A vibration parameter self-adaptive control method for lost foam sanding process

Abstract

The present application relates to the field of lost foam casting and adaptive control technology, and particularly relates to a kind of lost foam shakeout process vibration parameter adaptive control method, the method comprises: determining the area collapse characteristics of each monitoring area, and calculating the area collapse cumulative value of each monitoring area according to area collapse characteristics;Determine the area variation coefficient, determine whether the shakeout working condition of each monitoring area is area progress consistent;Determine the shakeout working condition;Based on the shakeout working condition, determine the compensation gain strategy to generate the correction amount set of current vibration parameter;According to the correction amount set, determine the corrected vibration parameter set;Adjust the preset area variation threshold value.The present application effectively solves the technical problems of uneven shakeout and dead angle residue in the prior art caused by relying on fixed vibration parameters, not setting multi-area collapse monitoring means and lacking shakeout effect closed-loop feedback mechanism through intelligent adaptive control of the shakeout process of complex structure castings.

Description

Technical Field

[0001] This invention relates to the field of lost foam casting and adaptive control technology, and in particular to an adaptive control method for vibration parameters in the lost foam casting process. Background Technology

[0002] Lost foam casting is a near-net-shape precision casting process. Its sand removal process, which uses vibration to separate the dry sand in the sand box from the casting, is a crucial step connecting casting and subsequent cleaning processes. For castings with complex internal cavities, deep holes, and narrow flow channels, such as engine blocks, pump housings, and hydraulic components, problems easily arise during sand removal, including uneven sand drop in certain areas, localized sand residue, blockage in dead corners, and difficulty in removing sand blocks from internal cavities. These problems not only reduce sand removal efficiency and prolong production cycle time but can also lead to surface damage or internal blockage of the casting, severely affecting casting quality and becoming one of the bottlenecks restricting the development of lost foam casting technology towards higher precision and efficiency.

[0003] Chinese Patent Publication No. CN121289404A discloses a lost foam casting equipment, including a functional box and a collection box that are interconnected. The collection box is located below the functional box, and a grid support plate is provided between the collection box and the functional box. The collection box is slidably connected to a base in the horizontal direction. A vibration component is provided on the base for driving the collection box and the functional box to reciprocate in the horizontal direction. A first support plate for placing sand molds is provided inside the functional box, and the first support plate is slidably connected to the functional box in the vertical direction. A box plate is provided on the box wall of the functional box. This lost foam casting equipment can realize the lost foam casting process of mold placement, sand filling, pouring, and sand removal. The first support plate and the box plate inside the functional box can be combined to form a sand box. The sand box can be used to place the model, fill sand, and seal and shape to obtain a sand mold. After pouring, the first support plate and the box plate are separated, and the sand mold on the first support plate can be moved to the bottom of the functional box for vibration and sand removal, realizing integrated casting.

[0004] Therefore, the existing technology has the following problems: it relies on fixed or preset vibration parameters to drive the vibration components, and cannot adjust the vibration parameters in real time according to the differences in the collapse progress of different areas in the sand box during the sand shedding process, which easily leads to uneven sand shedding; it relies on the overall vibration sand shedding method, lacks the ability to independently perceive the collapse state of each monitoring area, and cannot identify the regional inconsistency in the collapse progress of complex structure castings due to differences in cavity structure, which easily leads to dead corner residues that go unnoticed; it relies on the unidirectional drive connection of the mechanical structure to achieve sand shedding, and lacks an adaptive closed-loop adjustment mechanism based on the feedback of the sand shedding effect, which cannot optimize the subsequent sand shedding parameters according to the actual effect after the sand shedding is completed, which easily leads to the recurrence of sand shedding problems for the same type of casting. Summary of the Invention

[0005] To address this, the present invention provides an adaptive control method for vibration parameters in the lost foam casting process. This method overcomes the technical problems of uneven casting and dead zone residue caused by relying on fixed vibration parameters, lack of multi-regional collapse monitoring, and lack of closed-loop feedback mechanism for casting effect in the prior art through real-time monitoring of multi-region ultrasonic and vibration signals and adaptive parameter adjustment.

[0006] To achieve the above objectives, the present invention provides an adaptive control method for vibration parameters in the lost foam casting process, comprising:

[0007] The ultrasonic pulse signals and low-frequency vibration acceleration signals corresponding to each monitoring area in the sand box are acquired in real time during the sand removal process of complex structure castings, based on preset rules, in order to determine the regional collapse characteristics of each monitoring area, and calculate the cumulative regional collapse value of each monitoring area based on the regional collapse characteristics.

[0008] The regional variation coefficient is determined based on the statistical characteristics of the cumulative value of regional landslides, and the sandfall conditions in each monitoring area are determined to be consistent based on the comparison results between the regional variation coefficient and the preset regional variation threshold.

[0009] Based on the consistent progress of the regions, the total weight of the castings in the sand box is obtained, and the overall cumulative value of collapse is determined according to the cumulative value of collapse in each region. The structural acoustic-vibration coupling coefficient is calculated by combining the total weight of the castings and the low-frequency vibration acceleration signal within a preset first time period. Based on the threshold comparison result of the structural acoustic-vibration coupling coefficient, the sand falling condition is determined to be dead angle jamming or internal cavity imbalance.

[0010] Based on the aforementioned sandfall conditions, the current vibration parameters of the vibration table are obtained, and a compensation gain strategy is determined in conjunction with the aforementioned sandfall conditions to generate a set of correction values ​​for the current vibration parameters.

[0011] The set of corrected vibration parameters is determined based on the set of corrected values ​​to control the vibration table to perform sand dropping.

[0012] After the controlled vibration table completes the sandfall, the preset regional variation threshold is adjusted based on the spatial distribution uniformity characterization parameter of the accumulated collapse value in the region at the moment the sandfall stops and the rate of change of the accumulated collapse value in the region.

[0013] Furthermore, the process of determining the regional landslide characteristics of each monitoring area includes:

[0014] Within a preset second time period, envelope detection is performed on the ultrasonic pulse signal to extract the pulse echo amplitude sequence. When the pulse echo amplitude is lower than a preset departure threshold, it is determined that a collapse event has occurred in the monitored area.

[0015] Within the same preset second time period, the low-frequency vibration acceleration signal is integrated in the time domain to obtain the vibration velocity signal. When the energy decay rate of the vibration velocity signal exceeds a preset rate threshold, the collapse event is determined to be a valid collapse event.

[0016] The effective landslide events in each monitoring area within the preset second time period are arranged in chronological order to obtain the regional landslide characteristics of the monitoring area.

[0017] Furthermore, the process of calculating the cumulative regional landslide value for each monitoring area based on regional landslide characteristics includes:

[0018] The number of occurrences of the effective landslide events in the landslide characteristics of the region is counted to obtain the cumulative landslide frequency value;

[0019] Based on the determination time of the effective collapse event, the pulse echo amplitude at the same moment is extracted from the ultrasonic pulse signal and used as the single collapse energy characterization value of the effective collapse event.

[0020] The cumulative collapse energy value is obtained by accumulating the single collapse energy characterization values ​​of all the effective collapse events within the preset second time period;

[0021] The cumulative value of landslide frequency and the cumulative value of landslide energy are weighted and fused to obtain the regional landslide cumulative value of the monitored area.

[0022] Furthermore, the process of determining the regional coefficient of variation based on the statistical characteristics of the accumulated values ​​of landslides in the region includes:

[0023] Calculate the average cumulative collapse value of each monitored area within the preset second time period;

[0024] The mean of the sum of squared deviations of the cumulative landslide value in each monitoring area from the average value is calculated to obtain the cumulative landslide standard deviation.

[0025] The coefficient of variation for the region is obtained based on the cumulative standard deviation of the landslide and the cumulative average value of the landslide.

[0026] Furthermore, the process of determining whether the sandfall conditions in each monitoring area are consistent with the regional progress, based on the comparison results of the regional variation coefficient and the preset regional variation threshold, includes:

[0027] When the coefficient of variation of the region is less than or equal to the preset regional variation threshold, the sandfall condition is determined to be a region with consistent progress.

[0028] Furthermore, the process of calculating the acoustic-vibration coupling coefficient of the structure includes:

[0029] The effective vibration energy integral is obtained by integrating the square of the low-frequency vibration acceleration signal within the preset first time period.

[0030] Divide the total cumulative value of overall spalling by the total weight of the casting to obtain the cumulative value of spalling per unit weight;

[0031] The acoustic-vibration coupling coefficient of the structure is obtained by multiplying the cumulative value of collapse per unit weight by the integral of the effective vibration energy.

[0032] Furthermore, the process of determining whether the sandfall condition is a dead zone jam or internal cavity imbalance based on the threshold comparison results of the structural acoustic-vibration coupling coefficient includes:

[0033] When the acoustic-vibration coupling coefficient of the structure is less than or equal to the preset first coupling threshold, the sandfall condition is determined to be a dead angle jam.

[0034] When the acoustic-vibration coupling coefficient of the structure is greater than or equal to the preset second coupling threshold, the sandfall condition is determined to be an internal cavity imbalance.

[0035] The preset first coupling threshold is less than the preset second coupling threshold.

[0036] Furthermore, the process of determining the compensation gain strategy includes:

[0037] When the sandfall condition is a dead zone jam, the monitoring area where the cumulative value of regional collapse is lower than the preset collapse progress threshold is determined as the target monitoring area. The amplitude gain coefficient and frequency gain coefficient are determined based on the difference between the cumulative value of regional collapse in the target monitoring area and the preset collapse progress threshold.

[0038] When the sandfall condition is an internal cavity imbalance, the waveform adjustment coefficient and the excitation force adjustment coefficient are determined based on the distribution characteristics of the cumulative value of the regional collapse in each monitoring area.

[0039] Furthermore, the process of generating the correction set for the current vibration parameters includes:

[0040] Obtain the current vibration parameters of the vibration table, including the current amplitude, current frequency, current waveform parameters, and current excitation force;

[0041] When the sandfall condition is a dead angle jam, the amplitude gain coefficient is multiplied by the current amplitude to obtain the amplitude increase correction amount, the frequency gain coefficient is multiplied by the current frequency to obtain the frequency offset correction amount, and the amplitude increase correction amount and the frequency offset correction amount are combined to generate the correction amount set;

[0042] When the sand-falling condition is an internal cavity imbalance, the waveform adjustment coefficient is superimposed with the current waveform parameter to obtain the waveform parameter correction amount, the excitation force adjustment coefficient is multiplied with the current excitation force to obtain the excitation force correction amount, and the waveform parameter correction amount and the excitation force correction amount are combined to generate the correction amount set.

[0043] Furthermore, the process of adjusting the preset regional variation threshold based on the spatial distribution uniformity characterization parameter of the regional landslide accumulation value at the moment the landslide stops and the rate of change of the regional landslide accumulation value includes:

[0044] Obtain the duration from the start of sand falling to the stop of the vibrating table, as the sand falling completion time;

[0045] The ratio of the standard deviation to the mean of the cumulative landslide value in each monitoring area is calculated and used as a parameter characterizing the spatial distribution uniformity.

[0046] The ratio of the cumulative value of landslides in the region to the time it takes for the sand to fall is calculated as the rate of change of the cumulative value of landslides in the region.

[0047] When the spatial distribution uniformity characterization parameter is greater than the preset uniformity upper limit threshold, the preset regional variation threshold is reduced;

[0048] When the spatial distribution uniformity characterization parameter is less than the preset uniformity lower limit threshold and the rate of change of the cumulative value of regional collapse is greater than the preset efficiency threshold, the preset regional variation threshold is increased.

[0049] Compared with existing technologies, the advantages of this invention are as follows: By employing a dual confirmation mechanism of ultrasonic detection and vibration verification, this invention effectively eliminates misjudgments caused by non-collapse factors such as normal fluctuations in the sand layer and background interference from the vibration table, significantly improving the accuracy and reliability of collapse feature identification. By calculating the coefficient of variation of the cumulative collapse value in each monitoring area, it achieves an objective quantitative judgment of the consistency of regional collapse progress. By comparing the regional coefficient of variation with a preset regional variation threshold, subsequent condition judgments are only entered when the regional progress is consistent, avoiding misjudgments caused by the overall signal masking local anomalies when the regional collapse progress differences are too large. Furthermore, by calculating the structural acoustic-vibration coupling coefficient, differentiated compensation strategies are adopted for two different fault modes, addressing dead zone jamming through… Increasing the amplitude enhances vibration energy input, while decreasing the frequency increases vibration penetration, thus physically breaking down sand grain blockage. To address internal cavity imbalance, the waveform is adjusted to shift vibration energy towards lagging areas, and the excitation force is increased to improve overall collapse efficiency, thus improving the uniformity of collapse progress in each area. The spatial uniformity of collapse is quantified by calculating the ratio of the standard deviation to the mean of the cumulative collapse value in each monitored area after sandfall is completed. The average efficiency of sandfall is quantified by calculating the ratio of the overall cumulative collapse value to the sandfall completion time. When uniformity is poor, the preset regional variation threshold is reduced, allowing the system to identify regional progress differences earlier and adjust accordingly in future sandfalls. When uniformity is good and efficiency is high, the preset regional variation threshold is increased to prevent the system from becoming overly sensitive and causing unnecessary parameter adjustments. This closed-loop feedback mechanism enables the preset area variation threshold to continuously and adaptively evolve with the actual sand shedding effect, achieving a dynamic balance between sand shedding uniformity and efficiency. From signal acquisition, feature extraction, working condition determination to parameter adjustment and threshold optimization, a complete closed loop is formed, realizing intelligent adaptive control of the sand shedding process for complex structure castings. It effectively solves the technical problems of uneven sand shedding and dead zone residue caused by relying on fixed vibration parameters, lacking multi-area collapse monitoring methods, and lacking a closed-loop feedback mechanism for sand shedding effect in existing technologies. Attached Figure Description

[0050] Figure 1 This is a flowchart of the vibration parameter adaptive control method for the lost foam casting process in this embodiment;

[0051] Figure 2 This is a flowchart illustrating the calculation of the cumulative regional landslide value for each monitoring area in this embodiment;

[0052] Figure 3 This is the logic diagram for determining the sandfall condition based on the threshold comparison results of the structural acoustic-vibration coupling coefficient in this embodiment;

[0053] Figure 4 The logic diagram for adjusting the preset region variation threshold in this embodiment is shown. Detailed Implementation

[0054] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0055] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0056] Please see Figure 1 The diagram shows a flowchart of the adaptive control method for vibration parameters in the lost foam casting process of this embodiment. This embodiment provides an adaptive control method for vibration parameters in the lost foam casting process, including:

[0057] The ultrasonic pulse signals and low-frequency vibration acceleration signals corresponding to each monitoring area in the sand box are acquired in real time during the sand removal process of complex structure castings, based on preset rules, in order to determine the regional collapse characteristics of each monitoring area, and calculate the cumulative regional collapse value of each monitoring area based on the regional collapse characteristics.

[0058] The regional variation coefficient is determined based on the statistical characteristics of the cumulative value of regional landslides, and the sandfall conditions in each monitoring area are determined to be consistent based on the comparison results between the regional variation coefficient and the preset regional variation threshold.

[0059] Based on the consistent progress of the regions, the total weight of the castings in the sand box is obtained, and the overall cumulative value of collapse is determined according to the cumulative value of collapse in each region. The structural acoustic-vibration coupling coefficient is calculated by combining the total weight of the castings and the low-frequency vibration acceleration signal within a preset first time period. Based on the threshold comparison result of the structural acoustic-vibration coupling coefficient, the sand falling condition is determined to be dead angle jamming or internal cavity imbalance.

[0060] Based on the aforementioned sandfall conditions, the current vibration parameters of the vibration table are obtained, and a compensation gain strategy is determined in conjunction with the aforementioned sandfall conditions to generate a set of correction values ​​for the current vibration parameters.

[0061] The set of corrected vibration parameters is determined based on the set of corrected values ​​to control the vibration table to perform sand dropping.

[0062] After the controlled vibration table completes the sandfall, the preset regional variation threshold is adjusted based on the spatial distribution uniformity characterization parameter of the accumulated collapse value in the region at the moment the sandfall stops and the rate of change of the accumulated collapse value in the region.

[0063] In this embodiment, a complex structure casting refers to a lost foam casting with at least one of the following structural features: internal cavity, narrow flow channel, deep hole, uneven outer surface, or abrupt change in wall thickness. The above structural features cause the sand particle collapse path to be blocked, the flow direction to change abruptly, or local sand particles to accumulate during the sand falling process, thereby producing significant differences in the collapse progress between each monitoring area.

[0064] In this embodiment, the monitoring areas defined by preset rules refer to dividing the internal space of the sand box into several independent monitoring areas based on the structural features of the casting cavity and the geometric dimensions of the sand box. The specific preset rules are: dividing the sand box into grids with intervals of 80mm to 120mm along its length and 60mm to 100mm along its width, with each grid unit constituting a monitoring area; for local features with independent cavities or deep holes, the location of that feature is designated as a separate monitoring area. The basis for these interval values ​​is that values ​​below the lower limit would result in an excessive number of areas and a surge in data processing, adversely affecting real-time control, while values ​​above the upper limit would make it difficult to effectively identify local spalling differences. In this embodiment, for the engine block casting, a grid interval of 100mm × 80mm is used, and the locations of the four cylinder bores are each designated as a separate monitoring area, forming a total of 16 monitoring areas. This allows for effective monitoring of the spatial distribution of spalling progress while ensuring real-time control.

[0065] In this embodiment, the process of acquiring ultrasonic pulse signals and low-frequency vibration acceleration signals for each monitoring area includes: installing an ultrasonic probe at the position corresponding to each monitoring area above or to the side of the sand box; using the pulse reflection method, transmitting ultrasonic pulses into the sand box at a frequency of 0.5MHz to 2.5MHz and receiving echo signals to detect whether sand particles have collapsed; installing a low-frequency acceleration sensor at the position corresponding to each monitoring area at the bottom of the sand box, with a sampling frequency of 200Hz to 1000Hz and a range of ±5g to ±50g, to collect the dynamic response signal of the area to vibration under the excitation of the vibration table. When sand particles collapse, the mass and stiffness of the area change instantaneously, causing a change in the energy decay rate of the vibration velocity signal. By accelerating... The vibration velocity signal is obtained by time-domain integration of the intensity signal, and its energy decay rate is extracted. When the rate exceeds a preset threshold and is combined with the collapse event detected by ultrasound during the same period, it is jointly determined as a valid collapse event. To avoid signal interference, a time-division excitation strategy is adopted for the ultrasound probes in each monitoring area. The interval between adjacent transmissions is not less than twice the time required for the maximum detection travel, which is 5ms in this embodiment. Low-frequency acceleration signals are used as a reference to adaptively filter and cancel vibration interference coupled in the ultrasound echo. All sensor signal lines use independent shielded cables with single-end grounding, and the installation spacing is not less than 50mm. All sensors are synchronously acquired by a multi-channel data acquisition card and sent to the controller for processing. The sampling rate of the data acquisition card is not less than 5 times the transmission frequency of the ultrasound probe. In this embodiment, the ultrasound probe frequency is 1.0MHz, the acceleration sensor sampling frequency is 500Hz, the acquisition card sampling rate is 5MHz, the transmission interval is 5ms, and the installation spacing is 60mm, achieving accurate identification of collapse events without signal interference.

[0066] In this embodiment, the preset first duration refers to the time window for continuously sampling low-frequency vibration acceleration signals to calculate the structural acoustic-vibration coupling coefficient, which depends on the calculation timeliness and statistical stability. This embodiment uses an orthogonal experimental method to conduct sand-drop tests on the calibration casting, with the fluctuation rate of the structural acoustic-vibration coupling coefficient and the control response delay as optimization objectives. The test durations are set at five levels: 0.1s, 0.2s, 0.5s, 1s, and 2s. The preset first duration of 0.5s, with a fluctuation rate below 5% and the smallest response delay, is selected. This ensures the statistical stability of the structural acoustic-vibration coupling coefficient under normal sand-drop conditions and allows for rapid response during sudden changes in sand-drop conditions, providing a reliable real-time basis for accurate determination of subsequent dead-angle jamming or internal cavity imbalance.

[0067] By utilizing the physical principle of ultrasonic pulse echo amplitude attenuation, direct detection of sandfall events in each monitoring area is achieved. Utilizing the physical principle that sand fall causes transient changes in local mass and stiffness, altering vibration response characteristics, the validity of the sandfall event is verified through the energy attenuation rate of low-frequency vibration acceleration signals, forming a dual confirmation mechanism of ultrasound and vibration, effectively eliminating misjudgments from a single signal source. The regional coefficient of variation is used to eliminate the influence of absolute magnitude differences in sandfall stages on consistency evaluation, achieving an objective quantitative judgment of the consistency of regional sandfall progress. The structural acoustic-vibration coupling coefficient reflects the physical meaning of the amount of sand fall per unit casting weight under unit vibration energy input. Sandfall efficiency is used as a working condition criterion; a low structural acoustic-vibration coupling coefficient indicates that energy is not effectively converted into sandfall, identified as dead zone jamming; a high structural acoustic-vibration coupling coefficient indicates abnormally rapid sandfall, identified as internal cavity imbalance. To address sand grain congestion in dead zones, the system strengthens energy input by increasing amplitude and enhances vibration penetration by decreasing frequency, thus physically breaking down the sand grains. To address internal cavity imbalance, the system adjusts the waveform to shift vibration energy towards the lagging areas of sand shedding and increases excitation force to improve overall efficiency, thus physically improving the uniformity of sand shedding. Spatial uniformity is quantified by the ratio of the standard deviation to the mean of the cumulative sand shedding values ​​in each region after sand shedding, and sand shedding efficiency is quantified by the ratio of the overall cumulative sand shedding value to the sand shedding completion time. A closed-loop feedback mechanism adaptively adjusts the preset regional variation threshold, achieving a dynamic balance between sand shedding uniformity and efficiency. A complete closed loop is formed from signal acquisition, feature extraction, working condition determination to parameter adjustment, enabling intelligent adaptive control of the sand shedding process for complex casting structures. This effectively solves the technical problems of uneven sand shedding and dead zone residue caused by reliance on fixed vibration parameters, lack of multi-region sand shedding monitoring, and lack of a closed-loop feedback mechanism for sand shedding effects in existing technologies.

[0068] Specifically, the process of determining the regional landslide characteristics of each monitoring area includes:

[0069] Within a preset second time period, envelope detection is performed on the ultrasonic pulse signal to extract the pulse echo amplitude sequence. When the pulse echo amplitude is lower than a preset departure threshold, it is determined that a collapse event has occurred in the monitored area.

[0070] Within the same preset second time period, the low-frequency vibration acceleration signal is integrated in the time domain to obtain the vibration velocity signal. When the energy decay rate of the vibration velocity signal exceeds a preset rate threshold, the collapse event is determined to be a valid collapse event.

[0071] The effective landslide events in each monitoring area within the preset second time period are arranged in chronological order to obtain the regional landslide characteristics of the monitoring area.

[0072] In this embodiment, the preset second duration refers to the time window for continuous monitoring of ultrasonic pulse signals and low-frequency vibration acceleration signals to identify valid collapse events. This duration depends on a balance: it cannot be too short, leading to truncation of collapse events and missed or false detections; nor can it be too long, causing multiple events to overlap, making it impossible to distinguish the timing sequence and resulting in a delayed response. This embodiment comprehensively considers the physical duration of sand collapse and the sampling response speed of the control system. An orthogonal experimental method is used to conduct sand-fall tests on a calibration casting. The calibration casting is a representative casting with a structural complexity comparable to the casting to be produced. With the accuracy of collapse event identification and control response delay as optimization objectives, the test time window is set at five levels: 0.5 seconds, 1 second, 2 seconds, 3 seconds, and 5 seconds. The preset second duration of 2 seconds, which has the highest identification accuracy and the smallest response delay, is selected. This allows for the timely differentiation and extraction of collapse features while ensuring complete capture of a single collapse event.

[0073] In this embodiment, the preset detachment threshold is a critical value used to determine whether the ultrasonic pulse echo amplitude is low enough to indicate that sand particles have detached from the casting surface. It depends on a balance between false alarm rate and missed detection rate; it cannot be too high, causing normal sand layer attenuation to be misjudged as a collapse, nor too low, causing the echo amplitude to fail to reach the threshold after actual sand particle collapse, resulting in a missed detection. In this embodiment, before the sand falling begins, the amplitude of the ultrasonic pulse reflected by the intact sand layer in each monitoring area is collected as a reference amplitude. The preset detachment threshold is set to 30% of this reference amplitude. That is, when the pulse echo amplitude is lower than 30% of the reference amplitude, a collapse event is determined to have occurred. This eliminates false detections caused by normal sand layer fluctuations while ensuring that collapse events are captured promptly and accurately.

[0074] In this embodiment, the preset rate threshold is a critical value used to determine whether the energy decay rate of the vibration velocity signal is sufficient to indicate the occurrence of a valid landslide event. The value of this threshold depends on the balance between false alarms and missed detections: a threshold that is too low will misjudge normal background fluctuations of the vibration table as valid landslides, causing false alarms; a threshold that is too high will result in missed detections after actual sand landslides because the energy decay rate does not reach the threshold. In this embodiment, before the sandfall begins, vibration velocity signals are collected from each monitoring area when there are no landslide events, and their energy decay rate is calculated as a benchmark value. The preset rate threshold is set to twice this benchmark value. That is, when the real-time calculated energy decay rate exceeds twice the benchmark value, it is combined with the ultrasonic detection results to determine a valid landslide event, thus eliminating interference from normal vibration table fluctuations while ensuring accurate identification of real landslide events.

[0075] In this embodiment, the process of determining the regional collapse characteristics of each monitoring area is as follows: Within each preset second time period, the ultrasonic pulse signals collected in each monitoring area are first subjected to envelope detection processing to extract the sequence of pulse echo amplitude changes over time. When the pulse echo amplitude at a certain moment is lower than a preset detachment threshold, it is determined that a collapse event of sand particles detaching from the casting surface has occurred in that monitoring area at that moment. Then, within the same preset second time period, the low-frequency vibration acceleration signal collected in the same monitoring area is integrated in the time domain to convert it into a vibration velocity signal. The energy decay rate of the vibration velocity signal is calculated using the sliding window method. The length is set to 1 / 10 of the preset second time period. The integral value of the square of the vibration velocity signal amplitude within the window is calculated. The curves of the integral values ​​of adjacent windows changing with time are fitted, and the absolute value of the slope is taken as the energy decay rate. When the energy decay rate exceeds the preset rate threshold, it indicates that the collapse event has caused a significant change in the vibration response characteristics of the area. Based on this, the collapse event is determined as a valid collapse event. Finally, all valid collapse events detected in each monitoring area within the preset second time period are arranged in chronological order of occurrence to form a time sequence of valid collapse events in the monitoring area during that time period, thus obtaining the regional collapse characteristics of the monitoring area.

[0076] The system achieves direct detection of landslide events by utilizing the physical principle that the amplitude of ultrasonic echoes significantly decreases after sand avalanches. It independently verifies the validity of landslide events by leveraging the physical principle that sand avalanches cause transient changes in local mass and stiffness, altering vibration response characteristics. Furthermore, it eliminates misjudgments from single signal sources by using dual-signal correlation to eliminate the differences in physical mechanisms and independent noise distributions between ultrasound and vibration. Finally, it arranges valid landslide events chronologically, providing fundamental spatiotemporal correlation data for subsequent calculations of landslide frequency and energy accumulation. These processes, from a physical mechanism perspective, ensure the accuracy and reliability of landslide feature extraction.

[0077] Please see Figure 2 As shown, it is a flowchart for calculating the cumulative value of regional landslides in each monitoring area in this embodiment.

[0078] Specifically, the process of calculating the cumulative regional landslide value for each monitoring area based on regional landslide characteristics includes:

[0079] The number of occurrences of the effective landslide events in the landslide characteristics of the region is counted to obtain the cumulative landslide frequency value;

[0080] Based on the determination time of the effective collapse event, the pulse echo amplitude at the same moment is extracted from the ultrasonic pulse signal and used as the single collapse energy characterization value of the effective collapse event.

[0081] The cumulative collapse energy value is obtained by accumulating the single collapse energy characterization values ​​of all the effective collapse events within the preset second time period;

[0082] The cumulative value of landslide frequency and the cumulative value of landslide energy are weighted and fused to obtain the regional landslide cumulative value of the monitored area.

[0083] Specifically, the process of determining the regional coefficient of variation based on the statistical characteristics of the accumulated regional collapse values ​​includes:

[0084] Calculate the average cumulative collapse value of each monitored area within the preset second time period;

[0085] The mean of the sum of squared deviations of the cumulative landslide value in each monitoring area from the average value is calculated to obtain the cumulative landslide standard deviation.

[0086] The coefficient of variation for the region is obtained based on the cumulative standard deviation of the landslide and the cumulative average value of the landslide.

[0087] In this embodiment, the process of determining the regional coefficient of variation based on the statistical characteristics of the regional landslide cumulative value is as follows: First, the arithmetic mean of the regional landslide cumulative values ​​of each monitoring area within a preset second time period is calculated as the landslide cumulative average value; then, the mean of the squared deviations of the regional landslide cumulative values ​​of each monitoring area from the average value is calculated, and the square root is taken to obtain the landslide cumulative standard deviation; finally, the landslide cumulative standard deviation is divided by the landslide cumulative average value to obtain the regional coefficient of variation. The regional coefficient of variation is a dimensionless statistical indicator. The smaller the value, the more consistent the landslide progress among the monitoring areas; the larger the value, the more significant the difference in landslide progress between areas. It is used for subsequent comparison and judgment with a preset regional variation threshold.

[0088] By utilizing the coefficient of variation, the impact of the absolute magnitude difference in the cumulative value of landslides at different stages on the consistency evaluation is eliminated, making the comparison of landslide progress between regions comparable across different time windows. By using the statistical principle of the coefficient of variation to reflect the degree of data dispersion, the consistency of landslide progress in each region is quantified into a normalized index, which can identify whether the differences in landslide progress between regions are within an acceptable range, thus achieving a quantitative assessment of regional landslide obstacles.

[0089] Specifically, the process of determining whether the sandfall conditions in each monitored area are consistent with the regional progress, based on the comparison between the regional variation coefficient and the preset regional variation threshold, includes:

[0090] When the coefficient of variation of the region is less than or equal to the preset regional variation threshold, the sandfall condition is determined to be a region with consistent progress.

[0091] The preset regional variation threshold is a critical value used to determine whether the collapse progress of each monitored area is consistent, and it depends on the setting of the tolerance for regional progress differences. In this embodiment, taking into account both the complexity of the casting structure and the requirements for sand shedding uniformity, the initial value of the preset regional variation threshold is set to 0.2, with a range of 0.1 to 0.3. This can tolerate normal process fluctuations while effectively identifying significant differences in regional collapse progress, providing a reliable basis for subsequent working condition determination.

[0092] In this embodiment, when the regional coefficient of variation is less than or equal to the preset regional variation threshold, it indicates that the dispersion of the cumulative value of landslides between each monitoring area is within an acceptable range, that is, the difference in the landslide progress between each area is not significant. At this time, the landslide condition is determined to be consistent in regional progress. Conversely, when the regional coefficient of variation is greater than the preset regional variation threshold, it indicates that there is a significant difference in the landslide progress between each monitoring area, and it is determined to be inconsistent in regional progress.

[0093] In this embodiment, when the progress of different areas is inconsistent, the dispersion of the collapse progress in each monitored area is too large. At this time, the overall cumulative collapse value and the structural acoustic-vibration coupling coefficient cannot accurately reflect a single fault mode such as dead zone jamming or internal cavity imbalance, because the inconsistency of regional progress is caused by a mixture of multiple factors. Therefore, the regional progress is adjusted first to make it more consistent, and then a refined chemical condition diagnosis is performed under the premise of consistency to avoid misjudgment caused by mutual interference between areas.

[0094] By utilizing the coefficient of variation threshold, the ambiguous process problem of regional collapse progress consistency is transformed into a clear numerical comparison, achieving objective and quantitative judgment of the sandfall condition. By setting regional progress consistency as a prerequisite for entering the judgment of subsequent conditions, misjudgment caused by the overall signal masking local anomalies when the regional collapse progress difference is too large is avoided, improving the accuracy and pertinence of the judgment of dead zone jamming and internal cavity imbalance.

[0095] Specifically, the process of calculating the acoustic-vibration coupling coefficient of a structure includes:

[0096] The effective vibration energy integral is obtained by integrating the square of the low-frequency vibration acceleration signal within the preset first time period.

[0097] Divide the total cumulative value of overall spalling by the total weight of the casting to obtain the cumulative value of spalling per unit weight;

[0098] The acoustic-vibration coupling coefficient of the structure is obtained by multiplying the cumulative value of collapse per unit weight by the integral of the effective vibration energy.

[0099] In this embodiment, after determining that the progress of the monitoring areas is consistent, the total weight of the castings in the sand box is first obtained. This weight can be obtained by weighing before casting or by calculating the casting process parameters. Simultaneously, the cumulative collapse values ​​of each monitoring area are summed to obtain the overall cumulative collapse value. The process of calculating the structural acoustic-vibration coupling coefficient is as follows: Within a preset first time period, the low-frequency vibration acceleration signal is squared and integrated to obtain the effective vibration energy integral; the overall cumulative collapse value is divided by the total weight of the castings to obtain the cumulative collapse value per unit weight; the cumulative collapse value per unit weight is multiplied by the effective vibration energy integral to obtain the structural acoustic-vibration coupling coefficient. This coefficient reflects the amount of sand particles collapsing per unit weight of castings under a unit vibration energy input, i.e., the comprehensive efficiency of converting vibration energy into a collapse effect.

[0100] By using square integrals to convert vibration acceleration signals into total vibration energy, a quantitative characterization of the energy input of the vibration table is achieved. The ratio of the total cumulative spalling value to the total weight of the casting eliminates the influence of casting weight differences on the evaluation of spalling amount. The structural acoustic-vibration coupling coefficient obtained by multiplying the cumulative spalling value per unit weight by the effective vibration energy integral normalizes the two variables of casting weight and vibration energy, enabling the structural acoustic-vibration coupling coefficient to objectively reflect the spalling efficiency under different sand-falling conditions, and providing a unified quantitative basis for subsequent judgment of dead zone jamming or internal cavity imbalance.

[0101] Please see Figure 3 As shown, it is the judgment logic diagram for determining the sandfall condition based on the threshold comparison result of the structural acoustic-vibration coupling coefficient in this embodiment.

[0102] Specifically, the process of determining whether a sand-falling condition is a dead-angle jam or an internal cavity imbalance based on the threshold comparison results of the structural acoustic-vibration coupling coefficient includes:

[0103] When the acoustic-vibration coupling coefficient of the structure is less than or equal to the preset first coupling threshold, the sandfall condition is determined to be a dead angle jam.

[0104] When the acoustic-vibration coupling coefficient of the structure is greater than or equal to the preset second coupling threshold, the sandfall condition is determined to be an internal cavity imbalance.

[0105] The preset first coupling threshold is less than the preset second coupling threshold.

[0106] In this embodiment, the preset first coupling threshold is the critical value for determining that the spalling efficiency is too low and there is dead-angle jamming, while the preset second coupling threshold is the critical value for determining that the spalling efficiency is too high and there is internal cavity imbalance. Together, they constitute the judgment range for normal sand-falling conditions. The values ​​of the two thresholds are determined through the following calibration test: a calibration casting with a structural complexity comparable to the casting to be produced is selected, and a normal sand-falling test is conducted on a sand-falling device using empirically optimized vibration parameters. These vibration parameters should ensure uniform sand-falling, no dead-angle jamming, and no internal cavity imbalance. During the sand-falling process, the structural acoustic-vibration coupling coefficient is continuously calculated within each preset first time period, and the range of change of this coefficient throughout the entire sand-falling cycle is recorded. The average value is taken as the benchmark value. The fluctuation distribution of this coefficient is statistically analyzed, and the standard deviation σ is taken. The benchmark value minus 2σ is taken as the preset first coupling threshold, and the benchmark value plus 2σ is taken as the preset second coupling threshold. In this embodiment, the benchmark value obtained from the calibration test is 1.0, and the standard deviation σ is 0.25. Therefore, the preset first coupling threshold is set to 0.5, and the preset second coupling threshold is set to 1.5. With this setting, the coupling coefficient under normal operating conditions falls within the range between the two thresholds more than 95% of the time, effectively distinguishing between three operating conditions: normal sand falling, dead angle jamming, and internal cavity imbalance.

[0107] By utilizing the physical principle that the structural acoustic-vibration coupling coefficient reflects the efficiency of vibration energy conversion into collapse effect, this coefficient is used as the basis for determining the working condition. A coefficient that is too low indicates that the energy input is not effectively converted into collapse, suggesting the existence of dead zone jamming; a coefficient that is too high indicates an abnormally large collapse amount generated per unit energy, suggesting the existence of internal cavity imbalance. By using dual thresholds to divide the continuous coefficient into three intervals, automatic classification and identification of three working conditions are achieved, providing accurate working condition input for subsequent differentiated compensation strategies.

[0108] Specifically, the process of determining the compensation gain strategy includes:

[0109] When the sandfall condition is a dead zone jam, the monitoring area where the cumulative value of regional collapse is lower than the preset collapse progress threshold is determined as the target monitoring area. The amplitude gain coefficient and frequency gain coefficient are determined based on the difference between the cumulative value of regional collapse in the target monitoring area and the preset collapse progress threshold.

[0110] When the sandfall condition is an internal cavity imbalance, the waveform adjustment coefficient and the excitation force adjustment coefficient are determined based on the distribution characteristics of the cumulative value of the regional collapse in each monitoring area.

[0111] In this embodiment, the preset collapse progress threshold is a critical value used to determine whether the monitored area is lagging behind in collapse, and it depends on the current sand-falling progress. The specific determination method is as follows: First, calculate the ratio of the current cumulative collapse value to the preset final cumulative collapse value as the sand-falling completion progress. The preset final cumulative collapse value is obtained through calibration. A calibration casting with a structural complexity comparable to the casting to be produced is selected. A complete sand-falling test is conducted on the sand-falling equipment using empirically optimized vibration parameters, i.e., amplitude 1.5mm, frequency 25Hz, symmetrical waveform, and 80% of the rated excitation force. The test continues until the rate of change of the cumulative collapse value in the area within three consecutive monitoring cycles of the vibration table is lower than the preset stop threshold, at which point the test automatically stops. The cumulative total collapse value at the moment the sandfall stops is recorded, repeated at least three times, and the arithmetic mean is taken as the preset final cumulative total collapse value. Then, a threshold ratio coefficient is determined according to the sandfall completion progress: when the sandfall completion progress is less than 30%, it is the initial stage of sandfall, and the threshold ratio coefficient is 30%; when the sandfall completion progress is between 30% and 70%, it is the middle stage of sandfall, and the threshold ratio coefficient is 50%; when the sandfall completion progress is greater than 70%, it is the final stage of sandfall, and the threshold ratio coefficient is 70%. Finally, the cumulative total collapse value is multiplied by the threshold ratio coefficient to obtain the preset collapse progress threshold. In this embodiment, the threshold ratio coefficient of 50% at 50% completion is used as the benchmark point, and the threshold changes with the sandfall completion progress in a step function manner. Through this dynamic setting, the threshold is adaptively adjusted with the sandfall process to ensure that the lagging collapse area can be effectively identified in different stages of sandfall.

[0112] In this embodiment, the preset final overall spalling cumulative value is determined through the following calibration test: a calibration casting with a structural complexity comparable to the casting to be produced is selected, and a complete sand-drop test is conducted using empirically optimized vibration parameters. The overall spalling cumulative value at the end of the sand-drop test is used as the calibration value. To ensure the reliability of the calibration value, the above test is repeated at least three times, and the arithmetic mean of each calibration value is taken as the preset final overall spalling cumulative value. If the casting to be produced has a similar structure to a previously produced historical casting, the final overall spalling cumulative value of that historical casting can also be directly used as the expected value. In this embodiment, for the engine block casting, the final overall spalling cumulative values ​​obtained through three calibration tests are 12500, 12800, and 12200, respectively, and the average value of 12500 is taken as the preset final overall spalling cumulative value.

[0113] In this embodiment, the current vibration parameters of the vibration table are first obtained, including the current amplitude, current frequency, current waveform parameters, and current excitation force. When the sand falling condition is a dead zone jam, the monitoring area where the cumulative value of regional collapse is lower than the preset collapse progress threshold is determined as the target monitoring area. The preset collapse progress threshold is calculated by subtracting the cumulative value of regional collapse in the target monitoring area to obtain the collapse lag difference. The amplitude gain coefficient = 1 + collapse lag difference / preset collapse progress threshold. If the calculation result is greater than 1.5, then 1.5 is used, because an amplitude gain coefficient exceeding 1.5 will cause the vibration amplitude to exceed the safe allowable range of the equipment, which may cause damage to the casting or vibration table; this coefficient is used to increase the current amplitude to enhance the vibration energy input. The frequency gain coefficient = 1 - collapse lag difference / preset collapse progress threshold × 0.2. If the calculation result is less than 0.8, then 0.8 is used. The current frequency is multiplied by this coefficient to reduce the vibration frequency. The coefficient 0.2 is the frequency adjustment sensitivity factor, and its value is determined based on the following: 0.2 allows for a maximum frequency reduction of 20%, ensuring sufficient frequency reduction space to enhance vibration penetration while avoiding excessive frequency reduction that would prevent vibration energy from being effectively transferred to the sandbox; the lower limit of 0.8 corresponds to a maximum frequency reduction of 20%, exceeding which would cause the vibration frequency to be too low, exceeding the effective operating frequency band of the vibration table. If multiple target monitoring areas exist, the area with the smallest cumulative collapse value is selected to calculate the collapse hysteresis difference.

[0114] In this embodiment, the preset offset threshold is a critical value used to determine whether the center of gravity shift of the cumulative spalling value is significant and whether there is internal cavity imbalance. The value of this threshold depends on the tolerance for spalling uniformity. In this embodiment, based on the sand box length and the complexity of the casting structure, the preset offset threshold is set to 10% of the sand box length. This value is based on the following: when the offset distance is less than 10%, the center of gravity shift mainly originates from random fluctuations, and the uniformity of sand falling is still within an acceptable range; when the offset distance reaches more than 10%, the center of gravity shift has significantly deviated from the geometric center, indicating a systematic deviation in the spalling progress of each area, requiring waveform and excitation force adjustments. This setting effectively distinguishes between normal fluctuations and true imbalances, avoiding misadjustments and omissions.

[0115] In this embodiment, the specific process for determining the waveform adjustment coefficient and excitation force adjustment coefficient based on the distribution characteristics of the cumulative regional collapse values ​​of each monitoring area is as follows: First, a plane rectangular coordinate system is established with the geometric center of the sand box bottom surface as the origin. The X-axis is along the length direction of the sand box, and the Y-axis is along the width direction of the sand box. The coordinates of each monitoring area are taken as the center point of the corresponding projection area on the bottom surface of the sand box. Then, the spatial distribution analysis of the cumulative regional collapse values ​​of each monitoring area is performed, and the centroid coordinates of the cumulative collapse values ​​are calculated, that is, the weighted average of the coordinates of each monitoring area with its cumulative regional collapse value as the weight. The offset distance of the centroid coordinates relative to the geometric center of the sand box is taken as the straight-line distance from the centroid to the origin; the offset direction is determined by the sign of the centroid X coordinate, positive for forward bias and negative for backward bias, and the sign of the Y coordinate determines the left and right bias. When the center of gravity offset distance exceeds a preset offset threshold, an internal cavity imbalance is determined: if the center of gravity is biased towards the front, it indicates that the front collapses faster and the rear lags behind. In this case, the waveform adjustment coefficient is negative to increase the amplitude of the vibration waveform on the rear side and decrease the amplitude on the front side, thus accelerating the collapse of the lagging rear area. If the center of gravity is biased towards the rear, it indicates that the rear collapses faster and the front lags behind. In this case, the waveform adjustment coefficient is positive to increase the amplitude of the front side and decrease the amplitude on the rear side, thus accelerating the collapse of the lagging front area. The excitation force adjustment coefficient is greater than 1 to increase the total excitation force. If the center of gravity offset distance is less than or equal to the preset offset threshold, the waveform parameters remain unchanged, and only the excitation force is fine-tuned. In this embodiment, the waveform adjustment coefficient ranges from -0.3 to 0.3, with negative values ​​indicating a rearward offset and positive values ​​indicating a forward offset. The excitation force adjustment coefficient ranges from 0.7 to 1.3, with values ​​greater than 1 increasing the excitation force and values ​​less than 1 decreasing the excitation force. The greater the offset distance, the larger the absolute value of the waveform adjustment coefficient, and the greater the degree to which the excitation force adjustment coefficient deviates from 1.

[0116] Specifically, in this embodiment, the waveform adjustment coefficient and the excitation force adjustment coefficient are calculated according to the following formulas:

[0117] Waveform adjustment coefficient = offset distance / sand box length × 0.6. The absolute value of the calculation result is taken and the sign is determined according to the above rules. The value range is -0.3 to 0.3.

[0118] The excitation force adjustment coefficient = 1 + offset distance / sand box length × 0.6, with a value range of 0.7 to 1.3.

[0119] The value of 0.6 is the adjustment strength coefficient, which is determined based on the following: when the offset distance reaches the length of the sand box (i.e., the maximum offset), the absolute value of the waveform adjustment coefficient is 0.6. However, the actual offset distance is much smaller than the length of the sand box due to its geometric dimensions, so the actual coefficient range is controlled between -0.3 and 0.3. The theoretical value of the excitation force adjustment coefficient at the maximum offset is 1.6, which is controlled to be within 1.3 by limiting the amplitude. The value of 0.6 ensures that a sufficiently significant adjustment effect is produced within the typical offset range (i.e., the offset distance is less than half the length of the sand box), while avoiding over-adjustment and instability of the falling sand due to an excessively large coefficient.

[0120] By addressing the issue of stuck corners, the vibration amplitude is increased through an amplitude gain coefficient to enhance energy input, and the vibration frequency is reduced through a frequency gain coefficient to enhance vibration penetration. This allows vibration energy to be effectively transferred to the dead corner areas of the casting, thus physically eliminating sand jamming. For the issue of internal cavity imbalance, the physical principle that the uneven spatial distribution of the spalling progress in each area reflects the cumulative spalling value's center of gravity shift is utilized. The vibration waveform spatial distribution is altered through a waveform adjustment coefficient to shift energy towards the lagging area, and the total excitation force is increased through a vibration force adjustment coefficient to improve overall spalling efficiency. This physically improves the uniformity of spalling progress in each area, ensuring that vibration parameter adjustments match the physical causes of specific fault modes. This achieves differentiated and targeted compensation for stuck corners and internal cavity imbalance, improving the effectiveness of parameter correction and the adaptability of the sand shedding process.

[0121] Specifically, the process of generating the correction set for the current vibration parameters includes:

[0122] Obtain the current vibration parameters of the vibration table, including the current amplitude, current frequency, current waveform parameters, and current excitation force;

[0123] When the sandfall condition is a dead angle jam, the amplitude gain coefficient is multiplied by the current amplitude to obtain the amplitude increase correction amount, the frequency gain coefficient is multiplied by the current frequency to obtain the frequency offset correction amount, and the amplitude increase correction amount and the frequency offset correction amount are combined to generate the correction amount set;

[0124] When the sand-falling condition is an internal cavity imbalance, the waveform adjustment coefficient is superimposed with the current waveform parameter to obtain the waveform parameter correction amount, the excitation force adjustment coefficient is multiplied with the current excitation force to obtain the excitation force correction amount, and the waveform parameter correction amount and the excitation force correction amount are combined to generate the correction amount set.

[0125] In this embodiment, the process of generating the correction set of the current vibration parameters is as follows: First, the current vibration parameters of the vibration table are obtained, including the current amplitude, current frequency, current waveform parameters, and current excitation force. The waveform parameter is defined as the asymmetry coefficient of the vibration waveform, with a value ranging from -1 to 1. A positive value indicates that the amplitude on the front side of the waveform is greater than that on the back side, a negative value indicates that the amplitude on the back side is greater than that on the front side, and zero indicates a symmetrical waveform. When the sandfall condition is a dead-angle jam, the amplitude gain coefficient is multiplied by the current amplitude to obtain the amplitude increase correction, and the frequency gain coefficient is multiplied by the current frequency to obtain the frequency offset correction. The amplitude increase correction and the frequency offset correction are combined to generate the correction set. When the sandfall condition is an internal cavity imbalance, the waveform adjustment coefficient is added to the current waveform parameters to obtain the waveform parameter correction. If the calculation result exceeds the range of -1 to 1, the boundary value is taken. The excitation force adjustment coefficient is multiplied by the current excitation force to obtain the excitation force correction. The waveform parameter correction and the excitation force correction are combined to generate the correction set. The correction set consists of the target correction values ​​for each vibration parameter. The controller updates the current vibration parameters to the corrected values ​​before executing the sandfall.

[0126] For dead-angle jamming conditions, multiplication is used to generate amplitude and frequency corrections. The proportional scaling characteristic of multiplication ensures the adjustment is proportional to the current vibration parameters—when the amplitude gain coefficient is greater than 1, the larger the current amplitude, the greater the absolute value of the increase, ensuring large-amplitude equipment receives sufficient energy increment; when the frequency gain coefficient is less than 1, the higher the current frequency, the greater the absolute value of the frequency reduction, ensuring high-frequency equipment can effectively reduce frequency to enhance vibration penetration. For internal cavity imbalance conditions, addition is used to adjust the waveform parameters, utilizing the translational characteristic of addition to smoothly shift the spatial distribution of the vibration waveform towards the collapsing lag region. The excitation force is proportionally scaled using multiplication, matching the total energy input to the degree of center of gravity shift. These calculation methods adapt to the physical adjustment needs of different operating conditions, ensuring the correction amount changes smoothly and continuously within the equipment's safe range, avoiding sudden parameter changes that could impact the sand-falling process, and guaranteeing the adaptability of the adjustment effect to the current operating state.

[0127] Please see Figure 4 As shown, it is the logic diagram for adjusting the preset region variation threshold in this embodiment.

[0128] Specifically, the process of adjusting the preset regional variation threshold based on the spatial distribution uniformity characterization parameter of the accumulated landslide value at the moment the landslide stops and the rate of change of the accumulated landslide value includes:

[0129] Obtain the duration from the start of sand falling to the stop of the vibrating table, as the sand falling completion time;

[0130] The ratio of the standard deviation to the mean of the cumulative landslide value in each monitoring area is calculated and used as a parameter characterizing the spatial distribution uniformity.

[0131] The ratio of the cumulative value of landslides in the region to the time it takes for the sand to fall is calculated as the rate of change of the cumulative value of landslides in the region.

[0132] When the spatial distribution uniformity characterization parameter is greater than the preset uniformity upper limit threshold, the preset regional variation threshold is reduced;

[0133] When the spatial distribution uniformity characterization parameter is less than the preset uniformity lower limit threshold and the rate of change of the cumulative value of regional collapse is greater than the preset efficiency threshold, the preset regional variation threshold is increased.

[0134] In this embodiment, the preset uniformity upper threshold is the critical value for determining that the difference in the amount of spalling in different areas is too large and the spatial distribution is uneven after sand removal. The preset uniformity lower threshold is the critical value for determining that the spalling uniformity is good. Together, they constitute the evaluation range for sand removal uniformity. The values ​​of the two thresholds depend on the tolerance for sand removal uniformity. In this embodiment, based on the complexity of the casting structure and the requirements for sand removal quality, the initial value of the preset uniformity upper threshold is set to 0.2, and the initial value of the preset uniformity lower threshold is set to 0.1. This allows for automatic optimization of the preset area variation threshold based on the actual uniformity after sand removal, achieving closed-loop adaptive evolution of the sand removal control strategy.

[0135] In this embodiment, the process of adjusting the preset regional variation threshold based on the spatial distribution uniformity characterization parameter of the regional landslide accumulation value at the moment of sandfall cessation and the rate of change of the regional landslide accumulation value is as follows: First, the duration from the start of sandfall to the stop of the vibration table is obtained as the sandfall completion time; then, the ratio of the standard deviation to the mean of the regional landslide accumulation value of each monitoring area is calculated as the spatial distribution uniformity characterization parameter. This parameter reflects the spatial uniformity of the total landslide amount in each monitoring area after sandfall is completed. The smaller the value, the more uniform the landslide. At the same time, the ratio of the overall landslide accumulation value to the sandfall completion time is calculated as the rate of change of the regional landslide accumulation value. This parameter reflects the average speed of sandfall. The larger the value, the higher the sandfall efficiency. When the spatial distribution uniformity characterization parameter is greater than the preset uniformity upper limit threshold, the current preset regional variation threshold is multiplied by the preset first step adjustment coefficient; when the spatial distribution uniformity characterization parameter is less than the preset uniformity lower limit threshold and the rate of change of the regional landslide accumulation value is greater than the preset efficiency threshold, the current preset regional variation threshold is multiplied by the preset second step adjustment coefficient.

[0136] The preset first-step adjustment coefficient and the preset second-step adjustment coefficient are reciprocals of each other, with each adjustment increment being 10%. This ensures that the threshold can effectively respond to changes in sandfall effect while avoiding excessively large single adjustments that could cause drastic threshold fluctuations and system instability. If the adjusted preset region variation threshold is less than 0.05, it is set to 0.05; if it is greater than 0.5, it is set to 0.5, maintaining the threshold within a reasonable range and preventing the system from becoming overly sensitive due to an excessively small threshold or overly sluggish due to an excessively large threshold. Through this adaptive adjustment, the preset region variation threshold is continuously optimized in fixed steps based on the actual sandfall effect, automatically balancing the system between sandfall uniformity and efficiency.

[0137] In this embodiment, the preset first-step adjustment coefficient refers to the step ratio by which the preset region variation threshold is reduced when the sandfall is determined to be uneven, and the preset second-step adjustment coefficient refers to the step ratio by which the preset region variation threshold is increased when the sandfall is determined to be uniform and efficient. Both depend on the convergence speed and system stability. This embodiment comprehensively considers both convergence speed and system stability, setting each adjustment increment to 10%, i.e., the preset first-step adjustment coefficient is 0.9 and the preset second-step adjustment coefficient is 1.1. This ensures that the threshold effectively responds to changes in the sandfall effect while avoiding drastic fluctuations in the threshold that could lead to control instability.

[0138] Understandably, the adjusted preset regional variation threshold is set to 0.05 if it is less than 0.05, and to 0.5 if it is greater than 0.5, to prevent the system from being too sensitive if the threshold is too small or too sluggish if the threshold is too large.

[0139] The spatial uniformity of sandfall is quantified by using the ratio of the standard deviation to the mean of the cumulative sandfall values ​​in each monitoring area after sandfall is completed. The average efficiency of sandfall is quantified by using the ratio of the overall cumulative sandfall value to the sandfall completion time. When the uniformity parameter exceeds the upper threshold, it indicates excessive differences in sandfall progress between areas. By reducing the preset area variation threshold, the consistency judgment criteria for the next sandfall cycle become more stringent, allowing for earlier identification of regional progress differences and initiation of adjustments, thus achieving closed-loop improvement in sandfall uniformity. When the uniformity parameter is below the lower threshold and the efficiency parameter exceeds the efficiency threshold, the judgment criteria are relaxed by increasing the preset area variation threshold, avoiding oversensitivity that leads to unnecessary parameter adjustments, thus achieving adaptive matching between control sensitivity and sandfall efficiency. Through a closed-loop optimization mechanism, the preset area variation threshold continuously evolves with the actual sandfall effect, achieving a dynamic balance between sandfall uniformity and efficiency.

[0140] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive control method for vibration parameters in a lost foam casting process, characterized in that, include: The ultrasonic pulse signals and low-frequency vibration acceleration signals corresponding to each monitoring area in the sand box are acquired in real time during the sand removal process of complex structure castings, based on preset rules, in order to determine the regional collapse characteristics of each monitoring area, and calculate the cumulative regional collapse value of each monitoring area based on the regional collapse characteristics. The regional variation coefficient is determined based on the statistical characteristics of the cumulative value of regional landslides, and the sandfall conditions in each monitoring area are determined to be consistent based on the comparison results between the regional variation coefficient and the preset regional variation threshold. Based on the consistent progress of the regions, the total weight of the castings in the sand box is obtained, and the overall cumulative value of collapse is determined according to the cumulative value of collapse in each region. The structural acoustic-vibration coupling coefficient is calculated by combining the total weight of the castings and the low-frequency vibration acceleration signal within a preset first time period. Based on the threshold comparison result of the structural acoustic-vibration coupling coefficient, the sand falling condition is determined to be dead angle jamming or internal cavity imbalance. Based on the aforementioned sandfall conditions, the current vibration parameters of the vibration table are obtained, and a compensation gain strategy is determined in conjunction with the aforementioned sandfall conditions to generate a set of correction values ​​for the current vibration parameters. The set of corrected vibration parameters is determined based on the set of corrected values ​​to control the vibration table to perform sand dropping. After the controlled vibration table completes the sandfall, the preset regional variation threshold is adjusted based on the spatial distribution uniformity characterization parameter of the accumulated collapse value in the region at the moment the sandfall stops and the rate of change of the accumulated collapse value in the region.

2. The adaptive control method for vibration parameters in the lost foam casting process according to claim 1, characterized in that, The process of determining the regional landslide characteristics of each monitoring area includes: Within a preset second time period, envelope detection is performed on the ultrasonic pulse signal to extract the pulse echo amplitude sequence. When the pulse echo amplitude is lower than a preset departure threshold, it is determined that a collapse event has occurred in the monitored area. Within the same preset second time period, the low-frequency vibration acceleration signal is integrated in the time domain to obtain the vibration velocity signal. When the energy decay rate of the vibration velocity signal exceeds a preset rate threshold, the collapse event is determined to be a valid collapse event. The effective landslide events in each monitoring area within the preset second time period are arranged in chronological order to obtain the regional landslide characteristics of the monitoring area.

3. The adaptive control method for vibration parameters in the lost foam casting process according to claim 2, characterized in that, The process of calculating the cumulative regional landslide value for each monitoring area based on regional landslide characteristics includes: The number of occurrences of the effective landslide events in the landslide characteristics of the region is counted to obtain the cumulative landslide frequency value; Based on the determination time of the effective collapse event, the pulse echo amplitude at the same moment is extracted from the ultrasonic pulse signal and used as the single collapse energy characterization value of the effective collapse event. The cumulative collapse energy value is obtained by accumulating the single collapse energy characterization values ​​of all the effective collapse events within the preset second time period; The cumulative value of landslide frequency and the cumulative value of landslide energy are weighted and fused to obtain the regional landslide cumulative value of the monitored area.

4. The adaptive control method for vibration parameters in the lost foam casting process according to claim 3, characterized in that, The process of determining the regional coefficient of variation based on the statistical characteristics of the cumulative value of landslides in the region includes: Calculate the average cumulative collapse value of each monitored area within the preset second time period; The mean of the sum of squared deviations of the cumulative landslide value in each monitoring area from the average value is calculated to obtain the cumulative landslide standard deviation. The coefficient of variation for the region is obtained based on the cumulative standard deviation of the landslide and the cumulative average value of the landslide.

5. The adaptive control method for vibration parameters in the lost foam casting process according to claim 4, characterized in that, The process of determining whether the sandfall conditions in each monitoring area are consistent based on the comparison between the regional coefficient of variation and the preset regional variation threshold includes: When the coefficient of variation of the region is less than or equal to the preset regional variation threshold, the sandfall condition is determined to be a region with consistent progress.

6. The adaptive control method for vibration parameters in the lost foam casting process according to claim 5, characterized in that, The process of calculating the acoustic-vibration coupling coefficient of a structure includes: The effective vibration energy integral is obtained by integrating the square of the low-frequency vibration acceleration signal within the preset first time period. Divide the total cumulative value of overall spalling by the total weight of the casting to obtain the cumulative value of spalling per unit weight; The acoustic-vibration coupling coefficient of the structure is obtained by multiplying the cumulative value of collapse per unit weight by the integral of the effective vibration energy.

7. The adaptive control method for vibration parameters in the lost foam casting process according to claim 6, characterized in that, The process of determining whether a sand-falling condition is a dead-angle jam or an internal cavity imbalance based on the threshold comparison results of the structural acoustic-vibration coupling coefficient includes: When the acoustic-vibration coupling coefficient of the structure is less than or equal to the preset first coupling threshold, the sandfall condition is determined to be a dead angle jam. When the acoustic-vibration coupling coefficient of the structure is greater than or equal to the preset second coupling threshold, the sandfall condition is determined to be an internal cavity imbalance. The preset first coupling threshold is less than the preset second coupling threshold.

8. The adaptive control method for vibration parameters in the lost foam casting process according to claim 7, characterized in that, The process of determining the compensation gain strategy includes: When the sandfall condition is a dead zone jam, the monitoring area where the cumulative value of regional collapse is lower than the preset collapse progress threshold is determined as the target monitoring area. The amplitude gain coefficient and frequency gain coefficient are determined based on the difference between the cumulative value of regional collapse in the target monitoring area and the preset collapse progress threshold. When the sandfall condition is an internal cavity imbalance, the waveform adjustment coefficient and the excitation force adjustment coefficient are determined based on the distribution characteristics of the cumulative value of the regional collapse in each monitoring area.

9. The adaptive control method for vibration parameters in the lost foam casting process according to claim 8, characterized in that, The process of generating the correction set for the current vibration parameters includes: Obtain the current vibration parameters of the vibration table, including the current amplitude, current frequency, current waveform parameters, and current excitation force; When the sandfall condition is a dead angle jam, the amplitude gain coefficient is multiplied by the current amplitude to obtain the amplitude increase correction amount, the frequency gain coefficient is multiplied by the current frequency to obtain the frequency offset correction amount, and the amplitude increase correction amount and the frequency offset correction amount are combined to generate the correction amount set; When the sand-falling condition is an internal cavity imbalance, the waveform adjustment coefficient is superimposed with the current waveform parameter to obtain the waveform parameter correction amount, the excitation force adjustment coefficient is multiplied with the current excitation force to obtain the excitation force correction amount, and the waveform parameter correction amount and the excitation force correction amount are combined to generate the correction amount set.

10. The adaptive control method for vibration parameters in the lost foam casting process according to claim 9, characterized in that, The process of adjusting the preset regional variation threshold based on the spatial distribution uniformity characterization parameter of the accumulated landslide value at the moment the landslide stops and the rate of change of the accumulated landslide value includes: Obtain the duration from the start of sand falling to the stop of the vibrating table, as the sand falling completion time; The ratio of the standard deviation to the mean of the cumulative landslide value in each monitoring area is calculated and used as a parameter characterizing the spatial distribution uniformity. The ratio of the cumulative value of landslides in the region to the time it takes for the sand to fall is calculated as the rate of change of the cumulative value of landslides in the region. When the spatial distribution uniformity characterization parameter is greater than the preset uniformity upper limit threshold, the preset regional variation threshold is reduced; When the spatial distribution uniformity characterization parameter is less than the preset uniformity lower limit threshold and the rate of change of the cumulative value of regional collapse is greater than the preset efficiency threshold, the preset regional variation threshold is increased.

Images