Pouring temperature adjusting method and system combined with multi-dimensional sensor

Abstract

The invention discloses a pouring temperature adjusting method and system combined with a multi-dimensional sensor, and relates to the technical field of pouring temperature control. The method comprises the following steps: performing temperature control key anchor point analysis based on a historical casting quality detection record set to obtain a temperature control abnormal anchor point set; obtaining an infrared sensor array and a thermocouple temperature sensor array; performing pouring temperature time sequence acquisition according to a preset monitoring frequency to obtain an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; performing double analysis on temperature distribution uniformity and temperature fluctuation, and determining a temperature regulation and control scheme; and the pouring temperature of the target thin-wall automobile casting is dynamically regulated and controlled based on the temperature regulation and control scheme. The technical problem that in the prior art, the casting quality is unstable due to the fact that the pouring temperature of a thin-wall automobile casting is not accurately regulated and controlled is solved, accurate pouring temperature regulation and control are achieved through multi-sensor cooperative monitoring and temperature control key anchor point dynamic adjustment, and the technical effect of improving the casting quality stability is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pouring temperature control, in particular to a pouring temperature regulation method and system combined with multi-dimensional sensors. BACKGROUND

[0002] Under the trend of automobile lightweight development, thin-walled automobile castings are widely used in key parts such as vehicle body, engine and chassis due to their light weight, high strength and complex structure. However, due to the small wall thickness and fast cooling speed of thin-walled castings, they are easily affected by temperature fluctuations and uneven temperature distribution during pouring, resulting in the occurrence of casting defects such as shrinkage, porosity and cracks. In the prior art, a single thermocouple sensor is often used for pouring temperature monitoring. Limited by the limited measuring points and insufficient response speed, it is difficult to timely and comprehensively reflect the real temperature field of the pouring process, resulting in insufficient precision and dynamics of temperature regulation, and further causing poor quality stability and low yield of castings. SUMMARY

[0003] The present application provides a pouring temperature regulation method and system combined with multi-dimensional sensors, which solves the technical problem of inaccurate pouring temperature regulation of thin-walled automobile castings in the prior art, resulting in unstable casting quality.

[0004] In a first aspect, the present application provides a pouring temperature regulation method combined with multi-dimensional sensors, which comprises: Obtaining a set of historical casting quality detection records of a target thin-walled automobile casting, performing temperature control key anchor point analysis based on the set of historical casting quality detection records, and obtaining a set of temperature control abnormal anchor points; traversing the set of temperature control abnormal anchor points to arrange infrared sensors on the casting, and arranging thermocouple temperature sensors at the gates, risers and bosses to obtain an infrared sensor array and a thermocouple temperature sensor array; traversing the infrared sensor array and the thermocouple temperature sensor array to perform pouring temperature time sequence acquisition according to a preset monitoring frequency, and obtaining an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; performing dual analysis of temperature distribution uniformity and temperature fluctuation based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine a temperature regulation scheme; and dynamically regulating the pouring temperature of the target thin-walled automobile casting based on the temperature regulation scheme.

[0005] In a second aspect, the present application provides a pouring temperature regulation system combined with multi-dimensional sensors, which comprises: The historical data acquisition module acquires a historical casting quality detection record set of the target thin-walled automobile casting, performs temperature control key anchor point analysis based on the historical casting quality detection record set, and obtains a temperature control abnormal anchor point set; the sensor layout planning module iterates through the temperature control abnormal anchor point set to perform infrared sensor layout on the casting, and arranges thermocouple temperature sensors at the gates, risers and bosses to obtain an infrared sensor array and a thermocouple temperature sensor array; the temperature data acquisition module iterates through the infrared sensor array and the thermocouple temperature sensor array to perform pouring temperature time sequence acquisition according to a preset monitoring frequency, and obtains an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; the temperature analysis and decision module performs dual analysis of temperature distribution uniformity and temperature fluctuation based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, and determines a temperature control scheme; and the pouring temperature control module performs dynamic control on the pouring temperature of the target thin-walled automobile casting based on the temperature control scheme.

[0006] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: First, the historical casting quality detection record set of the target thin-walled automobile casting is acquired, and temperature control key anchor point analysis is performed based on the historical casting quality detection record set to obtain a temperature control abnormal anchor point set. Then, the infrared sensor layout is performed on the casting by iterating through the temperature control abnormal anchor point set, and the thermocouple temperature sensors are arranged at the gates, risers and bosses to obtain an infrared sensor array and a thermocouple temperature sensor array. Further, the pouring temperature time sequence acquisition is performed according to a preset monitoring frequency by iterating through the infrared sensor array and the thermocouple temperature sensor array, and an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array are obtained. Then, dual analysis of temperature distribution uniformity and temperature fluctuation is performed based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine a temperature control scheme. Finally, the pouring temperature of the target thin-walled automobile casting is dynamically controlled based on the temperature control scheme. The technical problem of inaccurate pouring temperature control of the thin-walled automobile casting in the prior art, which leads to unstable casting quality, is solved. The pouring temperature is accurately controlled through multi-sensor collaborative monitoring and dynamic adjustment of temperature control key anchors, and the technical effect of improving the stability of casting quality is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0008] Figure 1A schematic flowchart of a pouring temperature regulation method incorporating a multi-dimensional sensor, provided in an embodiment of this application; Figure 2 This is a schematic diagram of the pouring temperature control system combined with a multi-dimensional sensor, provided in an embodiment of this application.

[0009] Figure labeling: Historical data acquisition module 11, sensor deployment planning module 12, temperature data acquisition module 13, temperature analysis and decision-making module 14, casting temperature control module 15. Detailed Implementation

[0010] This application provides a method and system for adjusting the pouring temperature by combining multi-dimensional sensors, which solves the technical problem of inaccurate pouring temperature control in thin-walled automotive castings, leading to unstable casting quality.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0012] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0013] Example 1, as Figure 1 As shown, this application provides a method for regulating casting temperature by incorporating multi-dimensional sensors, wherein the method includes: Obtain a set of historical casting quality inspection records for the target thin-walled automotive casting, and perform temperature control key anchor point analysis based on the set of historical casting quality inspection records to obtain a set of temperature control anomaly anchor points.

[0014] In this embodiment, a set of historical casting quality inspection records for the target thin-walled automotive casting during its previous production process is obtained. This set of historical casting quality inspection records includes appearance inspection data, internal defect inspection data, and corresponding defect location and type information for castings produced under different batches and pouring conditions. Based on this set of historical casting quality inspection records, a set of temperature control critical anchor points is obtained through temperature control anomaly anchor point analysis.

[0015] Further, a set of historical casting quality detection records of the target thin-walled automobile casting is acquired, and a temperature control key anchor point analysis is performed based on the set of historical casting quality detection records to obtain a set of temperature control abnormal anchor points, including: Data extraction is performed on the set of historical casting quality detection records according to the abnormal type and the abnormal position to obtain a set of historical abnormal type-abnormal position; and the set of historical abnormal type-abnormal position is aggregated in the same category with the same historical abnormal type as the index to determine a plurality of sets of historical abnormal type-abnormal position in the same category; and the set of temperature control abnormal anchor points is obtained by performing a temperature control key anchor point analysis on the plurality of sets of historical abnormal type-abnormal position in the same category with the abnormal position as the analysis object.

[0016] Specifically, first, the set of historical casting quality detection records is parsed, and data extraction is performed on the detection data therein according to two dimensions of abnormal type and abnormal position to associate the abnormal type (such as shrinkage cavity, gas hole, crack, etc.) marked in each detection result with the corresponding abnormal position coordinates to form a set of historical abnormal type-abnormal position. Subsequently, the set of historical abnormal type-abnormal position is aggregated in the same category with the same historical abnormal type as the index to aggregate a plurality of records belonging to the same abnormal type into the same set, thereby determining a plurality of sets of historical abnormal type-abnormal position in the same category. Further, the set of temperature control abnormal anchor points is obtained by performing a temperature control key anchor point analysis on the plurality of sets of historical abnormal type-abnormal position in the same category with the abnormal position as the analysis object, and the key positions with higher temperature sensitivity and larger abnormal concentration are identified by comprehensively calculating the aggregation, frequency, and corresponding relationship with the key regions of the casting pouring process (such as the thin-walled region, the runner connection, the periphery of the riser, etc.) of the abnormal position in the spatial distribution.

[0017] Further, the set of temperature control abnormal anchor points is obtained by performing a temperature control key anchor point analysis on the plurality of sets of historical abnormal type-abnormal position in the same category with the abnormal position as the analysis object, including: A first set of historical abnormal type-abnormal position in the same category is extracted from the plurality of sets of historical abnormal type-abnormal position in the same category; a sliding is performed on the first set of historical abnormal type-abnormal position in the same category according to a preset sliding window, the number of abnormal positions in each sliding sub-region window is counted, and a set of sliding sub-region window abnormal position numbers is obtained; a sliding sub-region window corresponding to a maximum value in the set of sliding sub-region window abnormal position numbers is taken as a first initial anchor sliding sub-region window; a coordinate mean value screening of the abnormal position is performed in the first initial anchor sliding sub-region window to determine a first temperature control key anchor point; and the first temperature control key anchor point is added to the set of temperature control abnormal anchor points.

[0018] Specifically, first, a first same-type historical abnormality type-abnormality location set is extracted from a plurality of same-type historical abnormality type-abnormality location sets, the set containing a plurality of historical abnormality location data under the same abnormality type, wherein the first same-type historical abnormality type-abnormality location set does not represent an order and represents any one of the plurality of same-type historical abnormality type-abnormality location sets. Subsequently, a preset sliding window is used to slide step by step in the spatial coordinate interval corresponding to the first same-type historical abnormality type-abnormality location set, each sliding forming a sliding sub-region window, and the number of abnormality locations in the sliding sub-region window is counted, thereby obtaining a sliding sub-region window abnormality location number set. By comparing the values in the sliding sub-region window abnormality location number set, the sliding sub-region window corresponding to the maximum value is identified as the first initial anchor sliding sub-region window. Then, the coordinates of all abnormality locations contained in the first initial anchor sliding sub-region window are subjected to mean value screening, that is, the weighted average of all abnormality location coordinates is taken as the screening result, thereby determining the most representative temperature control key location in the region as the first temperature control key anchor. Finally, the first temperature control key anchor is added to the temperature control abnormality anchor set to form a comprehensive temperature control anchor set for multiple abnormality types and multiple abnormality locations, which is used for subsequent sensor layout and pouring temperature dynamic regulation.

[0019] Further, the coordinate screening of the abnormality locations in the first initial anchor sliding sub-region window to determine the first temperature control key anchor includes: An abnormality location is randomly selected from the first initial anchor sliding sub-region window as a coordinate screening starting point, and mean shift screening is performed on the coordinate screening starting point in the first initial anchor sliding sub-region window according to one-twentieth of the preset sliding window length, to determine the first temperature control key anchor.

[0020] Specifically, an abnormality location is randomly selected from the first initial anchor sliding sub-region window as a coordinate screening starting point, which serves as an initial center point for mean shift iteration. Subsequently, a neighborhood search is performed on other abnormality location coordinates within the first initial anchor sliding sub-region window from the coordinate screening starting point according to one-twentieth of the preset sliding window length as a step, and the mean value of all abnormality location coordinates in the neighborhood is calculated and taken as a new center point. Then, based on the updated center point, iterative mean value calculation and drift updating are continued within the same neighborhood range until the coordinate difference between the previous and subsequent center points is less than a preset convergence threshold or the maximum iteration number is reached. Finally, the converged center point coordinates are the first temperature control key anchor, which can represent the optimal aggregation point of the abnormality locations in the initial anchor sliding sub-region window.

[0021] Infrared sensors are arranged on the casting by traversing the temperature control abnormal anchor point set, and thermocouple temperature sensors are arranged at the gates, risers and bosses to obtain an infrared sensor array and a thermocouple temperature sensor array.

[0022] After obtaining the temperature control abnormal anchor point set, sensors are arranged on the target thin-walled automobile casting based on the temperature control abnormal anchor point set. Specifically, by traversing the temperature control abnormal anchor point set, each anchor point position is taken as a layout priority point, and an infrared sensor is arranged on the corresponding casting surface, so that the infrared sensor can monitor the temperature of the anchor point area in a non-contact manner, and thus an infrared sensor arrangement scheme covering the key abnormal areas of the casting is formed. On this basis, contact thermocouple temperature sensors are further arranged at typical temperature control key positions of the casting, including gate positions, riser positions and boss positions, so as to obtain real-time change data of the internal metal liquid temperature of the casting. Through the above arrangement, an infrared sensor array and a thermocouple temperature sensor array can be formed respectively, wherein the infrared sensor array is mainly used for monitoring the dynamic change of the surface temperature distribution of the casting, and the thermocouple temperature sensor array is mainly used for obtaining temperature data of the temperature conduction and solidification process in the casting, which complement each other, thereby realizing multi-dimensional temperature monitoring system in the pouring process of the thin-walled automobile casting and providing complete data basis for subsequent temperature distribution analysis and dynamic regulation.

[0023] Infrared sensor array and the thermocouple temperature sensor array are traversed to perform pouring temperature time sequence collection according to a preset monitoring frequency to obtain an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array.

[0024] After the arrangement of the infrared sensor array and the thermocouple temperature sensor array is completed, the temperature evolution of the target thin-walled automobile casting during the pouring process is collected in time sequence. Specifically, the infrared sensor array and the thermocouple temperature sensor array are traversed, and the temperature data of each sensor node is collected periodically according to a preset monitoring frequency. The infrared sensor array obtains the instantaneous temperature value of the key position of the casting surface in a non-contact manner, and stores the temperature value in time sequence as an infrared monitoring temperature value sequence; the thermocouple temperature sensor array obtains the temperature change data of the internal key positions such as the gate, riser and boss of the casting through contact temperature measurement, and records the data as a thermocouple monitoring temperature value sequence at the same monitoring frequency. With the continuous pouring process, the outputs of all sensors are sequentially collected according to the time dimension to form an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array, respectively.

[0025] Based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, temperature distribution uniformity and temperature fluctuation are analyzed to determine a temperature regulation scheme.

[0026] After obtaining the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, the system performs dual analysis of temperature distribution uniformity and temperature fluctuation based on the two types of temperature data, to determine the temperature regulation scheme of the target thin-walled automobile casting.

[0027] Further, based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, the dual analysis of temperature distribution uniformity and temperature fluctuation is performed to determine the temperature regulation scheme, which includes: The temperature mean values of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array are calculated respectively to determine the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array; the temperature distribution uniformity analysis is performed on the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array to determine the first temperature regulation factor; the temperature fluctuation analysis is performed on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively to determine the second temperature regulation factor; and the temperature regulation scheme is identified according to the first temperature regulation factor and the second temperature regulation factor to determine the temperature regulation scheme.

[0028] First, the mean value calculation is performed on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively in the entire pouring period to obtain the temperature mean values of different sensor nodes, and then the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array are constructed to represent the average temperature distribution of the casting surface and the interior. Subsequently, the temperature distribution uniformity analysis is performed on the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array, for example, by calculating the mean square deviation, the coefficient of variation or the normalized temperature difference index between nodes to evaluate the consistency of the overall temperature distribution, thereby obtaining the first temperature regulation factor reflecting the spatial temperature uniformity. Then, the temperature fluctuation analysis is performed on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively, for example, by sliding window variance analysis, frequency domain power spectrum analysis or time series stability discrimination method to extract temperature fluctuation characteristics, thereby obtaining the second temperature regulation factor reflecting the stability of temperature change over time. Finally, the system calls the preset temperature control rule library or intelligent recognition model according to the numerical size and combination relationship of the first temperature regulation factor and the second temperature regulation factor, performs temperature regulation scheme identification, and then determines the temperature regulation scheme.

[0029] Further, the temperature distribution uniformity analysis is performed on the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array to determine the first temperature regulation factor, which includes: The infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array are respectively subjected to intra-array temperature distribution fluctuation analysis to determine an infrared monitoring temperature distribution uniformity coefficient and a thermocouple monitoring temperature distribution uniformity coefficient; the infrared monitoring temperature distribution uniformity coefficient and the thermocouple monitoring temperature distribution uniformity coefficient are subjected to weighted analysis to determine the first temperature regulation factor.

[0030] The infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array are respectively subjected to intra-array temperature distribution fluctuation analysis. Specifically, the temperature mean values of each monitoring point in the infrared monitoring temperature mean value array are subjected to difference degree calculation, for example, in the form of variance, standard deviation or coefficient of variation, to obtain an infrared monitoring temperature distribution uniformity coefficient; similarly, the temperature mean values of each monitoring point in the thermocouple monitoring temperature mean value array are subjected to the same fluctuation analysis to obtain a thermocouple monitoring temperature distribution uniformity coefficient. Subsequently, the infrared monitoring temperature distribution uniformity coefficient and the thermocouple monitoring temperature distribution uniformity coefficient are subjected to weighted analysis, wherein the weights can be preset or dynamically adjusted according to the sensitivity of infrared monitoring in surface temperature distribution monitoring and the accuracy of thermocouple monitoring in internal temperature measurement. Through weighted summation or normalized weighted average, the final first temperature regulation factor is obtained, which can comprehensively reflect the overall temperature distribution uniformity level of the target thin-walled automobile casting during the pouring process.

[0031] Further, the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array are respectively subjected to intra-sequence temperature fluctuation analysis to determine a second temperature regulation factor, including: The temperature fluctuation feature recognizer is called to perform intra-sequence temperature fluctuation feature recognition on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively to obtain an infrared monitoring temperature fluctuation feature array and a thermocouple monitoring temperature fluctuation array; the infrared monitoring temperature fluctuation feature array and the thermocouple monitoring temperature fluctuation array are respectively subjected to feature interaction enhancement to obtain infrared monitoring temperature fluctuation enhanced features and thermocouple monitoring temperature fluctuation enhanced features; and the second temperature regulation factor is determined according to the infrared monitoring temperature fluctuation enhanced features and the thermocouple monitoring temperature fluctuation enhanced features.

[0032] The preset temperature fluctuation feature recognizer is called to perform time sequence feature extraction on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively. The temperature fluctuation feature recognizer can identify temperature fluctuation features from the time sequence based on a statistical model (such as a sliding window variance, a root mean square deviation), a frequency domain model (such as a fast Fourier transform to extract a frequency spectrum energy), or a deep learning model (such as a convolutional neural network, a recurrent neural network, etc.), and finally obtain an infrared monitoring temperature fluctuation feature array and a thermocouple monitoring temperature fluctuation feature array. Subsequently, the infrared monitoring temperature fluctuation feature array and the thermocouple monitoring temperature fluctuation feature array are respectively subjected to feature interaction enhancement. Specifically, the correlation between different time segments or different monitoring points is introduced into the calculation through similarity recognition and feature interaction matrix construction, and the infrared monitoring temperature fluctuation enhanced feature and the thermocouple monitoring temperature fluctuation enhanced feature are obtained by using the methods of graph convolution, attention weighting or feature fusion. Finally, according to the infrared monitoring temperature fluctuation enhanced feature and the thermocouple monitoring temperature fluctuation enhanced feature, a joint evaluation is performed, for example, a weighted average, a normalized fusion or a machine learning discriminant model is used, to obtain a second temperature regulation factor that comprehensively reflects the stability of the temperature change of the casting with time. The factor can accurately represent the temperature fluctuation level of the target thin-walled automobile casting during the whole pouring process.

[0033] Further, comprising: The similarity of any two infrared monitoring temperature fluctuation features in the infrared monitoring temperature fluctuation feature array is identified, and a feature interaction enhancement matrix is constructed according to the identification result. The graph convolution operation is performed on any two infrared monitoring temperature fluctuation features by using the feature interaction enhancement matrix to obtain an initial infrared monitoring temperature fluctuation enhanced feature. The similarity of any two infrared monitoring temperature fluctuation features in the infrared monitoring temperature fluctuation feature array is identified again, and when the identification number meets the preset number, a set of initial infrared monitoring temperature fluctuation enhanced features is obtained. The mean value of the set of initial infrared monitoring temperature fluctuation enhanced features is calculated to obtain the infrared monitoring temperature fluctuation enhanced feature.

[0034] First, similarity recognition is performed on any two infrared monitoring temperature fluctuation features in the infrared monitoring temperature fluctuation feature array respectively. The similarity can be calculated in the form of Euclidean distance, cosine similarity or Pearson correlation coefficient. The similarity recognition result is normalized, and the processed value is filled into the initially empty feature interaction enhancement matrix according to the corresponding feature index position, so as to establish the correlation between the features. Then, graph convolution operation is performed on any two infrared monitoring temperature fluctuation features by using the feature interaction enhancement matrix, the correlation information of adjacent features is aggregated through convolution, and the initial infrared monitoring temperature fluctuation enhancement feature is obtained, so as to realize the interaction enhancement between local features. Next, similarity recognition is performed on any two infrared monitoring temperature fluctuation features in the infrared monitoring temperature fluctuation feature array again, and the similarity recognition and graph convolution calculation are executed in a loop. When the number of similarity recognition meets the preset number, a set of initial infrared monitoring temperature fluctuation enhancement features is output, forming an initial infrared monitoring temperature fluctuation enhancement feature set. Finally, the initial infrared monitoring temperature fluctuation enhancement feature set is subjected to mean value calculation to eliminate the influence of local differences, so as to obtain the infrared monitoring temperature fluctuation enhancement feature representing the overall temperature fluctuation enhancement effect.

[0035] Further, comprising: According to the size of the first temperature regulation factor and the second temperature regulation factor, a first temperature monitoring frequency and a second temperature monitoring frequency are configured; the first temperature monitoring frequency is distributed to the infrared sensor array, and the second temperature monitoring frequency is distributed to the thermocouple temperature sensor array; and based on the infrared sensor array and the thermocouple temperature sensor array after the temperature monitoring frequency is updated, the target thin-walled automobile casting is subjected to pouring temperature monitoring.

[0036] When the first temperature regulation factor value is large, it indicates that the temperature uniformity of the casting in the spatial distribution is poor, at this time, the sampling frequency of the infrared sensor is preferentially increased, so as to capture the dynamic change of the casting surface temperature distribution with higher resolution; when the second temperature regulation factor value is large, it indicates that the temperature fluctuation of the casting in the time sequence evolution process is significant, at this time, the sampling frequency of the thermocouple temperature sensor is preferentially increased, so as to enhance the capture ability of the internal temperature transient change. Through the above adjustment, the first temperature monitoring frequency and the second temperature monitoring frequency are obtained, and the first temperature monitoring frequency is distributed to the infrared sensor array and the second temperature monitoring frequency is distributed to the thermocouple temperature sensor array, so that the two sensor arrays can operate cooperatively at different monitoring frequencies. Based on the updated temperature monitoring frequency, the infrared sensor array and the thermocouple temperature sensor array perform pouring temperature monitoring on the target thin-walled automobile casting, realize synchronous and differential sampling of the surface temperature field and the internal temperature field, and ensure that the obtained temperature data has spatial distribution coverage and time resolution advantage, thereby providing fine data support for dynamic temperature regulation.

[0037] The pouring temperature of the target thin-walled automobile casting is dynamically regulated based on the temperature regulation scheme.

[0038] The system adjusts the temperature control unit in real time during the pouring process according to the regulation instruction output in the temperature regulation scheme, including controlling the heat preservation and heating device of the molten metal, the cooling air duct device, and the start-stop and power size of the sprue compensation heating link. When the temperature distribution uniformity is insufficient, the temperature of the low-temperature area is increased by increasing the heating power or prolonging the heat preservation time, so as to make it consistent with the overall temperature field; when the temperature fluctuation is too large, the transient temperature change is balanced by reducing the metal liquid flow rate, opening the cooling air duct, or adjusting the heat dissipation rate of the riser, so as to suppress the rapid temperature fluctuation. The above regulation process is carried out in a closed loop manner throughout the pouring period, that is, the temperature data fed back by the sensor are obtained in real time and compared with the regulation scheme, and the control amount is continuously corrected according to the deviation, so as to realize dynamic and stable control of the pouring temperature of the target thin-walled automobile casting.

[0039] In summary, the embodiments of the present application have at least the following technical effects: First, a historical casting quality detection record set of the target thin-walled automobile casting is obtained, and a temperature control key anchor point analysis is performed based on the historical casting quality detection record set to obtain a temperature control abnormal anchor point set. Then, the casting is laid out with infrared sensors by traversing the temperature control abnormal anchor point set, and thermocouple temperature sensors are laid out at the sprue, riser and boss to obtain an infrared sensor array and a thermocouple temperature sensor array. Further, the infrared sensor array and the thermocouple temperature sensor array are traversed to perform pouring temperature time sequence acquisition according to a preset monitoring frequency to obtain an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array. Then, temperature distribution uniformity and temperature fluctuation are analyzed based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine a temperature regulation scheme. Finally, the pouring temperature of the target thin-walled automobile casting is dynamically regulated based on the temperature regulation scheme. The technical problem of inaccurate pouring temperature regulation of the thin-walled automobile casting in the prior art, which leads to unstable casting quality, is solved. The pouring temperature is accurately regulated through multi-sensor cooperative monitoring and dynamic adjustment of the temperature control key anchor point, and the technical effect of improving the stability of the casting quality is achieved.

[0040] Embodiment two, based on the same inventive concept as the pouring temperature regulation method combined with multi-dimensional sensors in the foregoing embodiments, as shown in Figure 2 The present application provides a pouring temperature regulation system combined with multi-dimensional sensors, wherein the system comprises: The historical data acquisition module 11 acquires a historical casting quality detection record set of the target thin-walled automobile casting, performs temperature control key anchor point analysis based on the historical casting quality detection record set, and obtains a temperature control abnormal anchor point set; the sensor layout planning module 12 performs infrared sensor layout on the casting by traversing the temperature control abnormal anchor point set, and arranges thermocouple temperature sensors at the gates, risers and bosses, and obtains an infrared sensor array and a thermocouple temperature sensor array; the temperature data acquisition module 13 performs pouring temperature time sequence acquisition according to a preset monitoring frequency by traversing the infrared sensor array and the thermocouple temperature sensor array, and obtains an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; the temperature analysis and decision module 14 performs dual analysis of temperature distribution uniformity and temperature fluctuation based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, and determines a temperature control scheme; and the pouring temperature control module 15 dynamically controls the pouring temperature of the target thin-walled automobile casting based on the temperature control scheme.

[0041] Further, the historical data acquisition module 11 is configured to perform the following method: The historical casting quality detection record set is subjected to data extraction according to abnormal types and abnormal positions, and a historical abnormal type-abnormal position set is obtained; the historical abnormal type-abnormal position set is subjected to same type aggregation with the same historical abnormal type as an index, and a plurality of same type historical abnormal type-abnormal position sets are determined; the plurality of same type historical abnormal type-abnormal position sets are subjected to temperature control key anchor point analysis with abnormal positions as analysis objects, and the temperature control abnormal anchor point set is obtained.

[0042] Further, the historical data acquisition module 11 is configured to perform the following method: A first same type historical abnormal type-abnormal position set is extracted from the plurality of same type historical abnormal type-abnormal position sets; the first same type historical abnormal type-abnormal position set is subjected to sliding according to a preset sliding window, and the number of abnormal positions in each sliding sub-region window is counted, and a sliding sub-region window abnormal position number set is obtained; a sliding sub-region window corresponding to a maximum value in the sliding sub-region window abnormal position number set is taken as a first initial anchor point sliding sub-region window; coordinate mean filtering of abnormal positions is performed in the first initial anchor point sliding sub-region window, and a first temperature control key anchor point is determined; and the first temperature control key anchor point is added to the temperature control abnormal anchor point set.

[0043] Further, the historical data acquisition module 11 is configured to perform the following method: Randomly selecting an abnormal position from the first initial anchor point sliding sub-region window as a coordinate screening starting point, performing mean shift screening on the coordinate screening starting point in the first initial anchor point sliding sub-region window according to one-twentieth of the preset sliding window length, and determining a first temperature control key anchor point.

[0044] Further, the temperature analysis and decision module 14 is configured to perform the following method: respectively calculating temperature means of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine an infrared monitoring temperature mean array and a thermocouple monitoring temperature mean array, performing temperature distribution uniformity analysis on the infrared monitoring temperature mean array and the thermocouple monitoring temperature mean array to determine a first temperature control factor, respectively performing intra-sequence temperature fluctuation analysis on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine a second temperature control factor, and performing temperature control scheme identification according to the first temperature control factor and the second temperature control factor to determine the temperature control scheme.

[0045] Further, the temperature analysis and decision module 14 is configured to perform the following method: respectively performing intra-array temperature distribution fluctuation analysis on the infrared monitoring temperature mean array and the thermocouple monitoring temperature mean array to determine an infrared monitoring temperature distribution uniformity coefficient and a thermocouple monitoring temperature distribution uniformity coefficient, and performing weighted analysis on the infrared monitoring temperature distribution uniformity coefficient and the thermocouple monitoring temperature distribution uniformity coefficient to determine the first temperature control factor.

[0046] Further, the temperature analysis and decision module 14 is configured to perform the following method: respectively performing intra-sequence temperature fluctuation feature identification on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array by calling a temperature fluctuation feature identifier to obtain an infrared monitoring temperature fluctuation feature array and a thermocouple monitoring temperature fluctuation array, respectively performing feature interaction enhancement on the infrared monitoring temperature fluctuation feature array and the thermocouple monitoring temperature fluctuation array to obtain infrared monitoring temperature fluctuation enhanced features and thermocouple monitoring temperature fluctuation enhanced features, and determining the second temperature control factor according to the infrared monitoring temperature fluctuation enhanced features and the thermocouple monitoring temperature fluctuation enhanced features.

[0047] Further, the temperature analysis and decision module 14 is configured to perform the following method: Similarity recognition is performed on any two infrared monitoring temperature fluctuation features in the array of infrared monitoring temperature fluctuation features, and a feature interaction enhancement matrix is constructed according to a recognition result; graph convolution operation is performed on any two infrared monitoring temperature fluctuation features by using the feature interaction enhancement matrix, to obtain initial infrared monitoring temperature fluctuation enhancement features; similarity recognition is again performed on any two infrared monitoring temperature fluctuation features in the array of infrared monitoring temperature fluctuation features, and when the number of times of recognition meets a preset number of times, an initial infrared monitoring temperature fluctuation enhancement feature set is obtained; a mean value of the initial infrared monitoring temperature fluctuation enhancement feature set is calculated, to obtain the infrared monitoring temperature fluctuation enhancement features.

[0048] Further, the temperature analysis decision module 14 is configured to perform the following method: According to the sizes of the first temperature regulation factor and the second temperature regulation factor, a first temperature monitoring frequency and a second temperature monitoring frequency are configured; the first temperature monitoring frequency is distributed to the infrared sensor array, and the second temperature monitoring frequency is distributed to the thermocouple temperature sensor array; and based on the infrared sensor array and the thermocouple temperature sensor array after the temperature monitoring frequency is updated, pouring temperature monitoring is performed on the target thin-walled automobile casting.

[0049] It should be noted that the above sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes a specific embodiment of the present application. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0050] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

[0051] The present specification and drawings are merely exemplary descriptions of the present application, and should be considered to cover any and all modifications, changes, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method of adjusting a casting temperature by using a multi-dimensional sensor, characterized by, The method comprises: obtaining a historical casting quality detection record set of a target thin-walled automobile casting, performing warm control key anchor point analysis based on the historical casting quality detection record set, and obtaining a warm control abnormal anchor point set; traversing the warm control abnormal anchor point set to arrange infrared sensors for the casting, and arranging thermocouple temperature sensors at the gates, risers and bosses to obtain an infrared sensor array and a thermocouple temperature sensor array; traversing the infrared sensor array and the thermocouple temperature sensor array to perform pouring temperature time sequence collection according to a preset monitoring frequency, and obtaining an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; performing temperature distribution uniformity and temperature fluctuation double analysis based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array to determine a temperature regulation scheme; performing dynamic regulation on the pouring temperature of the target thin-walled automobile casting based on the temperature regulation scheme.

2. The cast temperature adjustment method with multi-dimensional sensor binding of claim 1, wherein, obtaining a historical casting quality detection record set of a target thin-walled automobile casting, performing warm control key anchor point analysis based on the historical casting quality detection record set, and obtaining a warm control abnormal anchor point set, comprising: performing data extraction on the historical casting quality detection record set according to abnormal types and abnormal positions to obtain a historical abnormal type-abnormal position set; performing same type aggregation on the historical abnormal type-abnormal position set with the same historical abnormal type as the index to determine a plurality of same type historical abnormal type-abnormal position sets; performing warm control key anchor point analysis on the plurality of same type historical abnormal type-abnormal position sets respectively with the abnormal position as the analysis object to obtain the warm control abnormal anchor point set.

3. The cast temperature adjustment method incorporating multidimensional sensors as claimed in claim 2, wherein, performing warm control key anchor point analysis on the plurality of same type historical abnormal type-abnormal position sets respectively with the abnormal position as the analysis object to obtain the warm control abnormal anchor point set, comprising: extracting a first same type historical abnormal type-abnormal position set from the plurality of same type historical abnormal type-abnormal position sets; performing sliding on the first same type historical abnormal type-abnormal position set according to a preset sliding window, counting the number of abnormal positions in each sliding sub-region window, and obtaining a sliding sub-region window abnormal position number set; taking the sliding sub-region window corresponding to the maximum value in the sliding sub-region window abnormal position number set as a first initial anchor point sliding sub-region window; performing coordinate mean filtering of abnormal positions in the first initial anchor point sliding sub-region window to determine a first warm control key anchor point; adding the first warm control key anchor point to the warm control abnormal anchor point set.

4. The cast temperature adjustment method incorporating multidimensional sensors as claimed in claim 3 wherein, performing coordinate filtering of abnormal positions in the first initial anchor point sliding sub-region window to determine a first warm control key anchor point, comprising: randomly selecting an abnormal position in the first initial anchor point sliding sub-region window as a coordinate filtering starting point, performing mean shift filtering of the coordinate filtering starting point in the first initial anchor point sliding sub-region window according to one-twentieth of the preset sliding window length to determine a first warm control key anchor point.

5. The cast temperature adjustment method with multi-dimensional sensor binding of claim 1, wherein, The temperature distribution uniformity and temperature fluctuation are analyzed based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, and a temperature regulation scheme is determined, including: The temperature mean values of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array are calculated respectively to determine the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array; The temperature distribution uniformity of the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array is analyzed to determine the first temperature regulation factor; The temperature fluctuation of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array is analyzed respectively to determine the second temperature regulation factor; The temperature regulation scheme is identified according to the first temperature regulation factor and the second temperature regulation factor to determine the temperature regulation scheme.

6. The cast temperature adjustment method incorporating multidimensional sensors as claimed in claim 5 wherein, The temperature distribution uniformity of the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array is analyzed to determine the first temperature regulation factor, including: The temperature distribution fluctuation of the infrared monitoring temperature mean value array and the thermocouple monitoring temperature mean value array is analyzed respectively to determine the infrared monitoring temperature distribution uniformity coefficient and the thermocouple monitoring temperature distribution uniformity coefficient; The infrared monitoring temperature distribution uniformity coefficient and the thermocouple monitoring temperature distribution uniformity coefficient are analyzed to determine the first temperature regulation factor.

7. The cast temperature adjustment method incorporating multidimensional sensors as claimed in claim 5 wherein, The temperature fluctuation of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array is analyzed respectively to determine the second temperature regulation factor, including: The temperature fluctuation characteristic identifier is called to identify the temperature fluctuation characteristics of the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array respectively to obtain the infrared monitoring temperature fluctuation characteristic array and the thermocouple monitoring temperature fluctuation array; The feature interaction enhancement is performed on the infrared monitoring temperature fluctuation characteristic array and the thermocouple monitoring temperature fluctuation array respectively to obtain the infrared monitoring temperature fluctuation enhanced feature and the thermocouple monitoring temperature fluctuation enhanced feature; The second temperature regulation factor is determined according to the infrared monitoring temperature fluctuation enhanced feature and the thermocouple monitoring temperature fluctuation enhanced feature.

8. The cast temperature adjustment method incorporating multidimensional sensors as claimed in claim 7, wherein, Including: The similarity of any two infrared monitoring temperature fluctuation characteristics in the infrared monitoring temperature fluctuation characteristic array is identified, and a feature interaction enhancement matrix is constructed according to the identification result; The graph convolution operation is performed on any two infrared monitoring temperature fluctuation characteristics using the feature interaction enhancement matrix to obtain the initial infrared monitoring temperature fluctuation enhanced feature; The similarity of any two infrared monitoring temperature fluctuation characteristics in the infrared monitoring temperature fluctuation characteristic array is identified again, and the initial infrared monitoring temperature fluctuation enhanced feature set is obtained when the identification times meet the preset times; The mean value of the initial infrared monitoring temperature fluctuation enhanced feature set is calculated to obtain the infrared monitoring temperature fluctuation enhanced feature.

9. The cast temperature adjustment method with multi-dimensional sensor binding of claim 5, wherein, Including: The first temperature monitoring frequency and the second temperature monitoring frequency are configured according to the sizes of the first temperature regulation factor and the second temperature regulation factor; The first temperature monitoring frequency is distributed to the infrared sensor array, and the second temperature monitoring frequency is distributed to the thermocouple temperature sensor array, and the target thin-walled automobile casting is monitored for pouring temperature based on the infrared sensor array and the thermocouple temperature sensor array after the temperature monitoring frequency is updated.

10. A pour temperature regulation system incorporating multidimensional sensors, characterized in that, The system comprises: A historical data acquisition module: acquires a historical casting quality detection record set of the target thin-walled automobile casting, analyzes temperature control key anchor points based on the historical casting quality detection record set, and obtains a temperature control abnormal anchor point set; A sensor layout planning module: lays out infrared sensors on the casting by traversing the temperature control abnormal anchor point set, and lays out thermocouple temperature sensors at gates, risers and bosses, to obtain an infrared sensor array and a thermocouple temperature sensor array; A temperature data acquisition module: acquires pouring temperature time series based on a preset monitoring frequency by traversing the infrared sensor array and the thermocouple temperature sensor array, to obtain an infrared monitoring temperature value sequence array and a thermocouple monitoring temperature value sequence array; A temperature analysis and decision module: performs dual analysis of temperature distribution uniformity and temperature fluctuation based on the infrared monitoring temperature value sequence array and the thermocouple monitoring temperature value sequence array, to determine a temperature control scheme; A pouring temperature control module: dynamically controls the pouring temperature of the target thin-walled automobile casting based on the temperature control scheme.

Images