Semi-solid magnesium alloy injection casting temperature feedback control method and equipment

Abstract

The invention relates to the field of industrial data acquisition control, and provides a semi-solid magnesium alloy injection casting temperature feedback control method and equipment, and the method comprises the steps: monitoring and obtaining sensor temperature data at a plurality of moments in a semi-solid magnesium alloy injection casting process; acquiring a plurality of detection time periods and predicted temperature data thereof; analyzing the difference and variation trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantifying the temperature deviation factor of each detection time period; based on the correlation of the change trend between the sensor temperature data and the predicted temperature data in each detection time period, the change of the correlation under the time sequence change of the detection time period is analyzed, and the temperature deviation accuracy of each detection time period is obtained; and correcting the sensor temperature data at each moment to obtain melt temperature data at each moment, and performing feedback control. The invention aims to solve the problem that the overall temperature feedback control effect is influenced by local temperature change in the magnesium alloy injection casting process.

Description

Technical Field

[0001] The invention relates to the field of industrial data acquisition control, and in particular to a temperature feedback control method and equipment for semi-solid magnesium alloy injection casting. Background Art

[0002] Semi-solid magnesium alloy injection casting is a high-precision processing technology that uses the special fluidity of metal in a state of solid-liquid coexistence to form. In this technology, it is necessary to ensure that the melt temperature of the alloy is controlled between the solidus and the liquidus (usually 520℃~600℃). This temperature range can ensure that the solid phase particles are evenly distributed in the alloy, ensuring its good fluidity and formability; if the temperature is too low, the metal fluidity will be poor and it will not be able to fill the mold, which is prone to defects such as cold shut; if the temperature is too high, the liquid metal will wash the mold, resulting in a shortened mold life and may also change the alloy microstructure; therefore, it is necessary to feedback control the melt temperature during the magnesium alloy injection casting process to ensure that its solid-liquid phase is controlled within a normal range.

[0003] During the melting of solid magnesium alloy, the temperature of the melt will drop due to switching stirring and other processes, which will cause local temperature fluctuations in the magnesium alloy melt. When the sensor collects temperature and performs feedback adjustment, the local temperature change will cause a deviation between the collected temperature and the actual temperature of the melt as a whole. As a result, the temperature feedback regulation process cannot be accurately controlled, which affects the fluidity and formability of the solid phase particles in the magnesium alloy injection casting process and reduces the quality of the casting. Summary of the invention

[0004] The present invention provides a method and device for temperature feedback control of semi-solid magnesium alloy injection casting, so as to solve the problem that local temperature changes in the existing magnesium alloy injection casting process affect the overall temperature feedback control effect. The technical solution adopted is as follows:

[0005] The present invention proposes a temperature feedback control method for semi-solid magnesium alloy injection casting, which comprises the following steps:

[0006] Monitor and obtain sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process;

[0007] Based on the sensor temperature data at each moment, obtain several detection time periods and their predicted temperature data; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period;

[0008] Based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained;

[0009] According to the temperature deviation factor and temperature deviation accuracy of each detection time period, the sensor temperature data at each moment is corrected, the melt temperature data at each moment is obtained and feedback control is performed.

[0010] Optionally, the specific method for obtaining the several detection time periods and their predicted temperature data is as follows:

[0011] The time period of heating the melt in the melting stage is extracted and recorded as the heating period in the melting stage; starting from the heating period, the initial heating period of the initial duration is intercepted;

[0012] After the initial heating period is over, the remaining heating period and the periods at each stage of the injection casting process are divided into time periods according to the segment length to obtain a number of detection time periods;

[0013] For the first detection time period, the sensor temperature data at all moments before the detection time period are input into the ARIMA algorithm, and the predicted temperature data at the first moment in the detection time period are output; the predicted temperature data at the first moment is used as the input of the ARIMA algorithm, and the predicted temperature data at the second moment in the detection time period are continuously output; and so on, the predicted temperature data at each moment in the detection time period are obtained;

[0014] The predicted temperature data at each moment in the first detection time period is input into the ARIMA algorithm, and combined with the sensor temperature data at each moment in the previous initial heating period, the predicted temperature data at each moment in the second detection time period is output, and so on, to obtain the predicted temperature data for each detection time period.

[0015] Optionally, the specific method of analyzing the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period and quantifying the temperature deviation factor of each detection time period includes:

[0016] For any detection time period, the difference between the sensor temperature data at any moment in the detection time period and the predicted temperature data at that moment is calculated, and the difference is used as the temperature difference at that moment; the temperature differences at all moments in the detection time period are obtained, and the average of the temperature differences at all moments is used as the temperature deviation of the detection time period;

[0017] According to the variation trend of the temperature difference in adjacent detection time periods, combined with the temperature deviation, a temperature deviation factor for each detection time period is obtained.

[0018] Optionally, the temperature deviation factor of each detection time period is obtained according to the change trend of the temperature difference in adjacent detection time periods in combination with the temperature deviation, including the specific method of:

[0019] A coordinate system is constructed with the order value of each moment in any detection time period as the horizontal axis and the temperature difference as the vertical axis. The temperature difference at each moment in the detection time period is mapped to the coordinate system to obtain several data points, and all data points are fitted by the least squares method to obtain a temperature difference fitting line for the detection time period. The slope of the temperature difference fitting line is recorded as the temperature difference change trend of the detection time period; the temperature deviation factor δ of the i-th detection time period is i The calculation method is:

[0020] δ i =norm[(k i ―k i―1 )×|g i |]

[0021] Among them, k i represents the temperature difference change trend in the i-th detection time period, k i―1 Indicates the temperature difference change trend in the i-1th detection time period, g i represents the temperature difference in the i-th detection time period; || represents the absolute value function, and norm[] represents the linear normalization function.

[0022] Optionally, the temperature deviation accuracy of each detection time period is obtained based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and the change of the correlation under the time series change of the detection time period is analyzed, including the specific method of:

[0023] Obtain the correlation between the sensor temperature data and the predicted temperature data in the same detection time period;

[0024] The change of the correlation degree between adjacent detection time periods is analyzed to obtain the temperature deviation accuracy of each detection time period.

[0025] Optionally, the correlation between the sensor temperature data and the predicted temperature data in the same detection time period is specifically obtained by:

[0026] For any detection time period, the time series consisting of the sensor temperature data at all times in the detection time period is recorded as the sensor temperature series of the detection time period;

[0027] The time series consisting of the predicted temperature data at all moments in the detection time period is recorded as the predicted temperature series of the detection time period;

[0028] The Pearson correlation coefficient between the sensor temperature sequence and the predicted temperature sequence is used as the correlation degree between the sensor temperature data and the predicted temperature data in the detection time period.

[0029] Optionally, the analyzing the change in the correlation between adjacent detection time periods to obtain the temperature deviation accuracy of each detection time period includes the following specific methods:

[0030] The calculation method of the temperature deviation accuracy in the i-th detection time period is:

[0031]

[0032] Among them, f i represents the correlation deviation factor of the i-th detection time period, α i―1 、a i and a i+1 They represent the correlation degree of the i-1th detection time period, the correlation degree of the i-th detection time period, and the correlation degree of the i+1th detection time period, respectively. max Indicates the maximum value of the correlation degree of all detection time periods;

[0033] The correlation deviation factors of all detection time periods are normalized to their maximum and minimum values, where the minimum value is set to the minimum difference between the correlation degrees of two adjacent detection time periods, and the maximum value is set to the maximum value of the absolute value of the difference between the correlation degrees of two adjacent detection time periods. The normalized result is used as the temperature deviation accuracy of each detection time period.

[0034] Optionally, the sensor temperature data at each moment is corrected according to the temperature deviation factor and the temperature deviation accuracy of each detection time period to obtain the melt temperature data at each moment, including the specific method of:

[0035] For the sensor temperature data at any time in any detection time period, the product of the sensor temperature data and the temperature deviation factor at that time is used as the corrected temperature data at that time;

[0036] The temperature deviation accuracy of the detection time period is used as the weight of the corrected temperature data, and the difference obtained by subtracting the temperature deviation accuracy from 1 is used as the weight of the sensor temperature data at that moment, and the melt temperature data at that moment is obtained by weighted summation.

[0037] Optionally, the feedback control includes the following specific methods:

[0038] If the melt temperature data at the current moment is lower than the lower limit of the temperature range of the stage, the alloy is heated; if it is higher than the upper limit of the temperature range of the stage, the heating of the alloy is stopped.

[0039] The present invention also proposes a semi-solid magnesium alloy injection casting temperature feedback control device, the device comprising:

[0040] Temperature data acquisition module, used to monitor and obtain sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process;

[0041] The temperature data analysis module is used to obtain several detection time periods and their predicted temperature data based on the sensor temperature data at each moment; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period;

[0042] Based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained;

[0043] The temperature feedback control module is used to correct the sensor temperature data at each moment according to the temperature deviation factor and temperature deviation accuracy of each detection time period, obtain the melt temperature data at each moment and perform feedback control.

[0044] The beneficial effects of the present invention are as follows: the present invention analyzes the sensor temperature during the injection-casting process of the semi-solid magnesium alloy, predicts the subsequent temperature based on the fact that the temperature is less disturbed in the initial stage of heating, and quantifies the deviation of the sensor temperature and makes adjustments based on this; wherein the detection time period is extracted from the melting stage, and the temperature data of the subsequent detection time period is predicted based on the sensor temperature data that is not disturbed in the initial stage of heating, and the deviation of the sensor temperature data of each detection time period is further analyzed through the predicted temperature data to quantify the temperature deviation caused by the uneven heating caused by possible stirring and other processes, and the temperature deviation of each detection time period is adjusted through the change trend of the temperature deviation of adjacent detection time periods to more accurately reflect the temperature deviation of each detection time period; on the basis of the temperature deviation factor, by analyzing the change in the correlation between the sensor temperature data and the predicted temperature data between adjacent detection time periods, the detection time period with a reduced correlation has a less reliable predicted temperature data, and the temperature deviation accuracy should be reduced to reduce the influence of the temperature deviation factor; thereby reducing the misjudgment caused by the uneven melt temperature caused by stirring and other processes during the injection-casting process, thereby improving the accuracy of temperature feedback control. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0046] Figure 1A schematic flow chart of a temperature feedback control method for semi-solid magnesium alloy injection casting provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 , which shows a flow chart of a temperature feedback control method for semi-solid magnesium alloy injection casting provided by an embodiment of the present invention, the method comprising the following steps:

[0049] Step S001: monitor and obtain sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process.

[0050] The purpose of this embodiment is to reduce the impact of uneven melt temperature caused by stirring and other processes during the injection-casting of semi-solid magnesium alloys, and to avoid deviations and misjudgments during temperature feedback control. Therefore, it is necessary to collect the temperature of the semi-solid magnesium alloy injection-casting process in real time through sensors.

[0051] Specifically, the semi-solid magnesium alloy injection casting process includes multiple stages such as melting, insulation and injection, and each stage includes processes such as temperature stabilization, slag removal and stirring. In the melting stage, the alloy heating rate is 50℃~100℃ / hour, and the melting temperature needs to be controlled between 650℃~700℃, of which the slag removal time is 10~15 minutes, and stirring is required to ensure uniform temperature. In the insulation stage, the alloy needs to be maintained in the temperature range of 650℃~700℃, and the insulation time is 30~60 minutes. The sample needs to be regularly deslagging and evenly stirred; in the injection stage, the injection temperature is controlled between 640℃ and 680℃, the mold temperature is 150℃ to 250℃, the injection speed is 0.5-1.5m / s, the initial injection pressure is 10-20MPa, and the holding pressure is 20-40Mpa; and temperature collection is carried out from the melting stage. The temperature sensor is installed in the melt standby area, and the sampling frequency is set to 10Hz, that is, 10 times per second, to monitor and obtain the sensor temperature data at several moments in real time.

[0052] Step S002: Based on the sensor temperature data at each moment, obtain several detection time periods and their predicted temperature data; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period.

[0053] It should be noted that in the initial stage of the melting stage, the overall temperature of the melt is relatively low because it is in the initial heating process. During this period, the melt has not undergone processes such as stirring and slag removal, and the overall temperature of the melt is similar. It is necessary to extract the time period of the heating stage and divide the subsequent temperature stabilization stage into a detection time period; and in the detection time period, since the melt temperature needs to be kept stable (stable at a certain temperature or steadily rising), stirring and slag removal and other processes are usually required, which will cause uneven heating of the melt and produce local temperature deviations. The temperature data measured by the sensor may be the temperature of the local temperature-uneven area, so it is necessary to quantify the deviation of the uneven temperature relative to the overall temperature of the melt. As for the overall temperature of the melt, it can be predicted by the previous melt temperature, that is, based on the undisturbed temperature during the heating process.

[0054] Preferably, in one embodiment of the present invention, based on the sensor temperature data at each moment, a number of detection time periods and their predicted temperature data are obtained, including the specific method of:

[0055] The time period of melt heating in the melting stage is extracted and recorded as the heating period in the melting stage; the initial time length is preset, and the initial time length of this embodiment is processed as 5 minutes as an example. Starting from the heating period, the initial heating period of 5 minutes is intercepted; the segmented time length is preset, and the segmented time length of this embodiment is processed as 1 minute as an example. After the initial heating period ends, the remaining heating period and the time periods in each stage of the injection casting process are divided into time periods according to the segmented time length to obtain a number of detection time periods.

[0056] Furthermore, for the first detection time period, the sensor temperature data of all moments before the detection time period are input into the ARIMA algorithm, and the predicted temperature data of the first moment in the detection time period is output; the predicted temperature data of the first moment is used as the input of the ARIMA algorithm, and the predicted temperature data of the second moment in the detection time period is continuously output; and so on, the predicted temperature data of each moment in the detection time period is obtained; the predicted temperature data of each moment in the first detection time period is also input into the ARIMA algorithm, combined with the sensor temperature data of each moment in the previous initial heating period, and the predicted temperature data of each moment in the second detection time period is output, and so on, the predicted temperature data of each detection time period is obtained; wherein the ARIMA algorithm is a well-known technology in the field of data prediction, and will not be described in detail in this embodiment.

[0057] Preferably, in one embodiment of the present invention, the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period are analyzed to quantify the temperature deviation factor of each detection time period, including the specific method of:

[0058] It should be noted that the predicted temperature data is predicted from the very beginning based on the sensor temperature data in the initial heating period, which can reflect the actual data change trend in the detection time period, that is, the temperature data measured by the sensor may be affected by the temperature unevenness caused by processes such as stirring, while the predicted temperature data tends to the temperature data of the entire melt. The possible temperature deviation is quantified by the difference between the predicted temperature data and the sensor temperature data; at the same time, for adjacent detection time periods, the deviation between the sensor temperature data and the predicted temperature data at each moment will show a changing trend, and the greater the difference in the changing trend, the greater the change in temperature deviation, and the more it is necessary to adjust the temperature deviation to ensure that the temperature deviation in each detection time period is accurately reflected.

[0059] Specifically, for any detection time period, the difference obtained by subtracting the predicted temperature data at any moment in the detection time period from the sensor temperature data at that moment is calculated, and the difference is used as the temperature difference at that moment; the temperature differences of all moments in the detection time period are obtained, and the average of the temperature differences of all moments is used as the temperature deviation of the detection time period.

[0060] Furthermore, a coordinate system is constructed with the order value of each moment in the detection time period as the horizontal axis and the temperature difference as the vertical axis, and the temperature difference at each moment in the detection time period is mapped to the coordinate system to obtain a number of data points, and all the data points are fitted by the least squares method to obtain a temperature difference fitting line for the detection time period, and the slope of the temperature difference fitting line is recorded as the temperature difference change trend of the detection time period; the temperature deviation and temperature difference change trend of each detection time period are obtained according to the above method, wherein the temperature difference change trend of the initial heating period is set to 0, then the temperature deviation factor δ of the i-th detection time period is i The calculation method is:

[0061] δ i =norm[(k i ―k i―1 )×|g i |]

[0062] Among them, k i represents the temperature difference change trend in the i-th detection time period, k i―1 Indicates the temperature difference change trend in the i-1th detection time period, g i represents the temperature difference of the i-th detection time period; || represents the absolute value function, norm[] represents the linear normalization function, and the normalization object is (k i ―k i―1 )×|g i |, wherein when i=1, the i-1=0th detection time period is the initial heating period.

[0063] It should be noted that the greater the difference in the change trends of adjacent detection time periods, that is, the greater the change trend compared with the previous detection time period, the more it is necessary to amplify the temperature deviation of the current time period, so as to quantify the temperature deviation factor.

[0064] At this point, the detection time period is extracted for the melting stage, and the temperature data of the subsequent detection time period is predicted based on the undisturbed sensor temperature data in the initial heating stage. The deviation of the sensor temperature data of each detection time period is further analyzed through the predicted temperature data to quantify the temperature deviation caused by uneven heating due to possible processes such as stirring, and adjustments are made through the changing trend of the temperature deviation of adjacent detection time periods to more accurately reflect the temperature deviation of each detection time period.

[0065] Step S003: Based on the correlation between the change trends of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained.

[0066] It should be noted that the predicted temperature data of each detection time period is predicted based on the previous sensor temperature data and predicted temperature data. Therefore, its reflection of the actual melt temperature at the corresponding moment is affected by the predicted temperature data of the previous detection time period. It is necessary to perform a correlation analysis on the predicted temperature data and the sensor temperature data of the previous detection time periods. The greater the correlation change between adjacent detection time periods, the less accurate the predicted temperature data. The temperature deviation factor calculated based on it needs to reduce its reference, so as to quantify the accuracy of the temperature deviation.

[0067] Preferably, in one embodiment of the present invention, the specific method included in this step is:

[0068] Obtain the correlation between the sensor temperature data and the predicted temperature data in the same detection time period;

[0069] The change of the correlation degree between adjacent detection time periods is analyzed to obtain the temperature deviation accuracy of each detection time period.

[0070] As an example, the correlation between the sensor temperature data and the predicted temperature data in the same detection time period is obtained, including the following specific methods:

[0071] For any detection time period, the time series sequence composed of the sensor temperature data at all moments in the detection time period is recorded as the sensor temperature sequence of the detection time period; similarly, the time series sequence composed of the predicted temperature data at all moments in the detection time period is recorded as the predicted temperature sequence of the detection time period; the Pearson correlation coefficient between the sensor temperature sequence and the predicted temperature sequence is used as the correlation degree between the sensor temperature data and the predicted temperature data in the detection time period.

[0072] As an example, analyzing the change in the correlation between adjacent detection time periods to obtain the temperature deviation accuracy of each detection time period includes the following specific methods:

[0073] It should be noted that by analyzing the difference in correlation between adjacent detection time periods, on the basis of a smaller correlation, the more it decreases compared to the adjacent detection time periods on both sides, the greater the change in correlation, and the less accurate the temperature deviation calculated using the predicted temperature data.

[0074] Specifically, the correlation degree of the initial heating period is set to 1, and the calculation method of the temperature deviation accuracy of the i-th detection period is:

[0075]

[0076] Among them, f i represents the correlation deviation factor of the i-th detection time period, α i―1 、a i and a i+1 They represent the correlation degree of the i-1th detection time period, the correlation degree of the i-th detection time period, and the correlation degree of the i+1th detection time period, respectively. max Represents the maximum value of the correlation degree of all detection time periods; the correlation deviation factors of all detection time periods are normalized to the maximum and minimum values, where the minimum value is set to the minimum value of the difference between the correlation degrees of two adjacent detection time periods (which may be a negative number), and the maximum value is set to the maximum value of the absolute value of the difference between the correlation degrees of two adjacent detection time periods (here the adjacent calculation includes the initial heating period), and the normalized result is used as the temperature deviation accuracy of each detection time period; in particular, α 0 This is the degree of correlation during the initial heating period.

[0077] It should be noted that by analyzing the change in correlation between the current detection time period and the adjacent detection time periods, and assigning weights to the adjacent detection time periods according to their correlation, the greater the correlation, the more credible the correlation change, and the more the correlation decreases, the greater the difference in the overall change trend between the predicted temperature data and the sensor temperature data, the more inaccurate it is, and the lower the accuracy of the temperature deviation.

[0078] At this point, on the basis of the temperature deviation factor, by analyzing the change in the correlation between the sensor temperature data and the predicted temperature data between adjacent detection time periods, the more the correlation decreases in the detection time period, the less reliable its predicted temperature data is, and the temperature deviation accuracy should be reduced to reduce the impact of the temperature deviation factor.

[0079] Step S004: Correct the sensor temperature data at each moment according to the temperature deviation factor and temperature deviation accuracy of each detection time period, obtain the melt temperature data at each moment and perform feedback control.

[0080] Specifically, for the sensor temperature data at any moment in any detection time period, the product of the sensor temperature data and the temperature deviation factor at that moment is used as the corrected temperature data at that moment; the temperature deviation accuracy of the detection time period is used as the weight of the corrected temperature data, and the difference obtained by subtracting the temperature deviation accuracy from 1 is used as the weight of the sensor temperature data at that moment, and the melt temperature data at that moment is obtained by weighted summation; the melt temperature data at each moment in each detection time period is obtained according to the above method.

[0081] Furthermore, each stage in the semi-solid magnesium alloy injection casting process corresponds to a corresponding temperature range. If the melt temperature data at the current moment is lower than the lower limit of the temperature range of the stage, the alloy needs to be heated; conversely, if it is higher than the upper limit of the temperature range of the stage, the heating of the alloy is stopped.

[0082] It should be noted that the temperature feedback control is a real-time process. In the detection time period analysis process, there is no subsequent adjacent detection time period. In the above analysis process, only the previous adjacent detection time period is analyzed for calculation and processing.

[0083] So far, by analyzing the sensor temperature during the injection-casting process of semi-solid magnesium alloy, the subsequent temperature prediction is carried out based on the fact that the overall temperature is less disturbed during the initial heating period. The deviation of the sensor temperature is quantified and adjusted to reduce the misjudgment caused by uneven melt temperature due to processes such as stirring during the injection-casting process, thereby improving the accuracy of temperature feedback control.

[0084] Another embodiment of the present invention provides a semi-solid magnesium alloy injection casting temperature feedback control device, the device comprising:

[0085] Temperature data acquisition module: monitors and obtains sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process;

[0086] Temperature data analysis module: Based on the sensor temperature data at each moment, obtain several detection time periods and their predicted temperature data; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period;

[0087] Based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained;

[0088] Temperature feedback control module: According to the temperature deviation factor and temperature deviation accuracy of each detection time period, the sensor temperature data at each moment is corrected, the melt temperature data at each moment is obtained and feedback control is performed.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A temperature feedback control method for semi-solid magnesium alloy injection casting, characterized in that: The method comprises the following steps: Monitor and obtain sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process; Based on the sensor temperature data at each moment, obtain several detection time periods and their predicted temperature data; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period; Based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained; According to the temperature deviation factor and temperature deviation accuracy of each detection time period, the sensor temperature data at each moment is corrected, the melt temperature data at each moment is obtained and feedback control is performed.

2. A temperature feedback control method for semi-solid magnesium alloy injection casting according to claim 1, characterized in that: The specific method for obtaining the several detection time periods and their predicted temperature data is as follows: The time period of heating the melt in the melting stage is extracted and recorded as the heating period in the melting stage; starting from the heating period, the initial heating period of the initial duration is intercepted; After the initial heating period is over, the remaining heating period and the periods at each stage of the injection casting process are divided into time periods according to the segment length to obtain a number of detection time periods; For the first detection time period, the sensor temperature data at all moments before the detection time period are input into the ARIMA algorithm, and the predicted temperature data at the first moment in the detection time period are output; the predicted temperature data at the first moment is used as the input of the ARIMA algorithm, and the predicted temperature data at the second moment in the detection time period are continuously output; and so on, the predicted temperature data at each moment in the detection time period are obtained; The predicted temperature data at each moment in the first detection time period is input into the ARIMA algorithm, and combined with the sensor temperature data at each moment in the previous initial heating period, the predicted temperature data at each moment in the second detection time period is output, and so on, to obtain the predicted temperature data for each detection time period.

3. The method for temperature feedback control of semi-solid magnesium alloy injection casting according to claim 1, characterized in that: The specific method of analyzing the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period and quantifying the temperature deviation factor of each detection time period includes: For any detection time period, the difference between the sensor temperature data at any moment in the detection time period and the predicted temperature data at that moment is calculated, and the difference is used as the temperature difference at that moment; the temperature differences at all moments in the detection time period are obtained, and the average of the temperature differences at all moments is used as the temperature deviation of the detection time period; According to the variation trend of the temperature difference in adjacent detection time periods, combined with the temperature deviation, a temperature deviation factor for each detection time period is obtained.

4. A temperature feedback control method for semi-solid magnesium alloy injection casting according to claim 3, characterized in that: The temperature deviation factor of each detection time period is obtained according to the change trend of the temperature difference in adjacent detection time periods in combination with the temperature deviation, including the specific method of: A coordinate system is constructed with the order value of each moment in any detection time period as the horizontal axis and the temperature difference as the vertical axis. The temperature difference at each moment in the detection time period is mapped to the coordinate system to obtain several data points, and all data points are fitted by the least squares method to obtain a temperature difference fitting line for the detection time period. The slope of the temperature difference fitting line is recorded as the temperature difference change trend of the detection time period; the temperature deviation factor δ of the i-th detection time period is i The calculation method is: d i =norm[(k i ―k i―1 )×|g i |] Among them, k i represents the temperature difference change trend in the i-th detection time period, k i―1 Indicates the temperature difference change trend in the i-1th detection time period, g i represents the temperature difference in the i-th detection time period; || represents the absolute value function, and norm[] represents the linear normalization function.

5. The method for temperature feedback control of semi-solid magnesium alloy injection casting according to claim 1, characterized in that: The method of obtaining the temperature deviation accuracy of each detection time period based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period and analyzing the change of the correlation under the time series change of the detection time period includes: Obtain the correlation between the sensor temperature data and the predicted temperature data in the same detection time period; The change of the correlation degree between adjacent detection time periods is analyzed to obtain the temperature deviation accuracy of each detection time period.

6. A temperature feedback control method for semi-solid magnesium alloy injection casting according to claim 5, characterized in that: The degree of correlation between the sensor temperature data and the predicted temperature data in the same detection time period is specifically obtained by: For any detection time period, the time series consisting of the sensor temperature data at all times in the detection time period is recorded as the sensor temperature series of the detection time period; The time series consisting of the predicted temperature data at all moments in the detection time period is recorded as the predicted temperature series of the detection time period; The Pearson correlation coefficient between the sensor temperature sequence and the predicted temperature sequence is used as the correlation degree between the sensor temperature data and the predicted temperature data in the detection time period.

7. A temperature feedback control method for semi-solid magnesium alloy injection casting according to claim 5, characterized in that: The specific method of analyzing the change of the correlation degree between adjacent detection time periods to obtain the temperature deviation accuracy of each detection time period includes: The calculation method of the temperature deviation accuracy in the i-th detection time period is: Among them, f i represents the correlation deviation factor of the i-th detection time period, α i―1 、a i and a i+1 They represent the correlation degree of the i-1th detection time period, the correlation degree of the i-th detection time period, and the correlation degree of the i+1th detection time period, respectively. max Indicates the maximum value of the correlation degree of all detection time periods; The correlation deviation factors of all detection time periods are normalized to their maximum and minimum values, where the minimum value is set to the minimum difference between the correlation degrees of two adjacent detection time periods, and the maximum value is set to the maximum value of the absolute value of the difference between the correlation degrees of two adjacent detection time periods. The normalized result is used as the temperature deviation accuracy of each detection time period.

8. The method for temperature feedback control of semi-solid magnesium alloy injection casting according to claim 1, characterized in that: The method of correcting the sensor temperature data at each moment according to the temperature deviation factor and the temperature deviation accuracy of each detection time period to obtain the melt temperature data at each moment includes the following specific methods: For the sensor temperature data at any time in any detection time period, the product of the sensor temperature data and the temperature deviation factor at that time is used as the corrected temperature data at that time; The temperature deviation accuracy of the detection time period is used as the weight of the corrected temperature data, and the difference obtained by subtracting the temperature deviation accuracy from 1 is used as the weight of the sensor temperature data at that moment. The melt temperature data at that moment is obtained by weighted summation.

9. The method for temperature feedback control of semi-solid magnesium alloy injection casting according to claim 1, characterized in that: The feedback control comprises the following specific methods: If the melt temperature data at the current moment is lower than the lower limit of the temperature range of the stage, the alloy is heated; if it is higher than the upper limit of the temperature range of the stage, the heating of the alloy is stopped.

10. A temperature feedback control device for semi-solid magnesium alloy injection casting, characterized in that: The equipment includes: Temperature data acquisition module, used to monitor and obtain sensor temperature data at several moments during the semi-solid magnesium alloy injection casting process; The temperature data analysis module is used to obtain several detection time periods and their predicted temperature data based on the sensor temperature data at each moment; analyze the difference and change trend between the sensor temperature data and the predicted temperature data in each detection time period, and quantify the temperature deviation factor of each detection time period; Based on the correlation between the change trend of the sensor temperature data and the predicted temperature data in each detection time period, and analyzing the change of the correlation under the time series change of the detection time period, the temperature deviation accuracy of each detection time period is obtained; The temperature feedback control module is used to correct the sensor temperature data at each moment according to the temperature deviation factor and temperature deviation accuracy of each detection time period, obtain the melt temperature data at each moment and perform feedback control.

Images