the obtained optimal parameters lack physical uniqueness and are not the true optimal solution of the physical system.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing and analyzing process parameters in lost foam casting, comprising the following steps: The interface between the casting and the mold is divided into multiple interconnected sub-regions along the filling direction and the wall thickness distribution, and the heat flow initiation time of each sub-region is determined.
[0006] For each interface sub-region, multiple heat transfer stages are divided based on the heat flow initiation time. The interface heat transfer coefficient is iteratively adjusted until convergence through thermal conductivity iteration calculation and deviation comparison, and the reference interface heat transfer coefficient for each heat transfer stage is determined.
[0007] Based on the heat flow initiation time and the cavity filling ratio, the filling process is divided into three filling stages. Three optimization parameters are set for the three filling stages and their value ranges are limited.
[0008] Set the inlet boundary conditions and pressure boundary conditions, calculate the molten metal flow field and the position of the molten metal front within a single time step based on the reference interface heat transfer coefficient, update the interface heat transfer coefficient in combination with the current negative pressure and air permeability, until the filling is completed and the casting is completely solidified, and obtain the comprehensive porosity defect index corresponding to the three optimized parameters through the coupling calculation of filling and solidification.
[0009] The response relationship between three optimization parameters and the comprehensive porosity defect index was established, and the global optimal negative pressure parameter combination was obtained by solving the problem. After the deviation was verified to be within acceptable limits, it was used as the final process parameter combination.
[0010] A lost foam casting process parameter optimization and analysis system includes: The start time determination module is used to divide the interface between the casting and the mold into multiple interconnected sub-regions along the filling direction and in combination with the wall thickness distribution, and to determine the heat flow start time of each sub-region.
[0011] The heat transfer coefficient determination module is used to divide each interface sub-region into multiple heat transfer stages based on the heat flow initiation time. Through heat conduction iteration calculation and deviation comparison, the interface heat transfer coefficient is iteratively adjusted until convergence, and the reference interface heat transfer coefficient for each heat transfer stage is determined.
[0012] The optimization parameter determination module is used to divide the filling process into three filling stages based on the heat flow start time and the cavity filling ratio, and set three optimization parameters for the three filling stages and limit their value range.
[0013] The defect index determination module is used to set the boundary conditions of the pouring inlet and the pressure boundary conditions. Based on the reference interface heat transfer coefficient, it calculates the flow field of the molten metal and the position of the molten metal front within a single time step. It updates the interface heat transfer coefficient in combination with the current negative pressure and air permeability until the filling is completed and the casting is completely solidified. Then, the comprehensive porosity defect index corresponding to the three optimized parameters is obtained through the coupling calculation of filling and solidification.
[0014] The parameter combination determination module is used to establish the response relationship between three optimization parameters and the comprehensive porosity defect index, solve for the globally optimal negative pressure parameter combination, and use the final process parameter combination after verifying that the deviation is qualified.
[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention first adaptively divides the interface into sub-regions based on the filling direction and wall thickness distribution, and then calculates the reference interface heat transfer coefficient in segments based on the heat flow initiation time, ensuring the physical uniqueness of the heat transfer boundary of each region and fundamentally avoiding convergence to local spurious solutions based on guessed values. Secondly, the filling process is divided into three segments according to the actual filling ratio, and the negative pressure range is independently defined, ensuring precise matching between optimization parameters and the physical process. Furthermore, through coupled flow field calculation within a single time step and dynamic updating of the heat transfer coefficient driven by negative pressure / permeability, a comprehensive porosity defect index covering both air entrainment and precipitation mechanisms is introduced as an evaluation target, enabling the coupled calculation to truly reflect the evolution process of multiple physics fields. Finally, through uniform sampling and successive refinement optimization supplemented by verification recalculation, the solution of the globally optimal negative pressure parameter combination is ensured to possess both mathematical reliability and engineering reproducibility, thereby elevating process optimization from empirical trial and error to a scientific decision-making process with clear physical orientation and high robustness. Attached Figure Description
[0016] Figure 1 This is a flowchart of the method for optimizing and analyzing process parameters in lost foam casting according to the present invention.
[0017] Figure 2 This is a flowchart illustrating the iterative adjustment of the interface heat transfer coefficient until convergence in the lost foam casting process parameter optimization analysis method of this invention.
[0018] Figure 3 This is a flowchart illustrating the calculation of the molten metal flow field and the position of the molten metal front in the lost foam casting process parameter optimization analysis method of this invention.
[0019] Figure 4 This is a schematic diagram of the module connections of the lost foam casting process parameter optimization and analysis system of the present invention. Detailed Implementation
[0020] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Please refer to the accompanying drawings. Figure 1 This invention provides a technical solution: a method for optimizing and analyzing process parameters in lost foam casting, comprising the following steps: S1. Along the filling direction, combined with the wall thickness distribution, the interface between the casting and the mold is divided into multiple interconnected sub-regions, and the heat flow initiation time of each sub-region is determined.
[0021] Considering that the actual flow path of molten metal in lost foam casting is determined by the geometric normal of the ingate outlet and the farthest boundary of the cavity, it is necessary to start from the normal of the ingate outlet face and combine it with the farthest endpoint of the cavity to determine an accurate direction line that runs through the entire filling path.
[0022] Secondly, considering that the wall thickness of the casting has a decisive influence on the heat transfer capacity, regions with different wall thicknesses exhibit significant differences in heat capacity and cooling rate. If regions with vastly different wall thicknesses are mixed within the same sub-region to calculate the interfacial heat transfer coefficient, the resulting value cannot represent any true local heat transfer state, thus violating the physical constraint that each interfacial sub-region corresponds to a specific heat transfer behavior. To ensure the inherent consistency of heat transfer characteristics within each sub-region, the process of dividing multiple interconnected interfacial sub-regions is as follows: S101. Obtain the three-dimensional assembly geometric model of the casting and mold. Extract the ingate entity of the gating system from the three-dimensional assembly geometric model. Identify the outflow end face connecting the ingate and the cavity. Calculate the arithmetic mean of the normal vectors of all nodes on the outflow end face to obtain the normal direction. Take the direction of the normal direction pointing into the cavity as the initial filling direction. Traverse all boundary points of the cavity along the initial filling direction. Select the cavity endpoint farthest from the outflow end face of the ingate. Connect the geometric center point of the outflow end face of the ingate with the cavity endpoint. The resulting straight line direction is the filling direction of the molten metal.
[0023] S102. Extract all curved surfaces that fit between the outer surface of the casting and the inner surface of the mold in the 3D assembly geometry model as the contact interface. Along the filling direction, generate multiple parallel sections perpendicular to the filling direction according to a preset cross-sectional spacing. Each parallel section intersects with the contact interface to obtain multiple closed intersection lines. The contact interface area between two adjacent intersection lines is an interface segment. For each interface segment, extract the wall thickness values at all positions on the corresponding casting cross-section, and calculate the average value of all wall thickness values of the cross-section as the representative wall thickness of the interface segment. According to the filling order on the contact interface, the representative wall thickness difference of two adjacent interface segments is compared in turn. If the difference is not greater than the preset wall thickness threshold (10%-20% of the average wall thickness of the casting, preferably, for castings with a wall thickness of 10mm-50mm, the preset wall thickness threshold is 2mm-5mm), then the two interface segments are merged into a continuous area. The merging operation is repeated until all adjacent interface segments have been compared, and finally multiple sequentially connected interface sub-regions are obtained, and all interface sub-regions completely cover the contact interface.
[0024] This embodiment ensures that the ordering of all subsequent interface sub-regions, the matching of heat flow initiation times, and the division of filling stages are all based on the actual flow sequence by controlling the filling direction. This avoids the problem of misalignment of heat flow sequence and negative pressure switching nodes caused by incorrect direction. The partitioning strategy based on wall thickness ensures that the reference interface heat transfer coefficient corresponding to each interface sub-region accurately reflects the intrinsic heat transfer capacity.
[0025] It should be noted that the determination of the heat flow initiation moment must be based on real physical signals rather than numerical simulation inference. However, the temperature gradient at the interface between the molten metal and the mold is extremely large and significantly affected by fluctuations in the contact state. Direct measurement at the interface surface is easily disturbed and generates false signals. Considering the natural fluctuations in ambient temperature inside the mold, simply judging the arrival of heat flow by the absolute temperature exceeding a certain fixed value will lead to misjudgments due to initial temperature differences in different parts and sensor drift. It is necessary to eliminate false triggers caused by single random fluctuations to ensure that the temperature rise inflection point found is the first physical moment when the heat flow actually arrives in that sub-region.
[0026] Specifically, the process of determining the heat flow initiation time of each interface sub-region is as follows: S103. Inside the mold body corresponding to each interface sub-region, select the projection position of the geometric center of the corresponding interface sub-region on the mold side, and embed a temperature acquisition element at a preset depth (preferably 8mm to 12mm) from the contact interface along the direction perpendicular to the contact interface. Simultaneously start all temperature acquisition elements to synchronously acquire data, recording the temperature values at each position according to the preset acquisition time interval, thus obtaining the measured temperature data corresponding to each interface sub-region.
[0027] S104. The stable ambient temperature inside the mold before casting is taken as the temperature rise reference temperature. For the measured temperature data of a single interface sub-region, the temperature values are traversed point by point in chronological order of acquisition time, and the temperature difference between two adjacent acquisition times is calculated sequentially. When a preset number of adjacent temperature differences are continuously detected to be greater than the preset temperature rise threshold (set to 5%-10% of the temperature rise reference temperature), the acquisition time when the first temperature value in the continuous data segment is greater than the sum of the temperature rise reference temperature and the preset offset is backtracked and determined as the temperature rise inflection point, which is used as the heat flow start time of the corresponding interface sub-region.
[0028] Among them, the temperature acquisition element refers to the thermocouple used to measure temperature.
[0029] This embodiment acquires the heat flow arrival signal of each interface sub-region in a non-contact manner, ensuring that the measured temperature data of each sub-region accurately reflects the heat flow penetration time at that location. Through a judgment logic combining continuous temperature difference discrimination and inflection point backtracking, the heat flow initiation time of each interface sub-region has a consistent and repeatable judgment standard, solving the problem of heat exchange behavior in each sub-region being incorrectly classified into different stages due to misaligned initiation times.
[0030] S2. For each interface sub-region, multiple heat transfer stages are divided based on the heat flow initiation time. The interface heat transfer coefficient is iteratively adjusted until convergence through heat conduction iteration calculation and deviation comparison, and the reference interface heat transfer coefficient for each heat transfer stage is determined.
[0031] It should be noted that the physical range of the interfacial heat transfer coefficient is constrained by the limits of the contact gap and the thermal conductivity of the material. If the search is not conducted within the feasible boundary, it may converge to a non-physical value. Therefore, as... Figure 2 As shown, the process of iteratively adjusting the interface heat transfer coefficient until convergence is as follows: S201. For a single interface sub-region, retrieve the solidus temperature of the corresponding casting material. Using the heat flow initiation time of the interface sub-region as the starting point of the time reference, and combining the advancing velocity of the molten metal front during the filling process, calculate the filling completion time when the molten metal completely covers the cavity surface corresponding to the interface sub-region. The time interval from the heat flow initiation time to the filling completion time is defined as the filling contact stage. The time interval from the filling completion time to the casting interface temperature dropping to the solidus temperature is defined as the initial solidification stage. The time interval between the starting and ending points is defined as the middle and late solidification stages, with the casting interface temperature dropping to the solidus temperature as the starting point and the internal temperature of the mold falling back to within the range of ambient temperature fluctuations as the ending point. This completes the division of multiple heat transfer stages for the interface sub-region.
[0032] S202. Retrieve parameters such as the thermal conductivity of the casting material, the thermal conductivity of the mold material, the density of the mold material, the specific heat capacity of the mold material, the maximum and minimum contact gap between the casting and the mold, the thermal conductivity of the pyrolysis gas of the pattern, and the percentage of the interface contact area.
[0033] The lower limit of the physically feasible range of the calculated interface heat transfer coefficient is as follows: taking the maximum contact gap as the maximum thickness of the interface air gap, dividing the thermal conductivity of the pyrolysis gas of the pattern by the maximum contact gap value, and obtaining the interface heat transfer coefficient value under the condition that the air gap completely dominates the heat transfer, which is used as the lower limit of the range.
[0034] The upper limit of the physically feasible range for calculating the interface heat transfer coefficient is as follows: First, take the minimum contact gap as the air gap thickness. Divide the thermal conductivity of the pyrolysis gas of the pattern by the minimum contact gap to obtain the air gap thermal conductivity component. Then, multiply the harmonic average of the thermal conductivity of the casting material and the thermal conductivity of the mold material by the contact area ratio and divide by the contact thermal resistance thickness corresponding to the minimum contact gap to obtain the solid contact thermal conductivity component. Add the air gap thermal conductivity component and the solid contact thermal conductivity component to obtain the interface heat transfer coefficient value under tight contact condition, which is taken as the upper limit of the range.
[0035] Set the total number of guess values, and use the Latin hypercube sampling method to sample the physically feasible range of the interface heat transfer coefficient: divide the physically feasible range of the interface heat transfer coefficient into a number of sub-intervals with equal probability along the direction of numerical magnitude, the same number as the total number of groups, randomly select an independent value in each sub-interval, and perform the above sampling process on the three heat transfer stages respectively. Combine the sampled values of the three heat transfer stages one by one to obtain multiple sets of non-overlapping interface heat transfer coefficient guess values. Each set of interface heat transfer coefficient guess values contains an initial value for each of the three heat transfer stages.
[0036] S203. For a single heat transfer stage in a single interface sub-region, the time range corresponding to the heat transfer stage is extracted. The local area of the mold corresponding to the interface sub-region is selected as the forward heat conduction calculation domain. The forward heat conduction calculation domain is discretized into a one-dimensional grid along the direction perpendicular to the contact interface. A preset number of uniform grid nodes are set, and the spacing between adjacent grid nodes is a preset spatial step size. The first grid node is located at the contact interface between the casting and the mold. The temperature acquisition element is embedded at a grid node with one of the numbered positions. The last grid node is located on the outer surface of the mold. The initial temperature condition is set so that the initial temperature of all grid nodes in the forward heat conduction calculation domain is the ambient temperature before casting. The boundary conditions are set as follows: the first grid node at the contact interface adopts the third type of thermal boundary condition (also known as the convective heat transfer boundary condition, which specifies the convective heat transfer coefficient and fluid temperature between the object boundary and the surrounding fluid when solving the heat conduction differential equation). The interface heat transfer coefficient value of the current iteration and the metal liquid interface temperature of the corresponding stage are used as boundary input parameters. The last grid node on the outer side of the mold adopts the natural convection boundary condition, with the ambient temperature and the preset natural convection heat transfer coefficient as boundary input parameters.
[0037] Starting from the beginning of the heat transfer stage, the calculation proceeds step-by-step with a preset time step (set to 0.01s-0.1s): Within each time step, for all internal grid nodes, the temperature difference between a single internal grid node and its left and right adjacent internal grid nodes is calculated. This temperature difference is multiplied by the thermal conductivity of the casting material, multiplied by the time step, and divided by the square of the spatial step to obtain the heat conduction increment in both directions. The two heat conduction increments are added together and divided by the product of the density and specific heat capacity of the casting material. This is then added to the temperature value of the internal grid node at the previous moment to obtain the current temperature value of the internal grid node. After completing the calculation for all internal grid nodes, the current temperature values of the internal grid nodes at the first and last boundaries are calculated based on the boundary conditions. This step-by-step calculation continues until the end of the heat transfer stage. The temperature values of the corresponding internal grid nodes of the temperature acquisition elements at all time steps are extracted to form the calculated temperature values corresponding to the set of interface heat transfer coefficient values.
[0038] S204. Retrieve the measured temperature data corresponding to the heat transfer stage, extract the measured temperature values according to the same time step, and map the calculated temperature value to the measured temperature value one-to-one according to the time step. Subtract the measured temperature value of the corresponding time step from the calculated temperature value of each time step to obtain the temperature difference for each time step. Square all the temperature differences for each time step, and sum all the squaring results to obtain the sum of squares of the temperature deviations corresponding to the set of interface heat transfer coefficient values. Calculate the adjustment amount of the interface heat transfer coefficient using the gradient descent method: Apply a preset small increment to the current interface heat transfer coefficient value, repeat the positive heat conduction calculation once to obtain the sum of squares of the temperature deviations corresponding to the increment, subtract the original sum of squares of the temperature deviations after the increment, and divide by the small increment value to obtain the gradient value of the sum of squares of the temperature deviations with respect to the interface heat transfer coefficient. Set a preset iteration step size coefficient, invert the gradient value, multiply it by the iteration step size coefficient to obtain the adjustment amount of the interface heat transfer coefficient, and add the adjustment amount to the current iteration's interface heat transfer coefficient value to obtain the corrected interface heat transfer coefficient value.
[0039] S205. Record the sum of squared temperature deviations obtained in this iteration. Retrieve the sum of squared temperature deviations from the previous iteration. Subtract the two values and take the absolute value to obtain the change in deviation between the two iterations. Compare the change in deviation with a preset convergence threshold (set to a change in the sum of squared temperature deviations of less than 10⁻³℃²). If the change in deviation is less than the convergence threshold, the iteration is considered converged, the iteration loop stops, and the interface heat transfer coefficient value corresponding to this iteration is used as the converged heat transfer coefficient value for this heat transfer stage based on the guessed interface heat transfer coefficient value for this group. If the change in deviation is greater than or equal to the convergence threshold, the iteration is considered not converged. The corrected interface heat transfer coefficient value is used as the input value for the next iteration, and the forward heat conduction calculation process is returned to repeat the temperature calculation, deviation calculation, and coefficient correction operations.
[0040] Among them, solidus temperature refers to the critical temperature at which the casting material completely solidifies into a solid state. It is obtained by consulting the material thermophysical property handbook. The higher the value, the higher the temperature at which the material completes solidification.
[0041] The casting interface temperature refers to the temperature value of the side surface of the casting at the interface between the casting and the mold. When it drops to the solidus temperature, it is used as the basis for determining the end of the initial solidification stage.
[0042] The thermal conductivity of the gas produced by the thermal decomposition of the lost foam pattern refers to the thermal conductivity of the gas produced by the thermal decomposition of the lost foam pattern. The larger the value, the stronger the heat transfer capacity of the air gap.
[0043] The interface contact area ratio refers to the proportion of the actual solid contact area between the casting surface and the mold surface under close contact conditions to the nominal contact area. The value is between 0 and 1. The larger the value, the higher the contribution of solid contact to heat conduction.
[0044] The width of the physically feasible range of the interfacial heat transfer coefficient reflects the maximum possible variation of the interfacial heat transfer coefficient under this physical system.
[0045] The natural convection heat transfer coefficient refers to the heat transfer coefficient value when the outer surface of the mold exchanges heat with the surrounding ambient air through natural convection. It is preset, and the value determines the intensity of heat dissipation from the outer surface of the mold.
[0046] The magnitude of the sum of squares of temperature deviations reflects the cumulative degree of overall temperature calculation deviation under the current interface heat transfer coefficient value. The smaller the value, the better the calculated value matches the measured value.
[0047] A tiny increment refers to an extremely small increase applied based on the current value of the interfacial heat transfer coefficient. It is preset, and the magnitude of this value is used to calculate the gradient value.
[0048] The gradient value refers to the rate of change of the sum of squares of temperature deviation with respect to the interfacial heat transfer coefficient. The sign of the value indicates whether increasing the interfacial heat transfer coefficient will increase or decrease the deviation, and the magnitude of the value indicates the sensitivity of the deviation to changes in the coefficient.
[0049] The iteration step size coefficient is a proportional coefficient that controls the adjustment range of the interface heat transfer coefficient in each iteration. It is preset, and the magnitude of this value determines the correction step size.
[0050] This embodiment divides the heat transfer process into three stages based on physical events, performing reverse calculations for each stage. This eliminates systematic biases caused by cross-stage mixing and averaging, ensuring that the coefficients for each stage have a clear physical orientation. By calculating the lower and upper limits of the physically feasible interval, the search range is limited to the truly possible values, fundamentally avoiding convergence to non-physical values. Simultaneously, Latin hypercube sampling ensures that multiple sets of guessed values evenly cover the entire interval, guaranteeing that the global optimal solution is not overlooked. By comparing the change in bias with the convergence threshold, the iteration automatically terminates after reaching sufficient accuracy, resulting in a unique and reproducible physical solution for the final converged heat transfer coefficient value.
[0051] Furthermore, the process of determining the reference interface heat transfer coefficient for each heat exchange stage is as follows: S206. Perform iterative calculations on all predicted interface heat transfer coefficients for each heat transfer stage sequentially to obtain the convergent heat transfer coefficient value corresponding to each predicted value, forming a set of convergent heat transfer coefficient values. Substitute each value in the set of convergent heat transfer coefficient values into the forward heat conduction calculation process, repeating the calculation to obtain the sum of squared temperature deviations corresponding to each convergent heat transfer coefficient value, forming a set of deviation values. Sort all sums of squared temperature deviations in the set of deviation values by value, select the sum of squared temperature deviations with the smallest value, and determine the convergent heat transfer coefficient value corresponding to the smallest deviation value as the reference interface heat transfer coefficient for the corresponding heat transfer stage of the interface sub-region.
[0052] S207. According to the order of filling of the molten metal on the contact interface, for each interface sub-region, execute all processes from S201 to S206 in sequence until all interface sub-regions have been calculated, and finally obtain the reference interface heat transfer coefficients for each interface sub-region corresponding to the three heat transfer stages.
[0053] The number of values in the deviation value set is the same as the number of values in the convergent heat transfer coefficient value set. The magnitude of each value represents the overall deviation between the calculated temperature value and the measured temperature value under the corresponding convergent heat transfer coefficient value. The smaller the value, the more accurate the convergent heat transfer coefficient value.
[0054] In this embodiment, all converged heat transfer coefficient values are re-substituted into the positive heat conduction calculation and sorted and filtered one by one, ensuring that the final determined reference interface heat transfer coefficient is the one that minimizes the deviation between the calculated temperature value and the measured temperature value among all feasible converged solutions, that is, the physical optimal solution with the unique minimum deviation in the entire search space.
[0055] S3. Based on the heat flow start time and the cavity filling ratio, the filling process is divided into three filling stages. Three optimization parameters are set for the three filling stages and their value ranges are limited.
[0056] Considering that the spatial progress of molten metal filling the cavity is not linear and uniform along the filling direction, and the cavity volume corresponding to each interface sub-region is different, if the stages are divided according to geometric length or time, it will lead to an imbalance problem where large volume areas are divided into too many stages in a short period of time, and small volume areas are divided into the same stage for a long time. Therefore, it is necessary to establish an accurate mapping between heat flow time and filling volume by using the proportion of the cavity segment volume of each sub-region to the total volume, so as to ensure that the stage division is based on the actual filling volume of molten metal rather than spatial distance.
[0057] Furthermore, the adjustable range of negative pressure is limited by both the rated capacity of the negative pressure system equipment and the structural stability of the dry sand in the mold. Therefore, the smaller of the rated upper limit and the permeability limit must be used as the upper limit, which, together with the rated lower limit, constructs a feasible negative pressure range that simultaneously satisfies equipment feasibility and process safety. The specific process is as follows: S301. Retrieve the total volume parameters of the casting cavity and the pouring speed of the molten metal in the gating system. Divide the total cavity volume by the pouring speed to calculate the theoretical total filling time. Retrieve the heat flow start time of all interface sub-regions. Select the heat flow start time of the interface sub-region corresponding to the gate position as the filling start time, and select the heat flow start time of the end interface sub-region farthest from the gate as the filling completion time. Calculate the actual total filling time. Take the arithmetic mean of the theoretical total filling time and the actual total filling time as the benchmark time for dividing the filling stages.
[0058] S302. Along the direction of molten metal filling, calculate the cavity segment volume corresponding to each interface sub-region sequentially. Divide the volume of a single cavity segment by the total cavity volume to obtain the single-segment volume ratio corresponding to the interface sub-region. Accumulate the single-segment volume ratios of all interface sub-regions according to the filling order to obtain the cumulative cavity filling ratio for each interface sub-region. Match the heat flow initiation time of each interface sub-region with its corresponding cumulative cavity filling ratio to establish a mapping relationship between the heat flow initiation time and the cumulative cavity filling ratio.
[0059] S303. The preset cumulative cavity filling percentage range is 0% to 30% for the initial filling stage, 30% to 70% for the middle filling stage, and 70% to 100% for the final filling stage. Based on the mapping relationship between the heat flow initiation time and the cumulative cavity filling percentage, the first boundary time corresponding to a cumulative cavity filling percentage of 30% and the second boundary time corresponding to a cumulative cavity filling percentage of 70% are calculated using linear interpolation. The time period from the filling initiation time to the first boundary time is defined as the initial filling stage, the time period from the first boundary time to the second boundary time is defined as the middle filling stage, and the time period from the second boundary time to the filling completion time is defined as the final filling stage, thus completing the division of three continuous filling stages.
[0060] S304. Adjust the upper and lower limits of the rated working negative pressure of the mold-breaking casting negative pressure system, and determine the maximum allowable negative pressure (i.e., the air permeability limit negative pressure) for maintaining the stability of the molding sand structure with dry sand in the mold. Use the lower limit of the rated working negative pressure of the negative pressure system as the lower limit of the optimization parameter. Take the smaller value between the upper limit of the rated working negative pressure and the air permeability limit negative pressure as the upper limit of the optimization parameter. The lower and upper limits together constitute the feasible range of negative pressure values.
[0061] S305. The three optimization parameters correspond to the initial filling stage, the middle filling stage, and the final filling stage, respectively, and are named the initial filling stage negative pressure, the middle filling stage negative pressure, and the final filling stage negative pressure, respectively. The possible values of the initial filling stage negative pressure, the middle filling stage negative pressure, and the final filling stage negative pressure are all limited to the feasible value range of the negative pressure.
[0062] The cumulative cavity filling percentage refers to the value obtained by adding up the single-segment volume percentage of each interface sub-region in the order of filling. The value starts from the single-segment volume percentage of the first sub-region and gradually increases to 100%, representing the percentage of the cavity volume filled by molten metal to the total volume.
[0063] The permeability limit negative pressure value refers to the maximum allowable negative pressure that the dry sand in the mold can withstand to maintain the stability of the molding sand structure. Exceeding this value will cause the molding sand to collapse or disintegrate. The value represents the upper limit of negative pressure allowed by process safety. Below this value, the molding sand structure remains intact.
[0064] It should be noted that the filling start time is defined as the actual physical moment when molten metal begins to enter the mold cavity from the outflow end face of the ingate. This moment is preferentially based on the pressure surge moment measured by the pressure sensor located at the outlet of the gating system. If the deviation between the heat flow start time of the interface sub-region corresponding to the gate position and the pressure surge moment is within ±0.1s, then the heat flow start time can be used as the filling start time; if the deviation exceeds ±0.1s, then the actual outflow moment measured by the pressure sensor must be used as the filling start time to ensure the physical accuracy of the filling stage division.
[0065] This embodiment introduces the arithmetic mean of the theoretical total filling time and the actual total filling time as the benchmark time, avoiding extreme deviations caused by relying solely on theoretical flow rate or solely on heat flow signal. By precisely matching the heat flow start time with the cumulative cavity filling percentage obtained by accumulating the cavity segment volume percentages, and using linear interpolation to determine the boundary times corresponding to 30% and 70%, it ensures that the physical filling volume of the initial, middle, and final stages of filling is balanced and representative, avoiding misalignment between parameter control and physical processes. By introducing a permeability limit negative pressure value to constrain the upper limit of the value, it ensures that the optimization process is always carried out within the dual boundaries of equipment capacity and process safety.
[0066] S4. Set the boundary conditions for the pouring inlet and the pressure boundary conditions. Based on the reference interface heat transfer coefficient, calculate the flow field of the molten metal and the position of the molten metal front within a single time step. Combine the current negative pressure and air permeability to update the interface heat transfer coefficient until the filling is completed and the casting is completely solidified. Then, through the coupling calculation of filling and solidification, obtain the comprehensive porosity defect index corresponding to the three optimized parameters.
[0067] Considering that the heat transfer boundary in the physical simulation must maintain quantitative consistency with the reference interface heat transfer coefficient obtained under actual casting conditions, otherwise the flow field morphology measured in the plexiglass simulated cavity will deviate from reality due to the mismatch in heat transfer conditions. In addition, the inlet velocity, temperature, and negative pressure boundary of the simulation experiment must accurately reproduce the actual casting parameters and the variable negative pressure regime required for subsequent optimization, and the capture of the flow field and front position needs to be completed within each single time step. High-speed image acquisition and grayscale thresholding based on fluorescent tracer particles can obtain the velocity vector distribution and gas-liquid interface position of the entire field without disturbing the flow.
[0068] Therefore, as Figure 3 As shown, the process of calculating the molten metal flow field and the position of the molten metal front within a single time step, based on the reference interface heat transfer coefficient, with the setting of inlet and pressure boundary conditions, is as follows: S401. A simulated cavity made of transparent acrylic glass is fabricated at a 1:1 scale based on the three-dimensional geometric model of the casting cavity. The outer wall of the simulated cavity is vertically divided along the direction of molten metal filling at preset intervals to obtain multiple continuous annular wall segments. The casting wall thickness at the corresponding position of each annular wall segment is measured. Adjacent annular wall segments with a wall thickness difference not exceeding a preset wall thickness threshold are merged to obtain wall regions that correspond one-to-one with the number and spatial position of the aforementioned interface sub-regions. Each wall region is marked with the corresponding interface sub-region number.
[0069] S402. Assemble a heat transfer conditioning layer consisting of multiple layers of thin films with different thermal conductivity for each wall surface area. Retrieve the reference interface heat transfer coefficient value corresponding to each numbered interface sub-region, and adjust the overall heat transfer capacity of the wall surface by increasing or decreasing the number of thin film layers and changing the type of thin film in the corresponding wall surface area. Use a heat flux meter to detect the interface heat flux density of each wall surface area, and calibrate until the heat transfer effect is consistent with the corresponding reference interface heat transfer coefficient, thus completing the physical loading of the interface heat transfer boundary.
[0070] S403. Machining a connecting flange at the location of the inlet gate corresponding to the casting in the simulated cavity, and connecting the constant speed and temperature liquid supply unit through the flange. Retrieving the preset molten metal pouring speed parameters, adjusting the output frequency of the centrifugal pump of the liquid supply unit so that the normal flow velocity of the liquid at the outlet equals the molten metal pouring speed, and the flow direction is towards the inside of the cavity. Retrieving the preset molten metal pouring temperature parameters, adjusting the heating module of the liquid supply unit so that the outlet liquid temperature equals the molten metal pouring temperature, completing the setting of the pouring inlet boundary conditions.
[0071] S404. A sealed evacuation chamber is machined on the outer side of each wall area, and each sealed evacuation chamber is independently connected to a vacuum pumping unit. Three optimized parameters are retrieved: initial negative pressure, mid-filling negative pressure, and final filling negative pressure. These three negative pressure levels and corresponding switching time points are pre-set in the vacuum pumping unit's control system. After starting the pumping unit, the control system sequentially switches between the three negative pressure levels according to the filling duration, ensuring that the negative pressure value on the outer wall of the mold cavity at each filling stage matches the optimized parameters for that stage, thus completing the pressure boundary condition setting.
[0072] S405. Set the high-speed camera's frame rate so that the time interval between two adjacent frames equals the duration of a single time step. Add uniformly dispersed fluorescent tracer particles to the liquid storage tank of the liquid supply unit and stir until the fluorescent tracer particles are evenly distributed in the working fluid. Simultaneously start the liquid supply unit, vacuum pumping unit, and high-speed camera. The high-speed camera captures fluid flow images frame by frame, facing the front of the simulated cavity. For the fluid flow image corresponding to the current single time step, locate the planar coordinates of all fluorescent tracer particles. Compare these coordinates with the corresponding fluorescent tracer particles in the previous single time step image, calculate the displacement and direction of each fluorescent tracer particle within the single time step, and convert this into the fluid velocity vector at each position. The velocity vectors at all positions together constitute the molten metal flow field of the single time step.
[0073] S406. For the fluid flow image corresponding to the same single time step, distinguish the fluid region and the gas region in the fluid flow image by grayscale threshold segmentation. The grayscale value of the fluid region is lower than the preset grayscale threshold, and the grayscale value of the gas region is higher than the preset grayscale threshold. Extract the boundary contour line between the fluid region and the gas region, record the planar coordinates of all feature points on the contour line, and determine the spatial interface corresponding to the boundary contour line as the position of the metal liquid front at the single time step.
[0074] Among them, air permeability refers to the ability of dry sand in the mold to allow gas to pass through under a unit pressure difference. It is measured by an air permeability tester according to GB / T 2684-2009 "Test Methods for Foundry Sand and Mixtures", and the unit is Darcy (D).
[0075] A heat exchange regulating layer is a physical device installed on each wall area to regulate the overall heat exchange capacity of the wall. It is composed of multiple layers of thin films with different thermal conductivity, and the equivalent thermal conductivity can be changed by increasing or decreasing the number of thin film layers and changing the type of thin film.
[0076] Interfacial heat flux density refers to the amount of heat transferred per unit area of the contact interface per unit time. It is obtained by measuring the actual heat flux of each wall area using a heat flux meter. The value is used to compare with the target heat flux corresponding to the reference interface heat transfer coefficient to determine whether the calibration is complete.
[0077] This embodiment obtains the molten metal flow field and leading edge position through a physical experiment using transparent plexiglass to simulate the mold cavity. This avoids the model errors introduced by the complex interaction mechanism between the pyrolysis gas and the molten metal in pure numerical simulation. Simultaneously, by calibrating the heat transfer effect to the reference interface heat transfer coefficient block by block using a heat transfer adjustment layer, the thermal boundary conditions under which the flow observation occurs are strictly equivalent to the heat transfer characteristics obtained by reverse calculation under actual casting conditions, eliminating the potential for flow field measurement deviations from reality due to thermal boundary mismatch. High-speed image acquisition using fluorescent tracer particles obtains the molten metal flow field at each single time step and the molten metal leading edge position extracted through grayscale thresholding, ensuring that the fluid velocity vector and the gas-liquid interface position can be completely recorded within each time step.
[0078] It should be noted that the filling and solidification process is a dynamic, multi-physics coupled evolutionary process. The interfacial heat transfer coefficient changes in real time with the negative pressure and permeability, rather than being a fixed value. Furthermore, porosity defects originate from two different physical mechanisms: molten metal entrapment and dissolved gas precipitation, both influenced by flow, temperature, and pressure history. If the heat transfer coefficient is fixed or only a single defect is evaluated, subsequent calculations will fail to reflect the true physical evolution process, and the final optimization objective will deviate from the actual failure mode. Therefore, the process of updating the interfacial heat transfer coefficient based on the current negative pressure and permeability until filling is complete and the casting is fully solidified, and then obtaining the comprehensive porosity defect index corresponding to the three optimization parameters through coupled calculations of filling and solidification, is as follows: S407. Perform layered correction operations according to the coverage state of the molten metal front in different regions. The interface heat transfer coefficient of the current filling stage is updated by two coefficient corrections: negative pressure matching the air gap thickness and air permeability matching the convection intensity.
[0079] S408. Using the updated interface heat transfer coefficient as the boundary condition, and combining the molten metal flow field and the thermal conductivity of the material, calculate the heat change and final temperature of each element to determine the proportion of the molten metal front coverage and the proportion of solidified elements.
[0080] S409. The filling process is determined in stages based on the coverage ratio of the molten metal front and the proportion of solidified units. When the filling process is not completed, the next time step iteration is driven. When the process is completed, the iteration is terminated and the filling and solidification data is output.
[0081] S410. Based on the filling and solidification data, identify the entrapment region and the pore unit, quantify and calculate the volume ratio of entrapment defects and the number density of pores, and output the entrapment defect score and the pore defect score.
[0082] S411. Calculate the comprehensive porosity defect index by weighting the air entrapment defect score and the precipitated porosity defect score according to the preset weights.
[0083] It should be noted that the specific process of S407 is as follows: Substitute the current negative pressure during the filling stage into the negative pressure-air gap thickness lookup table to obtain the current interface air gap thickness. Calculate the difference between the current interface air gap thickness and the reference air gap thickness to obtain the air gap thickness deviation value. Calculate the product of the air gap thickness deviation value and the unit thickness correction ratio to obtain the total air gap correction ratio. Calculate the difference between value 1 and the total air gap correction ratio to obtain the first correction coefficient.
[0084] Calculate the difference between the current mold permeability and the reference permeability to obtain the permeability deviation value. Calculate the product of the permeability deviation value and the unit permeability correction ratio to obtain the total convection correction ratio. Calculate the sum of value 1 and the total convection correction ratio to obtain the second correction coefficient. Calculate the product of the reference interface heat transfer coefficient and the first correction coefficient to obtain the first corrected heat transfer coefficient. Calculate the product of the first corrected heat transfer coefficient and the second correction coefficient to obtain the corrected interface heat transfer coefficient for the current single time step of the interface sub-region.
[0085] The specific process of S408 is as follows: Multiply the updated interface heat transfer coefficient value by the contact heat transfer area value to obtain the first intermediate product. Subtract the unit temperature value of the previous single time step from the inner wall temperature value of the mold to obtain the temperature difference between the two sides of the interface. Multiply the first intermediate product by the temperature difference between the two sides of the interface, and then multiply by the time length value of the single time step to calculate the interfacial heat transfer value transmitted by the current casting unit through the interface between the casting and the mold in the current single time step.
[0086] Multiply the thermal conductivity of the casting material by the convective heat transfer area to obtain the second intermediate product. Subtract the unit temperature from the previous single time step from the molten metal temperature to obtain the temperature difference between the fluid and the unit. Multiply the second intermediate product by the temperature difference between the fluid and the unit, and then divide by the characteristic length to obtain the convective heat transfer rate. Multiply the convective heat transfer rate by the time step length to calculate the convective heat transfer between the current casting unit and the surrounding flowing molten metal within the current single time step.
[0087] The total cumulative heat transfer value of the current casting unit in the current single time step is calculated by adding the interfacial conduction heat transfer value and the convection heat transfer value. The material density and specific heat capacity values of the current casting unit are retrieved, as well as the geometric volume value. The material density, geometric volume, and specific heat capacity values are multiplied together to obtain the unit heat capacity value. The total cumulative heat transfer value is divided by the unit heat capacity value to calculate the temperature change value of the current casting unit in the current single time step.
[0088] Retrieve the unit temperature value saved at the end of the previous single time step for the current casting unit. Add this value to the temperature change value to calculate the final unit temperature value at the end of the current single time step. If the final unit temperature value is less than or equal to the solidus temperature value, the current casting unit is determined to be in a completely solid state and is marked as a solidified unit.
[0089] Repeat the above operations for each casting unit, including interface conduction heat transfer calculation, convection heat transfer calculation, total cumulative heat transfer calculation, temperature change calculation, unit final temperature calculation, and solidification state marking, until all casting units have completed the temperature value update and solidification state marking for the current single time step.
[0090] Retrieve the molten metal front position data for the current single time step, and sequentially traverse all casting elements. Compare the spatial coordinates of each casting element with the molten metal front position data to determine whether the casting element is located inside or on the boundary of the already filled area of the molten metal front. If the determination result is that it is located inside or on the boundary of the already filled area, then the casting element is counted as a molten metal covered element.
[0091] The total number of molten metal-covered units is calculated by counting all molten metal-covered units. The total number of all casting units is then retrieved. The percentage of molten metal front coverage in the current time step is calculated by dividing the total number of molten metal-covered units by the total number of casting units. The percentage of solidified units in the current time step is also calculated by dividing the total number of solidified units by the total number of casting units. When the percentage of molten metal front coverage reaches 1, the filling process is considered complete, and flow field iteration and front tracking are stopped. Subsequent time steps only continue temperature field calculations and solidification determination until the percentage of solidified units reaches 1, at which point the casting is considered completely solidified, and the entire coupled calculation terminates.
[0092] The solidification data includes the spatial coordinates and volume data of all casting elements at each calculation time step. It also includes the final temperature, liquid volume fraction, and solid volume fraction of all casting elements at each calculation time step, as well as the gas volume fraction, molten metal velocity vector, and pressure of all casting elements at each calculation time step. Finally, it includes the position of the molten metal front at each calculation time step. All data is stored sequentially according to the calculation time steps and indexed by time.
[0093] Specifically, regarding the liquid volume fraction and solid volume fraction: if the final temperature of the unit is greater than or equal to the liquidus temperature, the unit is considered to be in a completely liquid state, with a liquid volume fraction value of 1 and a solid volume fraction value of 0. If the final temperature of the unit is less than or equal to the solidus temperature, the unit is considered to be in a completely solid state, with a liquid volume fraction value of 0 and a solid volume fraction value of 1. If the final temperature of the unit is between the solidus temperature and the liquidus temperature, the unit is considered to be in a paste-like solidified state, and the liquid volume fraction and solid volume fraction are calculated using linear interpolation. The calculation formula converts the temperature value into a phase fraction value based on the lever law or the enthalpy method.
[0094] For the gas phase volume fraction: During the interface heat transfer coefficient update operation at each time step, the gas gap state at each unit location is simultaneously calculated. Specifically, based on the filling negative pressure and mold permeability values at the current time step, the accumulation volume of gas generated by pattern pyrolysis in the interface gap between the molten metal and the mold is calculated. This accumulation volume is divided by the total geometric volume of the unit to obtain the gas phase volume fraction of that unit at the current time step. This calculation process is performed synchronously with the interface heat transfer coefficient update, and the change in gas phase volume fraction directly reflects the influence of the current negative pressure and permeability on gas gap formation.
[0095] The specific process of S410 is as follows: Retrieve the gas phase volume fraction data of all casting units at each calculation time step, and mark casting units with a gas phase volume fraction greater than zero as gas-containing units. Traverse all gas-containing units, and determine whether the gas phase volume fraction of each gas-containing unit is greater than or equal to a preset gas entrapment judgment threshold. The gas entrapment judgment threshold is preset according to the casting material type and process requirements. Gas-containing units with a gas phase volume fraction greater than or equal to the gas entrapment judgment threshold are determined as gas entrapment units, and the area where all gas entrapment units are continuously distributed in space is the gas entrapment region. Calculate the individual volumes of all gas entrapment units, sum the individual volumes of all gas entrapment units, and obtain the total volume value of the gas entrapment region. Retrieve the sum of the volumes of all casting units to obtain the total casting volume value. Divide the total volume value of the gas entrapment region by the total casting volume value to calculate the gas entrapment defect volume ratio.
[0096] Retrieve the solidus temperature and liquidus temperature values of the casting material, as well as the solubility data of dissolved gases in the casting material at different temperatures. Traverse all casting units, extracting the final temperature and liquid volume fraction data of each unit at the end of solidification. Casting units that simultaneously meet the following two criteria are marked as candidate units for precipitation porosity: the first criterion is that the final temperature of the casting unit is less than or equal to the solidus temperature; the second criterion is that the liquid volume fraction of the casting unit at the end of solidification is zero. Retrieve the cooling rate data of all candidate units for precipitation porosity during the solidification process. The cooling rate data is calculated by dividing the temperature change of the unit within the solidification time range by the corresponding time length. Candidate units for precipitation porosity with a cooling rate less than or equal to a preset precipitation threshold are determined as precipitation porosity units. Traverse all precipitation porosity units, counting the number of individual pores formed in space by each precipitation porosity unit. The number density of precipitated pores is calculated by dividing the total number of precipitated pore units by the total volume of the casting.
[0097] A preset mapping relationship between the volume percentage of air entrapment defects and air entrapment defect scores is retrieved. This mapping relationship includes multiple consecutive volume percentage intervals, each corresponding to an air entrapment defect score value. The calculated air entrapment defect volume percentage value is compared with the mapping relationship to determine the interval to which the volume percentage value belongs, and the score value corresponding to that interval is retrieved as the air entrapment defect score output. A preset mapping relationship between the number density of precipitated pores and the precipitated pore defect score is retrieved. This mapping relationship includes multiple consecutive number density intervals, each corresponding to an precipitated pore defect score value. The calculated precipitated pore number density value is compared with the mapping relationship to determine the interval to which the number density value belongs, and the score value corresponding to that interval is retrieved as the precipitated pore defect score output. The air entrapment defect score and the precipitated pore defect score are output as quantitative evaluation indicators of pore defects under the current parameter combination.
[0098] The negative pressure-air gap thickness reference table refers to a data table establishing the correspondence between negative pressure values and interface air gap thicknesses through experimental calibration or numerical simulation. This table is obtained through prior experimental measurements, mapping each air gap thickness value to its corresponding negative pressure value. The magnitude of this value determines the benchmark for air gap thickness under different negative pressures. Specifically, it is established as follows: In a transparent simulated cavity, multiple different negative pressure levels (e.g., -20kPa, -30kPa, -40kPa, -50kPa, -60kPa) are set. At each negative pressure level, a laser displacement sensor is used to measure the air gap thickness between the mold wall and the molten metal front. Each negative pressure level is measured at least five times, and the average value is taken. The negative pressure is then mapped one-to-one with the corresponding average air gap thickness to establish the reference table. If the negative pressure value at the current filling stage falls between two adjacent calibration levels, linear interpolation is used to calculate the corresponding air gap thickness value.
[0099] The reference air gap thickness refers to the value of the interface air gap thickness under the standard reference state corresponding to the reference interface heat transfer coefficient.
[0100] The unit thickness correction ratio refers to the percentage of the interface heat transfer coefficient correction corresponding to the unit air gap thickness deviation. The magnitude of the value determines the sensitivity of the air gap thickness change to the heat transfer coefficient.
[0101] The value of the total air gap correction ratio represents the total correction range caused by changes in air gap thickness.
[0102] The value of the first correction factor is used to make an initial correction to the air gap effect on the heat transfer coefficient of the reference interface. The closer the value is to 1, the smaller the air gap deviation.
[0103] The unit air permeability correction ratio refers to the percentage correction of the interfacial heat transfer coefficient corresponding to the unit air permeability deviation. The magnitude of the value determines the sensitivity of the air permeability change to the convective heat transfer coefficient.
[0104] The magnitude of the total convection correction ratio represents the total correction magnitude caused by changes in air permeability.
[0105] The value of the second correction factor is used to make a secondary correction to the convection effect on the first correction heat transfer coefficient. The closer the value is to 1, the smaller the deviation of the air permeability.
[0106] The contact heat transfer area refers to the effective area of heat transfer between the current casting unit and the mold.
[0107] The temperature of the inner wall of the mold refers to the temperature value of the mesh node at the inner wall of the mold in the current single time step.
[0108] Interface conduction heat transfer refers to the amount of heat transferred to the current casting unit through the interface between the casting and the mold in the current single time step, reflecting the heat transferred to the unit by conduction in that time step.
[0109] The thermal conductivity of casting materials refers to the material's ability to conduct heat.
[0110] The convective heat transfer area refers to the effective area of the current casting unit that undergoes convective heat transfer with the surrounding flowing molten metal.
[0111] Characteristic length refers to the geometric characteristic dimension used to calculate convective heat transfer, such as the unit equivalent diameter. The magnitude of the value affects the convective heat transfer rate.
[0112] Convection heat transfer refers to the amount of heat transferred in or out by the current casting unit and the surrounding flowing molten metal during the current single time step.
[0113] The final temperature value of a unit refers to the result obtained by adding the unit temperature value of the previous single time step to the temperature change value, representing the final temperature of the current casting unit at the end of the current single time step.
[0114] The liquidus temperature is the critical temperature at which a casting material completely melts when it transitions from a solid to a liquid state.
[0115] Liquid phase volume fraction refers to the volume ratio of liquid phase in the current casting unit, with a value between 0 and 1, where 1 is completely liquid and 0 is completely solid.
[0116] Solid volume fraction refers to the volume ratio of solid phase in the current casting unit. The value is between 0 and 1, with 1 being completely solid and 0 being completely liquid.
[0117] The gas phase volume fraction refers to the proportion of the gas phase formed by the accumulation of pattern pyrolysis gas in the interfacial gaps in the current casting unit. The value directly reflects the degree of influence of the current negative pressure and air permeability on the formation of air gaps.
[0118] The molten metal front coverage ratio refers to the ratio obtained by dividing the total number of molten metal coverage units by the total number of all casting units. Its value represents the proportion of the volume of molten metal that has been filled to the total volume of the cavity at the current time step.
[0119] The mapping relationship between the volume percentage of air entrapment defects and the air entrapment defect score refers to a correspondence table containing multiple continuous volume percentage intervals, each interval corresponding to a specific air entrapment defect score value. This table is used to convert the volume percentage values of air entrapment defects into scores. For example, the mapping relationship between the volume percentage of air entrapment defects and the air entrapment defect score is as follows: a volume percentage less than 0.1% corresponds to a score of 0, 0.1% to 0.5% corresponds to a score of 1, 0.5% to 1.0% corresponds to a score of 2, 1.0% to 2.0% corresponds to a score of 3, and greater than 2.0% corresponds to a score of 4. The mapping relationship between the number density of precipitated pores and the precipitated pore defect score is as follows: a pore number density less than 1 / cm³ corresponds to a score of 0, 1 / cm³ to 5 / cm³ corresponds to a score of 1, 5 / cm³ to 10 / cm³ corresponds to a score of 2, 10 / cm³ to 20 / cm³ corresponds to a score of 3, and greater than 20 / cm³ corresponds to a score of 4. In the preset weights, the weight of the air condensation defect score is 0.6, the weight of the precipitated pore defect score is 0.4, and the sum of the two is 1.
[0120] The mapping relationship between precipitated pore number density and precipitated pore defect score refers to a correspondence table containing multiple continuous number density intervals, each interval corresponding to a precipitated pore defect score value, used to convert precipitated pore number density values into scores.
[0121] This embodiment corrects air gap heat conduction based on negative pressure deviation and convection intensity based on air permeability deviation within the current single time step. This allows the updated interface heat transfer coefficient to respond in real time to changes in negative pressure and air permeability during the filling process, avoiding the disconnect between thermal boundary conditions and physical reality caused by using a constant interface heat transfer coefficient. By calculating the interface conduction heat transfer and convection heat transfer unit by unit and tracking the unit's final temperature, liquid phase volume fraction, solid phase volume fraction, metal liquid front coverage ratio, and solidified unit ratio, accurate determination of whether the filling process is complete is achieved, ensuring that the iteration terminates at the correct physical moment. Simultaneously, it identifies air entrapment regions and quantifies the volume ratio of air entrapment defects, identifies candidate units for precipitated pores and determines precipitated pore units based on cooling rate to quantify the number density of precipitated pores, and then combines the two with preset weights to form a comprehensive pore defect index. This allows a single index to take into account the two main sources of pore defects, providing a clear and comprehensive evaluation basis for establishing the response relationship between optimization parameters and this index and solving for the globally optimal negative pressure parameter combination.
[0122] S5. Establish the response relationship between the three optimization parameters and the comprehensive porosity defect index, solve for the global optimal negative pressure parameter combination, and use the result as the final process parameter combination after verifying that the deviation is qualified.
[0123] Considering that the global optimal combination of the three optimization parameters cannot be directly solved using analytical formulas, numerical experimentation is necessary for optimization. Furthermore, the coupled computation of each parameter combination is time-consuming; performing a dense grid search across the entire value range would result in excessive computation and low efficiency, while simple random sampling might miss the optimal region. Therefore, the specific process is as follows: S501. Determine the value range of the three optimization parameters: initial filling negative pressure, mid-filling negative pressure, and final filling negative pressure. These ranges are used as the initial value intervals for the three optimization parameters. Using a uniform experimental design method, generate several sets of initial parameter samples within the initial value intervals. Each set of initial parameter samples contains a set of one-to-one corresponding values for initial filling negative pressure, mid-filling negative pressure, and final filling negative pressure. Number all initial parameter samples according to the generation order to obtain the initial parameter sample group.
[0124] S502. Retrieve the initial negative pressure, mid-filling negative pressure and final negative pressure values from each set of parameter samples in the order of their numbers. Substitute each set of initial parameter samples into the coupling calculation process of filling and solidification, and sequentially execute the interface heat transfer coefficient update, temperature field calculation, process iteration and porosity defect statistics operations to calculate the comprehensive porosity defect index value corresponding to each set of initial parameter samples. Bind and store each set of initial parameter samples with the corresponding comprehensive porosity defect index value to obtain the initial parameter-defect index set.
[0125] S503. Select the smallest comprehensive porosity defect index from the parameter-defect index set as the current optimal defect index, and at the same time, select the corresponding parameter sample group as the current optimal parameter combination.
[0126] S504. Taking the three optimized parameters in the current optimal parameter combination as the center, reduce the value range of the negative pressure at the beginning of filling, the negative pressure at the middle of filling, and the negative pressure at the end of filling according to the preset interval shrinkage ratio (set to 0.5-0.8) to obtain the refined value range. Use the uniform experimental design method to generate several sets of supplementary parameter samples in the refined value range. After adding all the supplementary parameter samples in sequence and numbering them, add them to the parameter-defect index set.
[0127] S505. Calculate the ratio of the length of the refined value interval to the length of the corresponding initial value range interval to obtain the interval shrinkage rate. Compare the interval shrinkage rate with the preset optimization accuracy threshold. If the interval shrinkage rate is greater than the optimization accuracy threshold, sequentially retrieve all supplementary parameter samples and substitute them into the filling and solidification coupling calculation process to calculate the comprehensive porosity defect index value corresponding to each set of supplementary parameter samples and update the parameter-defect index set. Return to the current optimal parameter combination screening step and repeat the execution. If the interval shrinkage rate is less than or equal to the optimization accuracy threshold, mark the current optimal parameter combination as the optimal parameter combination to be verified.
[0128] S506. Substitute the optimal parameter combination to be verified into the filling and solidification coupling calculation process for recalculation to obtain the comprehensive porosity defect index value for verification. Calculate the relative deviation between the comprehensive porosity defect index value for verification and the current optimal defect index value as the verification deviation. Compare the verification deviation with a preset deviation threshold: if the verification deviation is less than the preset deviation threshold, the deviation is deemed acceptable, and the optimal parameter combination to be verified is determined as the final process parameter combination. If the verification deviation is greater than or equal to the preset deviation threshold, the deviation is deemed unacceptable. The optimal parameter combination to be verified and the corresponding comprehensive porosity defect index value for verification are added to the parameter-defect index set. Return to the refinement interval supplementary parameter sample generation step and repeat until the verification deviation is acceptable, and output the final process parameter combination.
[0129] It should be noted that the process of generating several sets of supplementary parameter samples within the refined value interval using the uniform experimental design method is as follows: First, the refined value intervals of the three parameters—initial filling negative pressure, mid-filling negative pressure, and final filling negative pressure—are respectively taken as the three dimensions of the three-dimensional parameter space. The number of levels in each dimension (i.e., the number of discrete values in that dimension) is determined according to the preset number of supplementary sample sets. Then, based on the uniform design table or uniform random sampling algorithm, a set of sampling points uniformly distributed in space is generated in the three-dimensional parameter space. Each sampling point corresponds to a set of triplet samples (containing one value each for initial filling negative pressure, mid-filling negative pressure, and final filling negative pressure), so that the frequency of each level of each factor is equal and uniformly distributed within the experimental range, thereby covering the entire range of the refined interval with the fewest sample sets.
[0130] Among them, the interval shrinkage ratio refers to the fixed ratio coefficient used to shrink the current value interval of each of the three optimization parameters according to the current optimal parameter combination as the center in each refinement. It is preset and the value is between 0 and 1. The smaller the value, the faster the shrinkage and the fewer the refinement iterations.
[0131] The magnitude of the interval shrinkage rate reflects the degree to which the current refinement iteration has shrunk relative to the initial range; the smaller the value, the closer it is to the optimal solution.
[0132] This embodiment generates several initial parameter samples within the initial value range using a uniform experimental design method. This ensures that the initial samples are evenly distributed throughout the search space without clustering, allowing the overall trend of the comprehensive porosity defect index to be fully reflected with the minimum number of samples. By gradually narrowing the value range according to a preset range contraction ratio and generating supplementary parameter samples, centered on the current optimal parameter combination, the search gradually focuses on the optimal region. Through multiple refinements to approach the global optimum, blind exhaustive searching within the entire range is avoided. The iteration termination time is controlled by comparing the range contraction rate with the optimization accuracy threshold, stopping promptly when the accuracy requirements are met, effectively balancing optimization accuracy and computational cost. By comparing the verification deviation with a preset deviation threshold, the reproducibility of the final output optimal parameter combination is ensured, thereby selecting the globally optimal negative pressure parameter combination that is equally effective in physical experiments.
[0133] A system for optimizing and analyzing process parameters in lost foam casting, such as Figure 4 As shown, it includes: The start time determination module is used to divide the interface between the casting and the mold into multiple interconnected sub-regions along the filling direction and in combination with the wall thickness distribution, and to determine the heat flow start time of each sub-region.
[0134] The heat transfer coefficient determination module is used to divide each interface sub-region into multiple heat transfer stages based on the heat flow initiation time. Through heat conduction iteration calculation and deviation comparison, the interface heat transfer coefficient is iteratively adjusted until convergence, and the reference interface heat transfer coefficient for each heat transfer stage is determined.
[0135] The optimization parameter determination module is used to divide the filling process into three filling stages based on the heat flow start time and the cavity filling ratio, and set three optimization parameters for the three filling stages and limit their value range.
[0136] The defect index determination module is used to set the boundary conditions of the pouring inlet and the pressure boundary conditions. Based on the reference interface heat transfer coefficient, it calculates the flow field of the molten metal and the position of the molten metal front within a single time step. It updates the interface heat transfer coefficient in combination with the current negative pressure and air permeability until the filling is completed and the casting is completely solidified. Then, the comprehensive porosity defect index corresponding to the three optimized parameters is obtained through the coupling calculation of filling and solidification.
[0137] The parameter combination determination module is used to establish the response relationship between three optimization parameters and the comprehensive porosity defect index, solve for the globally optimal negative pressure parameter combination, and use the final process parameter combination after verifying that the deviation is qualified.
[0138] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.
[0139] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0140] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0141] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0142] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for optimizing and analyzing process parameters in lost foam casting, characterized in that, Includes the following steps: The interface between the casting and the mold is divided into multiple interconnected sub-regions along the filling direction and the wall thickness distribution, and the heat flow initiation time of each sub-region is determined. For each interface sub-region, multiple heat transfer stages are divided based on the heat flow initiation time. The interface heat transfer coefficient is iteratively adjusted until convergence through thermal conductivity iteration calculation and deviation comparison, and the reference interface heat transfer coefficient for each heat transfer stage is determined. Based on the heat flow initiation time and the cavity filling ratio, the filling process is divided into three filling stages. Three optimization parameters are set for the three filling stages and their value ranges are limited. Set the inlet boundary conditions and pressure boundary conditions, calculate the molten metal flow field and the molten metal front position within a single time step based on the benchmark interface heat transfer coefficient, update the interface heat transfer coefficient in combination with the current negative pressure and air permeability, until the filling is completed and the casting is completely solidified, and obtain the comprehensive porosity defect index corresponding to the three optimized parameters through the coupled calculation of filling and solidification. The response relationship between three optimization parameters and the comprehensive porosity defect index was established, and the global optimal negative pressure parameter combination was obtained by solving the problem. After the deviation was verified to be within acceptable limits, it was used as the final process parameter combination.
2. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The process of dividing the interface between the casting and the mold into multiple interconnected sub-regions along the filling direction and considering the wall thickness distribution is as follows: Extract all contact interfaces between the casting and the mold, identify the outflow end face of the ingate and determine its normal direction, connect the geometric center point of the outflow end face of the ingate with the endpoint farthest from the outflow end face in the cavity, and use the direction of the connecting line as the filling direction of the molten metal. The contact interface is divided into equal-distance sections along the filling direction to obtain multiple continuous interface segments. Adjacent interface segments with a wall thickness difference not exceeding a preset wall thickness threshold are merged to obtain multiple sequentially connected interface sub-regions.
3. The method for optimizing and analyzing process parameters in lost foam casting according to claim 2, characterized in that, The process of determining the heat flow initiation time of each interface sub-region is as follows: Temperature acquisition elements are embedded inside the mold side corresponding to each interface sub-region, along the direction perpendicular to the contact interface of the interface sub-region, and the measured temperature data of each temperature acquisition element are collected synchronously. Temperature rise characteristics are identified for the measured temperature data corresponding to each interface sub-region, and the inflection point time is extracted and determined as the heat flow start time of the corresponding interface sub-region.
4. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The process of iteratively adjusting the interface heat transfer coefficient until convergence is as follows: Based on the heat flow initiation time of the interface sub-region, combined with the thermophysical parameters of the casting material and the advancing velocity of the molten metal front, multiple heat transfer stages corresponding to different solidification states are obtained. Based on the material's thermophysical parameters and the physical limits of the contact state, the physically feasible range of values for the interface heat transfer coefficient is determined, and multiple sets of guessed values for the interface heat transfer coefficient are generated through uniform sampling. The estimated value of the heat transfer coefficient of a single interface is used as the initial input for iteration, and the temperature is calculated step by step using Fourier's law of thermal conductivity. The sum of squared temperature deviations is obtained by comparing the calculated temperature values with the measured temperature data point by point, and the gradient descent method is used to correct the value of the interface heat transfer coefficient. The process of correcting the heat transfer coefficient at the cyclic interface continues until the sum of squared temperature deviations meets the convergence threshold, at which point the converged heat transfer coefficient value is obtained.
5. The method for optimizing and analyzing process parameters in lost foam casting according to claim 4, characterized in that, The process for determining the reference interface heat transfer coefficient for each heat exchange stage is as follows: All convergent heat transfer coefficient values are compared for deviation, and the convergent heat transfer coefficient value with the smallest deviation is selected as the reference interface heat transfer coefficient for the corresponding heat transfer stage. By traversing all interface sub-regions, the following steps are performed sequentially: heat transfer stage division, generation of interface heat transfer coefficient guess, positive heat conduction calculation, deviation calculation and coefficient correction, iterative convergence determination, and selection of benchmark interface heat transfer coefficient, to obtain the benchmark interface heat transfer coefficient for each heat transfer stage in all interface sub-regions.
6. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The filling process is divided into three stages based on the cavity filling ratio. Correspondingly, the filling negative pressure is broken down into three optimization parameters with defined value ranges as follows: The theoretical filling time is calculated by combining the cavity volume and pouring speed. At the same time, the actual filling time is obtained based on the heat flow start time of each interface sub-region. The average of the theoretical filling time and the actual filling time is taken as the benchmark time. Based on the baseline duration, the single-segment volume ratio is calculated by combining the cavity segment volume corresponding to each interface sub-region. The cumulative cavity filling ratio of each interface sub-region is obtained by accumulating segment by segment according to the filling order. This ratio is matched one by one with the corresponding heat flow start time to establish a mapping relationship. Based on the mapping relationship, the boundary time between the two stages is calculated by linear interpolation, and the three continuous stages are divided. The feasible range of negative pressure values is obtained by combining the rated operating parameters of the negative pressure system with the mold permeability limit constraint. The negative pressure of the filling process is broken down into optimization parameters corresponding to the three filling stages, and the value range is limited to the feasible range of negative pressure values.
7. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The process of setting the inlet boundary conditions and pressure boundary conditions, and calculating the molten metal flow field and the position of the molten metal front within a single time step based on the reference interface heat transfer coefficient is as follows: Along the filling direction, combined with the wall thickness distribution, the outer wall of the transparent simulated cavity is divided into wall regions that correspond one-to-one with the interface sub-regions; Adjust the thermal conductivity of the heat exchange conditioning layer corresponding to each wall region so that the heat exchange state of each wall region matches the reference interface heat transfer coefficient of the corresponding interface sub-region. Set the liquid outlet speed of the liquid supply device to the preset molten metal pouring speed, set the liquid outlet temperature to the preset molten metal pouring temperature, and set the liquid outlet direction to the inside of the transparent simulated cavity; The operating negative pressure of each air extraction device is switched sequentially according to the time nodes of the filling stage, so that the negative pressure of the outer wall of the transparent simulated cavity in each filling stage matches the optimized parameters of the corresponding stage. The fluid flow image inside the transparent simulated cavity is acquired frame by frame according to the duration of a single time step. The position of the tracer particle in the fluid flow image of the current single time step and the previous single time step is combined to obtain the metal liquid flow field. For fluid flow images, the interface contour between the fluid region and the gas region is extracted to obtain the corresponding position of the molten metal front.
8. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The process of updating the interface heat transfer coefficient based on the current negative pressure and permeability until the filling is complete and the casting is fully solidified, and then obtaining the comprehensive porosity defect index corresponding to the three optimized parameters through coupled calculation of filling and solidification, is as follows: According to the coverage state of the molten metal front, a layered correction operation is performed in different regions. The interface heat transfer coefficient is updated by two coefficient corrections: negative pressure matching the air gap thickness and air permeability matching the convection intensity. Using the updated interfacial heat transfer coefficient as the boundary condition, the heat change and final temperature of each element are calculated to determine the proportion of the molten metal front covered and the proportion of solidified elements. The filling process is determined in stages based on the proportion of molten metal front coverage and the proportion of solidified units. When the filling process is not completed, the next time step iteration is driven. When the process is completed, the iteration is terminated and the filling and solidification data is output. Based on the filling solidification data, the volume ratio of air entrapment defects and the number density of precipitated pores are quantitatively calculated, and the air entrapment defect score and the precipitated pore defect score are output. The comprehensive porosity defect index is obtained by weighted summing of the air entrapment defect score and the precipitated porosity defect score.
9. The method for optimizing and analyzing process parameters in lost foam casting according to claim 1, characterized in that, The process of establishing the response relationship between three optimization parameters and the comprehensive porosity defect index, solving for the globally optimal negative pressure parameter combination, and verifying that the deviation is within acceptable limits to obtain the final process parameter combination is as follows: Numbered parameter sample groups are generated based on the value ranges of the three optimization parameters; Each parameter sample group is retrieved sequentially according to the numbering order, and the corresponding comprehensive porosity defect index is calculated. The parameter-defect index set is then output. The minimum comprehensive porosity defect index is selected from the parameter-defect index set as the current optimal defect index, and the corresponding parameter sample group is selected as the current optimal parameter combination. Centered on the current optimal parameter combination, the range of values is narrowed to obtain a refined range and supplementary samples are generated to update the parameter-defect index set; The shrinkage rate is calculated based on the refined value range, and the accuracy is judged. If the accuracy is not met, the sample supplementation and the optimal defect index are iterated. When the accuracy is met, the optimal parameter combination to be verified is output. The optimal parameter combination to be verified is substituted into the filling and solidification coupling calculation process for recalculation to obtain the verification deviation. If the verification deviation is qualified, the final process parameter combination is output; otherwise, supplementary samples are added to continue the iteration.
10. A system for optimizing and analyzing process parameters in lost foam casting, characterized in that, include: The start time determination module is used to divide the interface between the casting and the mold into multiple interconnected interface sub-regions along the filling direction and in combination with the wall thickness distribution, and to determine the heat flow start time of each interface sub-region. The heat transfer coefficient determination module is used to divide each interface sub-region into multiple heat transfer stages based on the heat flow initiation time, and iteratively adjust the interface heat transfer coefficient until convergence through heat conduction iteration calculation and deviation comparison, and determine the reference interface heat transfer coefficient for each heat transfer stage. The optimization parameter determination module is used to divide the filling process into three filling stages based on the heat flow initiation time and the cavity filling ratio, and set three optimization parameters for the three filling stages and limit their value range. The defect index determination module is used to set the boundary conditions of the pouring inlet and pressure boundary conditions. Based on the reference interface heat transfer coefficient, it calculates the flow field of the molten metal and the position of the molten metal front within a single time step. It updates the interface heat transfer coefficient in combination with the current negative pressure and air permeability until the filling is completed and the casting is completely solidified. Then, through the coupling calculation of filling and solidification, the comprehensive porosity defect index corresponding to the three optimized parameters is obtained. The parameter combination determination module is used to establish the response relationship between three optimization parameters and the comprehensive porosity defect index, solve for the globally optimal negative pressure parameter combination, and use the final process parameter combination after verifying that the deviation is qualified.
