Method for controlling temperature gradient time-varying disturbance and shrinkage defect in casting solidification process

Abstract

The present application relates to the technical field of casting solidification process control and casting defect inhibition, and particularly relates to a method for controlling temperature gradient time-varying disturbance and shrinkage defect in casting solidification process. Temperature data of hot spot area, riser neck area, feeding path area and near-surface of the mold are obtained; cooling rate and projected temperature gradient along the feeding direction are determined; Niyama estimation result with bias and margin residual are calculated; cooling loop control instruction is solved according to feeding path state and influence coefficient; integral term is suspended or removed according to mechanical in-place hard feedback signal of regulating valve, and unexecuted cooling demand is transferred to adjacent healthy loop when there is no response. The present application provides a method for controlling temperature gradient time-varying disturbance and shrinkage defect in casting solidification process, which can reduce problems such as criterion jump in low value interval of cooling rate, integral accumulation caused by valve hysteresis and misjudgment of feeding path state.

Description

Technical Field

[0001] This invention relates to the field of casting solidification process control and casting defect suppression technology, and in particular to a method for controlling time-varying temperature gradient disturbance and shrinkage defects during casting solidification process. Background Technology

[0002] Large castings often experience problems during solidification, such as delayed solidification in hot spots, narrowing of the feeding channels in the riser neck area, and changes in mold cooling conditions over time. Current production methods often employ riser feeding, chill placement, and adjustment of air or water cooling branches to alter the solidification temperature field, and use casting simulation software or on-site temperature measurement data to identify areas at risk of shrinkage porosity.

[0003] Existing multi-loop cooling control typically uses temperature at measurement points, cooling rate, or process curves obtained from offline simulation as the control basis. Some schemes use the Niyama criterion for online judgment of shrinkage porosity and adjust the opening of cooling branches according to the criterion values ​​of each hot zone. The above approach can reflect temperature field changes in the early stage of solidification, but in the later stage of solidification or during the eutectic plateau, when the cooling rate is in the low range, the denominator in the Niyama criterion is sensitive to temperature measurement noise. Small drifts in the field sampling signal can cause jumps in the calculation results, and the controller frequently adjusts the valve opening accordingly. The main pipe pressure is prone to fluctuations due to the operation of multiple valves.

[0004] Conventional controllers typically use the completion of software commands as the end condition for the control cycle. There is a time lag between the valve core, valve stem, and the flow of the cooling medium in the regulating valve. When valve jamming, pipeline air resistance, or liquid supply delay occurs, the integral term will still accumulate the unexecuted cooling error. After the valve resumes operation, the accumulated control quantity may be released into the cooling circuit all at once, causing localized overcooling. For the feeding path, the direction of the temperature gradient between the riser region, riser neck region, and hot spot region changes with the solidification process. If the cooling effect is judged solely by the absolute value of the temperature gradient or the amount of local temperature drop, it is impossible to accurately distinguish critical states such as the maintenance of the feeding path, the risk of premature solidification of the riser neck, and the failure of the feeding channel.

[0005] Therefore, a casting solidification process control method is provided based on the projected temperature gradient, biased Niyama estimation, mechanical positioning hard feedback, and residual transfer of adjacent healthy loops for closed-loop scheduling. This method enables the temperature field criterion, actuator mechanical state, and cooling resource allocation to be determined in a correlated manner within the same control cycle. Summary of the Invention

[0006] To achieve the above-mentioned objectives, this invention provides a method for controlling time-varying disturbances and shrinkage defects in the temperature gradient during the solidification process of castings, thereby mitigating problems such as criterion jumps in the low-value range of cooling rate, valve lag, and misjudgment of the state of the feeding path.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for controlling time-varying temperature gradient disturbance and shrinkage defects during the solidification process of castings, comprising:

[0008] Acquire temperature data for the hot spot region of the casting, the riser neck region, the feeding path region, and the near surface of the mold, and determine the cooling rate and the projected temperature gradient along the feeding direction from the temperature data;

[0009] The margin residual is determined based on the biased Niyama estimation results and the preset lower limit of the compensation target;

[0010] The scheduling weights and influence coefficients are adjusted based on the status of the compensation path and the influence of the cooling loop on the margin residuals of each region.

[0011] Solve for the control commands under constraints of cooling circuit opening, total liquid supply capacity, circuit status, and single-step change.

[0012] After issuing the control command, the mechanical positioning hard feedback signal of the corresponding loop is monitored, and the controller integral term of the corresponding controlled area is suspended or released according to the mechanical positioning hard feedback signal. If the mechanical positioning hard feedback signal is not received within the preset response time limit, the unsatisfied margin residual corresponding to the unexecuted effective control increment of the unresponsive loop is transferred to the adjacent healthy loop according to the preset spatial influence matrix and loop health status. The adjacent healthy loop is a cooling loop that has an adjacent influence relationship with the unresponsive loop on the same controlled area in the spatial influence matrix, is in an available state, and can output an effective mechanical positioning hard feedback signal.

[0013] To further realize the present invention, the following technical solutions may be preferred:

[0014] Preferably, the acquisition of temperature data of the hot spot region, riser neck region, feeding path region and near surface of the mold includes: acquiring temperature sequence by setting a temperature measuring array in the hot spot region, riser neck region, feeding path region and near surface of the mold, and performing sampling time alignment, abnormal jump rejection and consistency judgment of adjacent measuring points on the temperature sequence.

[0015] Preferably, determining the cooling rate and the projected temperature gradient along the feeding direction from the temperature data includes: determining a non-negative cooling rate based on the ratio of the temperature drop at adjacent sampling times to the sampling time interval at the same temperature measurement point, and determining the projected temperature gradient based on the dot product of the unit vector of the feeding direction pointing from the center of the hot spot region to the center of the corresponding riser region and the temperature gradient vector.

[0016] Preferably, determining the margin residual based on the biased Niyama estimation result and the preset lower limit of the compensation target includes: determining the ratio of the temperature drop at adjacent sampling times to the sampling time interval at the same temperature measurement point or the same controlled area as the non-negative cooling rate; introducing a bias term that decays as the cooling rate increases into the non-negative cooling rate; and using the non-negative part of the projected temperature gradient along the compensation direction as the numerator to obtain continuous Niyama estimation results; and using the positive difference between the preset lower limit of the compensation target and the Niyama estimation result as the margin residual.

[0017] Preferably, the step of correcting the scheduling weight and influence coefficient based on the relationship between the state of the feeding path and the influence of the cooling circuit on the margin residual of each region includes: identifying the state of the feeding path based on the temperature gradient and its rate of change between the riser region, the riser neck region and the hot spot region, and increasing the scheduling weight of the corresponding hot spot region when the state of the feeding path is the risk of premature solidification of the riser neck or the failure of the feeding channel.

[0018] Preferably, the step of correcting the scheduling weights and influence coefficients based on the compensation path status and the influence relationship between the cooling loop and the margin residuals of each region further includes: updating the influence coefficients based on the changes in the Niyama estimation results of the controlled region before and after the change in the cooling loop control quantity, wherein the influence coefficients are positive, negative, or zero.

[0019] Preferably, the step of solving the control commands under the constraints of cooling circuit opening, total liquid supply capacity, circuit status and single-step change includes: taking the weighted margin residual square sum and control quantity change penalty term as the target, and solving the control commands of each cooling circuit under the constraints of single circuit opening range, total liquid supply capacity, circuit availability status, single-step opening change amount and mold safety temperature.

[0020] Preferably, the step of monitoring the mechanical positioning hard feedback signal of the corresponding loop after issuing the control command, and suspending or releasing the integral term according to the mechanical positioning hard feedback signal includes: suspending the integral term of the corresponding controlled area after the control command is issued, and releasing the integral suspension of the corresponding loop and the corresponding controlled area after the hardware interrupt signal output by the valve stem displacement sensor of the regulating valve is captured when the deviation between the actual opening position and the target opening position is less than a preset threshold.

[0021] Preferably, transferring the unexecuted cooling demand to the adjacent healthy loop when the mechanical positioning hard feedback signal is not received includes: freezing the controller integral term of the corresponding controlled area when the mechanical positioning hard feedback signal is not captured within a preset response time limit; converting the difference between the target opening degree and the actual feedback opening degree of the unresponsive loop into a flow-equivalent unexecuted effective control increment; and determining the unmet margin residual based on the product of the unexecuted effective control increment and the influence coefficient of the unresponsive loop on the controlled area; and allocating the unmet margin residual to the adjacent healthy loop only when the influence coefficient of the adjacent healthy loop on the controlled area is positive, the loop is in an available state, and the constraints of mold safety temperature, maximum cooling rate, and total liquid supply capacity are met, according to the spatial influence matrix.

[0022] Preferably, the method further includes: determining the confidence level of the measuring point based on the signal change rate of the temperature measuring point and the consistency of adjacent measuring points, and reducing the weight of the corresponding controlled area in the control command solution when the confidence level of the measuring point decreases; when the overall confidence level of the measuring points in the key area is lower than the threshold, switching to a preset conservative cooling curve for control.

[0023] The beneficial effects of this invention are:

[0024] This invention calculates the margin residual by using the projected temperature gradient along the feeding direction and the biased Niyama estimation results, so that it can still output a continuous shrinkage risk assessment quantity when the cooling rate of the post-solidification section is in the low range. Furthermore, by identifying the feeding path state and updating the influence coefficient, the allocation of cooling resources corresponds to the temperature change state between the riser region, riser neck region, and hot spot region.

[0025] This invention monitors the actual position of the valve stem using a displacement sensor of the regulating valve, and suspends or releases the integral term accordingly. When no valid positioning signal is received, the unexecuted cooling demand is conditionally transferred to an adjacent healthy loop, so that the control command, the mechanical action of the regulating valve, and the transfer of cooling demand are executed according to the determinable signal state. This is used to reduce the integral accumulation when the valve is lag-or jammed and the cooling shock after the action is restored. Attached Figure Description

[0026] Figure 1 This is an overall flowchart of the control method of the present invention.

[0027] Figure 2 This is a schematic diagram of the integral suspension, interrupt release, and residual transfer logic based on mechanical hard feedback of the present invention.

[0028] Figure 3 This is a schematic diagram of the structure of the present invention, which involves the measurement point confidence level and the cooling circuit status in scheduling. Detailed Implementation

[0029] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Example 1

[0032] This embodiment provides a method for controlling time-varying temperature gradient disturbances and shrinkage defects during the solidification process of castings. This method can be implemented jointly by a temperature measuring array, a cooling execution network, a mechanical state hard feedback link, and an edge controller. The temperature measuring array is used to acquire temperature data of the hot spot region, riser neck region, feeding path region, and near-surface of the mold during the casting solidification process. The cooling execution network is used to adjust the cooling medium flow rate of each cooling branch according to the control commands output by the edge controller. The mechanical state hard feedback link is used to connect the mechanical displacement state of the regulating valve stem to the hardware interrupt input channel of the edge controller. The edge controller is used to perform temperature data processing, Niyama estimation, margin residual calculation, feeding path state identification, influence coefficient correction, constrained scheduling, integral suspension, and residual transfer processing.

[0033] In this embodiment, the casting can be ductile iron, cast steel, aluminum alloy, or titanium alloy. The hot spot region of the casting can be determined based on the casting process simulation results, riser arrangement, wall thickness variation region, and historical shrinkage location. The riser neck region is the connection area between the riser and the casting body used for feeding molten metal flow. The feeding path region is the area from the riser region through the riser neck region to the hot spot region. The near-surface of the mold is the temperature measurement area on the mold surface near the cooling circuit arrangement.

[0034] The temperature sensing array may include type B thermocouples, type S thermocouples, type K thermocouples, or fiber optic temperature sensing elements suitable for high-temperature casting environments. Temperature sensing points in the array are respectively arranged in the hot spot region, riser neck region, feeding path region, and near the mold surface. The number of sensing points can be set according to the casting size, the number of hot spots, and the number of cooling loops. The temperature signals output from the sensing points are input to the edge controller via an isolation acquisition module. The edge controller performs sampling time alignment, abnormal jump rejection, and consistency judgment on adjacent sensing points to form a temperature sequence.

[0035] The cooling pipeline network may include a variable frequency main pump, a main pipe, a distributor, multiple independent cooling branches, a return water pipe, regulating valves, and diaphragm-type gas-liquid accumulators. The multiple independent cooling branches correspond to different heat-generating areas of the casting or the mold cooling area. The regulating valves can be proportional regulating valves, electric ball valves, or industrial regulating valves with opening control. The diaphragm-type gas-liquid accumulators are connected in parallel at the main pipe junction or at locations where main pipe pressure fluctuations are concentrated; their pre-charge pressure can be determined based on the main pipe's rated pressure and the cooling medium flow rate.

[0036] The mechanical state hard feedback link includes a displacement sensor installed on the valve stem of each regulating valve to detect the actual position of the valve stem throughout its full stroke. This displacement sensor can be a magnetostrictive displacement sensor, an LVDT displacement sensor, or a potentiometer-type displacement sensor. The output of the displacement sensor is hardwired into the high-speed interrupt input channel of the edge controller. When the edge controller determines that the deviation between the actual valve stem position and the target opening position is less than a preset position tolerance error and remains stable for a preset time, it generates a mechanical positioning hard feedback signal.

[0037] See Figure 1 As shown, the method includes sequential or cyclical data acquisition, cooling rate calculation, projected temperature gradient calculation, Niyama estimation, margin residual calculation, compensation path state identification, influence coefficient correction, control command solution, mechanical hard feedback judgment, integral term processing, and residual transfer processing. The method can be executed according to a fixed sampling period, which can be set based on the casting size, temperature measurement system response time, and regulating valve action time.

[0038] In the data acquisition step, temperature sequences are obtained by setting temperature measurement arrays in the hot spot region, the riser neck region, the feeding path region, and the near-surface of the mold. The temperature sequences are then aligned with sampling times, anomaly jumps are eliminated, and consistency between adjacent measurement points is determined. Subsequently, a non-negative cooling rate is determined based on the ratio of the temperature drop at the same measurement point at adjacent sampling times to the sampling time interval. The projected temperature gradient is determined by the dot product of the unit vector of the feeding direction pointing from the center of the hot spot region to the center of the corresponding riser region and the temperature gradient vector.

[0039] In one specific implementation, the cooling rate is obtained by differential processing of the temperature sequence of the same controlled region, and the temperature gradient vector is determined by the spatial coordinate difference and temperature difference between the same controlled region and its adjacent temperature measurement points. The unit vector of the compensation direction points from the center coordinates of the hot spot region to the center coordinates of the riser region, and the projected temperature gradient is calculated according to the following formula:

[0040]

[0041] in, Let be the projected temperature gradient of region k along the shrinkage direction at time t. Let be the temperature gradient vector of region k at time t. Let be the unit vector representing the shrinkage direction corresponding to region k. As the temperature increases along the direction from the hot spot region to the riser region... For positive; when A value less than zero indicates that the temperature gradient along the feeding direction is reversed or that the feeding channel is at risk of failure.

[0042] In the margin residual calculation step, the cooling rate is defined as a non-negative cooling rate. This is the ratio of the temperature drop at adjacent sampling times to the sampling time interval at the same temperature measurement point or the same controlled area; when the difference result is less than zero, Set to zero. The non-negative part of the projected temperature gradient along the feeding direction is taken as zero. This is used for Niyama estimation, and a bias term that decays with increasing cooling rate is introduced into the cooling rate to obtain a continuous Niyama estimation result. This Niyama estimation result is calculated according to the following formula:

[0043]

[0044] in, Here are the Niyama estimates for region k at time t. Let k be the non-negative cooling rate of region k at time t. The bias coefficient, Attenuation coefficient. Bias coefficient. The unit is consistent with the cooling rate, and the attenuation coefficient is... The unit is the reciprocal of the cooling rate. and The value can be calibrated based on the casting material, temperature measurement noise level, and sampling cycle.

[0045] Subsequently, the lower limit of the preset compensation target will be set. Compared with Niyama's estimation results The positive difference between them is taken as the margin residual, i.e. If the Niyama estimate is lower than the preset lower limit of the shrinkage target, the margin residual is positive; otherwise, it is zero. This margin residual is used to represent the degree of non-compliance of the current controlled area with respect to the shrinkage safety target.

[0046] In the feeding path status identification step, the feeding path status is identified based on the temperature gradient and its rate of change between the riser region, riser neck region, and hot spot region. When the temperature decrease rate of the riser neck region is higher than that of the hot spot region, and the temperature gradient from the riser neck region to the hot spot region continuously decreases, the edge controller marks the corresponding feeding path as having a risk of premature solidification of the riser neck; when the projected temperature gradient along the feeding direction is less than or equal to zero and continues to exceed the set judgment window, the edge controller marks the corresponding feeding path as a feeding channel failure.

[0047] When a risk of premature solidification of the riser neck or failure of the feed channel is identified, the scheduling weight of the corresponding hot section area is increased, so that cooling resources are tilted more towards that hot section area.

[0048] The influence coefficient is used to represent the direction and magnitude of the impact of changes in the equivalent flow control quantity of cooling loop j on the Niyama estimation result or margin residual change in region k. This influence coefficient can be expressed as the change in Niyama estimation result caused by a unit flow equivalent control increment, or as the reduction in margin residual caused by a unit flow equivalent control increment. It is updated online based on the changes in the Niyama estimation result or margin residual of the controlled region before and after the change in the cooling loop control quantity, and can be positive, negative, or zero.

[0049] In the control command solution step, the control commands for each cooling circuit are solved under constraints such as the weighted margin residual sum of squares and the penalty term for control quantity changes, with the objectives being the single-loop opening range, total liquid supply capacity, circuit availability, single-step opening change, and mold safety temperature. This problem can be solved using constrained quadratic programming, and its objective function can be expressed as:

[0050]

[0051] in, Let J be the change in the opening degree of cooling loop j during the current control cycle. Let k be the scheduling weight. For the margin residual of region k, Let j be the influence coefficient of cooling loop j on region k. This is a penalty coefficient for changes in control quantities. Constraints include the upper and lower limits of each cooling circuit's opening, the upper limit of total liquid supply capacity, the available status of each circuit, the upper limit of single-step opening change, and the mold's safe temperature limit. The edge controller generates control commands for each cooling circuit based on the solution results.

[0052] See Figure 2 As shown, after the control command is issued, the edge controller immediately puts the integral term of the corresponding controlled area into a suspended state. In the suspended state, the integral term does not accumulate any unexecuted cooling errors. When the displacement sensor of the regulating valve detects that the deviation between the actual position of the valve stem and the target opening position is less than a preset threshold and triggers a hardware interrupt, the edge controller releases the integral suspension of the corresponding loop and the corresponding controlled area, allowing the integral term to resume normal accumulation in the next cycle.

[0053] If the mechanical positioning hard feedback signal is not captured within the preset response time limit, the edge controller will keep the controller integral term of the corresponding controlled area frozen, and obtain the flow-equivalent unexecuted effective control increment based on the target opening degree of the unresponsive loop, the actual feedback opening degree, and the valve opening degree-flow calibration relationship. For the controlled region k, the unmet margin residual caused by the failure to implement effective control increments can be calculated as follows: It is confirmed that, among them, Let be the influence coefficient of cooling loop j on the controlled area k. Subsequently, the edge controller, based on the spatial influence matrix, selects adjacent healthy loops that have a positive influence on the controlled area, are in an available state, and can output effective mechanical positioning hard feedback signals. Only if the allocation confirms that it will not trigger the mold near-surface safety temperature, maximum cooling rate, or total liquid supply capacity constraints, and the adjacent healthy loops have not reached the single-loop opening limit and the single-step opening change limit, will the unmet margin residual be allocated to adjacent healthy loops for compensation according to the proportion of the positive influence coefficient of each adjacent healthy loop. Unresponsive loops are marked as unavailable or limited available until the fault is cleared.

[0054] The spatial influence matrix is ​​used to characterize the spatial adjacency relationship and influence direction between each cooling loop and each controlled area. When two cooling loops have substitutable positive influences on the same controlled area or spatially adjacent controlled areas, they are marked as adjacent in the spatial influence matrix. The healthy loop refers to a cooling loop that is not marked as faulty or restricted unavailability, has normal mechanical feedback of the regulating valve, meets the confidence level of the temperature measurement associated area, and whose current control quantity has not reached the safety constraint boundary.

[0055] See Figure 3 As shown, the temperature measurement array can also be configured with a measurement point confidence processing unit. This unit determines the confidence level of a measurement point based on the rate of change of the temperature signal, the consistency of adjacent measurement points, the status of the acquisition channel, and the sensor disconnection status. When the confidence level of a certain measurement point is lower than a preset threshold, the edge controller reduces the weight of the corresponding area in the margin residual calculation and scheduling. When multiple measurement points in a critical area are in a low confidence state, the edge controller can call a preset conservative cooling curve and control the opening of the cooling loop according to the preset conservative cooling curve to ensure process safety.

[0056] In this embodiment, the diaphragm-type gas-liquid accumulator is used to absorb the pressure fluctuations generated in the main pipe when multiple regulating valves operate simultaneously. When the edge controller outputs multiple cooling circuit control commands, it can also sort the commands or set the action interval. The action sorting or action interval setting can be determined based on the output value of the main pipe pressure sensor, the number of regulating valves, and the total liquid supply capacity.

[0057] In the verification embodiment, a test system was constructed consisting of an edge controller, a PLC high-speed interrupt input module, eight cooling branches, eight proportional control valves and matching displacement sensors, a variable frequency main pump, a diaphragm gas-liquid accumulator, and sixteen high-temperature measurement points. The test casting was a large ductile iron part, divided into four hot-section regions, each corresponding to a riser region and at least one riser neck region.

[0058] During the plateau phase of solidification, a low cooling rate was observed in a certain hot spot region. Using biased Niyama estimation, the estimated value remained continuously changing, and the control valve opening output by the edge controller did not exhibit reverse switching between adjacent cycles, indicating that main pipe pressure fluctuations were effectively absorbed by the accumulator. In the case of premature cooling in the riser neck region, a continuous decrease in the temperature gradient from the riser neck region to the hot spot region was observed. The edge controller marked the corresponding feeding path as a risk of premature solidification of the riser neck and increased the scheduling weight of the hot spot region. Subsequently, cooling resources were preferentially allocated to loops that had a positive impact on this hot spot region. No continuous shrinkage defects were found in the corresponding area during casting dissection.

[0059] In the case of valve jamming in the cooling circuit, after receiving an opening command, the regulating valve of a certain branch failed to trigger a mechanical hard feedback signal within a preset time limit. The edge controller kept the integral term of the corresponding hot spot area of ​​the branch frozen and conditionally allocated the unexecuted cooling demand to the adjacent healthy circuit according to the spatial influence matrix. The test log showed that no sudden increase in opening occurred after the branch recovered. In the case of temperature point drift, the temperature difference between a near-surface temperature point of a mold and the adjacent temperature point deviated for a long time. The measurement point confidence processing unit reduced the weight of the measurement point, and the edge controller did not concentrate all cooling resources to the area corresponding to the abnormal measurement point; when multiple measurement points in the same critical area were below the confidence threshold, the system switched to the preset conservative cooling curve control state.

[0060] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for controlling time-varying temperature gradient disturbance and shrinkage porosity defects during the solidification process of castings, characterized in that, include: Acquire temperature data of the hot spot region of the casting, the riser neck region, the feeding path region, and the near surface of the mold, and determine the cooling rate and the projected temperature gradient along the feeding direction from the temperature data; The margin residual is determined based on the biased Niyama estimation results and the preset lower limit of the compensation target; The scheduling weights and influence coefficients are adjusted based on the status of the compensation path and the influence of the cooling loop on the margin residuals of each region. Solve for the control commands under constraints of cooling circuit opening, total liquid supply capacity, circuit status, and single-step change. After issuing the control command, the mechanical positioning hard feedback signal of the corresponding loop is monitored, and the controller integral term of the corresponding controlled area is suspended or released according to the mechanical positioning hard feedback signal. If the mechanical positioning hard feedback signal is not received within the preset response time limit, the unsatisfied margin residual corresponding to the unexecuted effective control increment of the unresponsive loop is transferred to the adjacent healthy loop according to the preset spatial influence matrix and loop health status. The adjacent healthy loop is a cooling loop that has an adjacent influence relationship with the unresponsive loop on the same controlled area in the spatial influence matrix, is in an available state, and can output an effective mechanical positioning hard feedback signal.

2. The method according to claim 1, characterized in that, The acquisition of temperature data for the hot spot region, riser neck region, feeding path region, and near-surface of the mold includes: acquiring temperature sequences by setting temperature measurement arrays in the hot spot region, riser neck region, feeding path region, and near-surface of the mold, and performing sampling time alignment, abnormal jump rejection, and consistency judgment on the temperature sequences.

3. The method according to claim 1, characterized in that, The step of determining the cooling rate and the projected temperature gradient along the feeding direction from the temperature data includes: determining the non-negative cooling rate based on the ratio of the temperature drop at adjacent sampling times to the sampling time interval at the same temperature measurement point, and determining the projected temperature gradient based on the dot product of the unit vector of the feeding direction pointing from the center of the hot spot region to the center of the corresponding riser region and the temperature gradient vector.

4. The method according to claim 1, characterized in that, The step of determining the margin residual based on the biased Niyama estimation result and the preset lower limit of the compensation target includes: determining the ratio of the temperature drop at adjacent sampling times to the sampling time interval at the same temperature measurement point or the same controlled area as the non-negative cooling rate; introducing a bias term that decays as the cooling rate increases into the non-negative cooling rate; and using the non-negative part of the projected temperature gradient along the compensation direction as the numerator to obtain continuous Niyama estimation results; and using the positive difference between the preset lower limit of the compensation target and the Niyama estimation result as the margin residual.

5. The method according to claim 1, characterized in that, The method of adjusting the scheduling weight and influence coefficient based on the relationship between the state of the feeding path and the influence of the cooling circuit on the margin residual of each region includes: identifying the state of the feeding path based on the temperature gradient and its rate of change between the riser region, the riser neck region and the hot spot region, and increasing the scheduling weight of the corresponding hot spot region when the state of the feeding path is the risk of premature solidification of the riser neck or the failure of the feeding channel.

6. The method according to claim 5, characterized in that, The method of correcting the scheduling weights and influence coefficients based on the compensation path status and the influence relationship between the cooling loop and the margin residuals of each region also includes: updating the influence coefficients based on the changes in the Niyama estimation results of the controlled region before and after the change in the control quantity of the cooling loop, wherein the influence coefficients are positive, negative or zero.

7. The method according to claim 1, characterized in that, The process of solving the control commands under constraints of cooling circuit opening, total liquid supply capacity, circuit status, and single-step change includes: taking the weighted margin residual square sum and control quantity change penalty term as objectives, and solving the control commands of each cooling circuit under constraints of single circuit opening range, total liquid supply capacity, circuit availability status, single-step opening change amount, and mold safety temperature.

8. The method according to claim 1, characterized in that, The step of monitoring the mechanical positioning hard feedback signal of the corresponding loop after issuing the control command, and suspending or releasing the integral term according to the mechanical positioning hard feedback signal includes: suspending the integral term of the corresponding controlled area after the control command is issued, and releasing the integral suspension of the corresponding loop and the corresponding controlled area after the hardware interrupt signal output by the valve stem displacement sensor of the regulating valve is captured when the deviation between the actual opening position and the target opening position is less than a preset threshold.

9. The method according to claim 1, characterized in that, The step of transferring unexecuted cooling requirements to adjacent healthy loops when the mechanical positioning hard feedback signal is not received includes: freezing the controller integral term of the corresponding controlled area when the mechanical positioning hard feedback signal is not captured within a preset response time limit; converting the difference between the target opening and the actual feedback opening of the unresponsive loop into a flow-equivalent unexecuted effective control increment; and determining the unmet margin residual based on the product of the unexecuted effective control increment and the influence coefficient of the unresponsive loop on the controlled area; and allocating the unmet margin residual to the adjacent healthy loop only when the influence coefficient of the adjacent healthy loop on the controlled area is positive, it is in an available state, and it meets the constraints of mold safety temperature, maximum cooling rate, and total liquid supply capacity, according to the spatial influence matrix.

10. The method according to claim 1, characterized in that, The method further includes: determining the confidence level of the measuring point based on the signal change rate of the temperature measuring point and the consistency of adjacent measuring points, and reducing the weight of the corresponding controlled area in the control command solution when the confidence level of the measuring point decreases; when the overall confidence level of the measuring points in the key area is lower than the threshold, switching to the preset conservative cooling curve for control.

Images