Real-time monitoring method and system for parameters of die casting equipment

Abstract

The invention relates to the technical field of data processing, in particular to a real-time monitoring method and system for parameters of die casting equipment. The method comprises the following steps: determining a critical point in a pressure data time sequence of the die-casting equipment according to a speed change condition of the die-casting equipment; according to the injection pressure between any moment and all moments in the critical point interval closest to the moment, the injection pressure change fluctuation degree at the moment is obtained; recording the difference between the injection pressure change fluctuation degree at the moment and the injection pressure change fluctuation degree at the previous moment as an adjustment factor at the moment; if the adjustment factor of the moment is greater than the length of the interval where the moment is located, taking the length of the interval where the moment is located as the number of neighbor data at the moment; otherwise, rounding up the adjustment factor at the moment and taking the number of the neighbor data at the moment as the number of the neighbor data at the moment; and the abnormal probability of each moment is calculated based on the SOS algorithm, so that the real-time monitoring result of the parameters of the die-casting equipment is obtained, and the parameter monitoring accuracy of the die-casting equipment is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a method and system for real-time monitoring of die-casting equipment parameters. Background Art

[0002] Die casting is a metal casting process that uses high pressure to inject molten metal into a mold to cast the required aluminum alloy castings. By precisely controlling the die-casting speed and die-casting pressure during die casting, the quality of the castings can be improved.

[0003] During the long-term operation of die-casting equipment, problems such as wear and hydraulic pipeline leakage are likely to occur, which will cause the injection pressure to leak and fail to reach the expected value. The molten metal cannot fill the mold cavity, and the castings produced have incomplete contours, missing edges or thin-walled parts, affecting the quality of the finished products. Therefore, it is necessary to monitor the injection pressure during the injection process in real time to identify abnormal changes. The Stochastic Outlier Selection (SOS) algorithm is an unsupervised anomaly detection algorithm that identifies outliers by calculating the affinity between data points and their neighboring data points, which can avoid false alarms triggered by vibrations, noises, etc., and the real-time performance of this algorithm is relatively high, and it can timely detect abnormal pressures during the operation of die-casting equipment.

[0004] However, the number of neighboring data points of the data points obtained in the SOS algorithm is a fixed value. During the operation of die-casting equipment, the pressure changes differently in different stages. If a fixed number of neighboring data points are set for all pressure data, it may occur that the neighboring data cannot accurately reflect the normal behavior pattern of the current data point, thereby reducing the monitoring accuracy of die-casting equipment parameters.

[0005] Based on this, how to set the number of neighboring data points for each data according to the pressure changes in different stages during the operation of die-casting equipment, so as to accurately reflect the normal behavior pattern of the current data point and improve the monitoring accuracy of die-casting equipment, is a problem that needs to be solved currently. Summary of the Invention

[0006] In order to solve the technical problem of how to set the number of neighboring data points for each data according to the pressure changes in different stages during the operation of die-casting equipment, so as to accurately reflect the normal behavior pattern of the current data point and improve the monitoring accuracy of die-casting equipment, the present invention provides a method and system for real-time monitoring of die-casting equipment parameters.

[0007] In the first aspect, the present invention provides a method for real-time monitoring of die-casting equipment parameters, adopting the following technical solutions: A method for real-time monitoring of die-casting equipment parameters includes the steps: Determine the critical points in the time series sequence of the die-casting equipment pressure data based on the speed change of the die-casting equipment; obtain the degree of change and fluctuation of the injection pressure at this moment by calculating the average difference of the injection pressures between each moment in the interval between any moment and the nearest critical point; multiply the ratio of the degree of change and fluctuation of the injection pressure at this moment and the previous moment by the preset number of nearest neighbors at this moment, and denote it as the adjustment factor at this moment; if the adjustment factor at this moment is greater than the length of the interval where this moment is located, then use the length of the interval where this moment is located as the number of nearest neighbor data at this moment; otherwise, round up the adjustment factor at this moment and use it as the number of nearest neighbor data at this moment; calculate the anomaly probability of each moment using the number of nearest neighbor data at each moment in the SOS algorithm to obtain the real-time monitoring result of the die-casting equipment parameters.

[0008] The present invention can effectively avoid interferences such as vibration and noise by processing the injection pressure data of the die-casting equipment in real time through the SOS algorithm, so as to accurately obtain the real-time monitoring result of the die-casting equipment parameters. In this process, the present invention considers that the number of nearest neighbor data in the SOS algorithm is a fixed value, and underfitting or overfitting may occur in different change stages of the pressure data; based on this, the present invention adaptively calculates the optimal number of nearest neighbor data at each moment by analyzing the fluctuation of the pressure data, so that the SOS algorithm can accurately obtain the neighborhood data according to the fluctuation of the pressure at each moment for anomaly probability calculation, effectively improving the accuracy of the real-time monitoring result of the die-casting equipment pressure data. On this basis, the present invention also considers that there may be different change patterns between the pressure data in each stage, and directly determining the possible abnormal pressure data in each pressure stage as the stage critical point based on the pressure data; based on this, the present invention accurately realizes the division of each pressure stage by obtaining the speed data corresponding to the pressure data at each moment according to the change of the speed data, thereby effectively improving the accuracy of the real-time monitoring result of the die-casting equipment pressure data.

[0009] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, before determining the critical points in the time series sequence of the die-casting equipment pressure data based on the speed change of the die-casting equipment, it further includes: obtaining the injection pressure and speed value of the die-casting equipment in real time at each moment to obtain the time series sequence of the pressure data and the time series sequence of the speed data of the die-casting equipment.

[0010] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, determining the critical points in the time series sequence of the die-casting equipment pressure data based on the speed change of the die-casting equipment includes: calculating the degree of speed mutation at the i-th moment : ; , , They are the velocity values at the i-th moment, the moment, and the moment respectively; by comparing the velocity mutation degrees between each moment and the previous moment of this moment, determine the moment corresponding to the critical point.

[0011] The present invention provides an accurate calculation method for the velocity mutation degree at each moment. By the velocity change situation between each moment and its previous moment, the moment with a large velocity change difference from the previous moment can be accurately obtained. Such a moment is the moment corresponding to the critical point, and the injection pressure corresponding to this moment is the injection pressure corresponding to the critical point.

[0012] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, the step of determining the moment corresponding to the critical point by comparing the velocity mutation degrees between each moment and the previous moment of this moment includes: if the velocity mutation degree of a moment is greater than the velocity mutation degree of the previous moment of this moment, then this moment is the moment corresponding to the critical point in the pressure data time series.

[0013] The present invention provides an accurate method for determining the critical point in the pressure stage. By analyzing the velocity mutation degree of the injection speed, accurate division of each stage in the die-casting process is realized, so that monitoring can be carried out for different characteristics of each stage, and misjudgment caused by feature differences across stages can be avoided.

[0014] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, take the time difference between each moment and the moment corresponding to each critical point as the distance between each moment and each critical point.

[0015] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, the step of obtaining the injection pressure change fluctuation degree of this moment by the average value of the differences in injection pressure between any moment and each moment in the critical point interval with the closest distance includes: ; is the injection pressure change fluctuation degree at the i-th moment, is the number of moments in the critical point interval with the closest distance to the i-th moment, and are the injection pressures at the -th moment and the -th moment in the critical point interval with the closest distance to the i-th moment respectively, is the linear normalization function.

[0016] The present invention provides an accurate calculation method for the fluctuation degree of injection pressure change. By analyzing the injection pressure difference between the current moment and the initial moment of the pressure stage where the current moment is located, the pressure fluctuation condition in the interval where the current moment is located can be accurately obtained. The injection pressure change fluctuation degree at the i-th moment is used to characterize the pressure fluctuation condition in the interval where the i-th moment is located.

[0017] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, calculating the anomaly probability of each moment using the number of neighbor data at each moment in the SOS algorithm includes: constructing a dissimilarity matrix through the Euclidean distance between each moment and the neighbor data; based on the dissimilarity matrix, calculating the correlation matrix between moments, normalizing the correlation matrix into a probability distribution, and obtaining the anomaly probability of each moment.

[0018] The present invention provides the SOS algorithm to calculate the anomaly probability of pressure data at each moment, which can effectively avoid the interference of noise data while obtaining the anomaly monitoring result of pressure data in real time, and effectively improve the real-time monitoring efficiency and accuracy of the pressure parameters of die-casting equipment.

[0019] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, calculating the anomaly probability of each moment using the number of neighbor data at each moment in the SOS algorithm to obtain the real-time monitoring result of die-casting equipment parameters includes: if the anomaly probability of any moment is greater than the anomaly probability threshold, the real-time monitoring result of the pressure parameter of the die-casting equipment corresponding to this moment is abnormal; otherwise, the real-time monitoring result of the pressure parameter of the die-casting equipment corresponding to this moment is normal.

[0020] According to a real-time monitoring method for die-casting equipment parameters provided by the present invention, after obtaining the real-time monitoring result of die-casting equipment parameters, it further includes: in response to the real-time monitoring result of the pressure parameter of the die-casting equipment being abnormal, sending an anomaly prompt and intercepting the casting corresponding to this moment.

[0021] In a second aspect, the present invention provides a real-time monitoring system for die-casting equipment parameters, adopting the following technical solution: A real-time monitoring system for die-casting equipment parameters includes: a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned real-time monitoring method for die-casting equipment parameters is implemented.

[0022] By adopting the above technical solution, the above-mentioned real-time monitoring method for die-casting equipment parameters is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, which is convenient to use.

[0023] The present invention has the following technical effects: Based on the above technical solution, for a real-time monitoring method and system of die-casting equipment parameters provided by the present invention, when obtaining the real-time monitoring result of the pressure data of the die-casting equipment, the pressure data of the die-casting equipment is processed in real time through the SOS algorithm, which can effectively avoid interferences such as vibration and noise, so as to accurately obtain the real-time monitoring result of the die-casting equipment parameters. In this process, the present invention takes into account that the number of neighboring data in the SOS algorithm is a fixed value, and under-fitting or over-fitting phenomena may occur in different change stages of the pressure data; based on this, the present invention analyzes the fluctuation of the pressure data and adaptively calculates the optimal number of neighboring data at each moment, so that the SOS algorithm can accurately obtain the neighborhood data according to the fluctuation of the pressure at each moment for abnormal probability calculation, effectively improving the accuracy of the real-time monitoring result of the die-casting equipment pressure data. On this basis, the present invention also takes into account that there may be different change patterns between the pressure data in each stage, and directly determining based on the pressure data that abnormal pressure data may be misidentified as the stage critical point in each pressure stage; based on this, the present invention obtains the speed data corresponding to the pressure data at each moment, and obtains the speed mutation point according to the change of the speed data, accurately realizing the division of each pressure stage, so as to effectively improve the accuracy of the real-time monitoring result of the die-casting equipment pressure data. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic flowchart of a real-time monitoring method for die-casting equipment parameters provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments.

[0026] An embodiment of the present invention discloses a real-time monitoring method for die-casting equipment parameters. This method can set the number of neighboring data points of each data in the SOS algorithm according to the pressure changes in different stages during the working process of the die-casting equipment, so as to accurately reflect the normal behavior mode of the current data point and improve the accuracy of die-casting equipment monitoring.

[0027] Specifically, please refer to Figure 1 as shown in Figure 1 It is a schematic flowchart of a real-time monitoring method for die-casting equipment parameters provided by an embodiment of the present invention. The method specifically includes the following steps: S1: Obtain the injection pressure corresponding to each moment in the pressure data time series of the die-casting equipment.

[0028] Exemplarily, in the embodiments of the present invention, obtaining the injection pressure corresponding to each moment in the time series of pressure data of the die-casting equipment includes: obtaining the injection pressure of the die-casting equipment in real time at each moment, and preprocessing to obtain the time series of pressure data of the die-casting equipment.

[0029] Specifically, when obtaining the time series of pressure and speed data of the die-casting equipment, the injection pressure during the working process of the die-casting equipment can be collected by a pressure sensor inside the die-casting equipment, and the time series of pressure data can be obtained after preprocessing.

[0030] Among them, the acquisition frequency of the pressure sensor can be set to be acquired once per second, and the preprocessing can be interpolation of missing data, etc.; the acquisition frequency and the preprocessing method can be specifically set according to actual needs, and the embodiments of the present invention do not limit this too much here.

[0031] After the injection pressure data during the die-casting alloy process of the die-casting equipment is collected in real time based on the above steps, the time series of pressure data can be monitored.

[0032] It should be noted that the process of die-casting alloy by the die-casting equipment generally includes three stages, namely the filling stage, the boosting stage, and the holding pressure stage. Among them, in the filling stage, the molten metal needs to be gradually filled into the cavity, and the pressure will rapidly rise from zero, but the overall pressure data is relatively low; at the end of the filling stage, the pressure in the boosting stage rises sharply to the peak value to compensate for the solidification shrinkage of the metal, enhance the compactness of the casting and reduce shrinkage cavities; finally, in the holding pressure stage, the pressure gradually drops slowly from the peak value, but maintains a relatively high pressure to ensure that the metal continues to compensate for shrinkage during solidification and prevent internal defects.

[0033] Obviously, there are normal fluctuations in the pressure data in different stages. When the SOS algorithm uses a fixed number of neighborhood data points to monitor the pressure data, it cannot adapt to such normal changes. For example, in the boosting stage, the data density is high, and overfitting may occur in the fixed neighborhood, misjudging normal fluctuations as anomalies, while in the holding pressure stage, the data density is low, and underfitting may occur in the fixed neighborhood, resulting in the inability to detect small anomalies, etc.

[0034] Based on this, the embodiments of the present invention can divide the time series of pressure data of the die-casting equipment into different stages according to the change of the pressure data, and obtain the degree of abnormality of the pressure data at each moment according to the change of the pressure data in the current stage corresponding to each moment, so as to accurately obtain the real-time monitoring result of the pressure of the die-casting equipment, that is, execute the following steps.

[0035] S2: Determine the critical points in the time series of pressure data of the die-casting equipment by using the change of the speed of the die-casting equipment.

[0036] It should be noted that during the process of die-casting alloy by die-casting equipment, in the filling stage, in order to fill the cavity, the injection speed will increase gradually at equal intervals. In the boosting stage, the injection speed will gradually decrease, and finally in the holding pressure stage, the injection speed will maintain the minimum value.

[0037] Based on this, in order to reduce the influence of abnormal pressure data on the accuracy of stage division when dividing the pressure change stage of die-casting equipment based on the time series of pressure data, the embodiments of the present invention can obtain the mutation points of speed data by analyzing the change of injection speed at each moment, and the moment corresponding to the mutation point is the moment corresponding to the critical point of the time series of pressure data.

[0038] Exemplarily, in the embodiments of the present invention, before determining the critical point in the time series of pressure data of die-casting equipment by using the speed change of die-casting equipment, it further includes: obtaining the speed value of the injection pressure of die-casting equipment in real time at each moment to obtain the time series of speed data of die-casting equipment.

[0039] Specifically, when obtaining the time series of speed data of die-casting equipment, the speed value during the injection pressure process of die-casting equipment can be collected by a speed sensor inside the die-casting equipment, and the time series of speed data can be obtained after preprocessing.

[0040] Among them, the acquisition frequency of the speed sensor can be set to collect once per second, and the preprocessing can be interpolation of missing data, etc.; the acquisition frequency of the speed sensor can be the same as that of the pressure sensor, and the specific preprocessing method can be set according to actual needs, and the embodiments of the present invention do not limit it too much here.

[0041] It can be understood that the injection pressure and speed value at each moment in the finally obtained time series of speed data and time series of pressure data based on the above steps correspond one by one.

[0042] Exemplarily, in the embodiments of the present invention, determining the critical point in the time series of pressure data of die-casting equipment by using the speed change of die-casting equipment includes: calculating the degree of speed mutation at each moment, and determining the moment corresponding to the critical point through the comparison result of the degree of speed mutation between each moment and the previous moment of this moment.

[0043] Exemplarily, in the embodiments of the present invention, calculating the degree of speed mutation at the i-th moment , specifically, the following relational expression can be referred to: ; is the degree of speed mutation at the i-th moment, is the speed value at the i-th moment, is the moment's speed value, is the moment's speed value.

[0044] In the above formula, represents the velocity change amount from the i-th moment to the moment, represents the moment to the moment of the velocity change amount, and The greater the difference between, the more significant the velocity change between the i-th moment and the moment is, the higher the possibility of being a velocity mutation point is, and the i-th moment may be the moment when the injection process enters the boosting stage from the filling stage, or the moment when the boosting stage enters the holding pressure stage, and the possibility of belonging to the critical point is greater.

[0045] The degree of velocity mutation at the i-th moment is used to characterize the possibility of this moment being a critical point. The larger this value is, the greater the possibility that the i-th moment is a critical point.

[0046] After obtaining the degree of velocity mutation at each moment based on the above steps, the moment corresponding to the critical point can be obtained according to the degree of velocity mutation between moments.

[0047] Exemplarily, in the embodiment of the present invention, the moment corresponding to the critical point is determined by comparing the degree of velocity mutation between each moment and the previous moment of this moment, including: if the degree of velocity mutation of a moment is greater than the degree of velocity mutation of the previous moment of this moment, then this moment is the moment corresponding to the critical point in the pressure data time series.

[0048] After obtaining the moments corresponding to all critical points based on the above steps, the critical points in the pressure data time series of the die-casting equipment can be obtained according to the moments corresponding to the critical points, so that the degree of change and fluctuation of the injection pressure at each moment can be obtained based on the distance between the pressure data at each moment and the critical points, that is, continue to execute the following steps.

[0049] S3: Obtain the degree of change and fluctuation of the injection pressure at this moment through the average value of the differences in the injection pressures between any moment and each moment in the critical point interval with the closest distance.

[0050] Exemplarily, in the embodiment of the present invention, the time difference between each moment and the moment corresponding to each critical point is used as the distance between each moment and each critical point.

[0051] It should be noted that based on the above steps, the critical points in each stage of the pressure data time series can be obtained. However, during the working process of the die-casting equipment, the injection system may have problems such as metal residue, continuous die-casting resulting in wear of the injection punch or leakage of the hydraulic pipeline, which may cause the injection pressure to not reach the expected value and affect the die-casting quality. It can be understood that the change of the injection pressure in each stage during the normal working process of the die-casting equipment is relatively stable.

[0052] Based on this, the embodiments of the present invention can obtain the abnormally fluctuating pressure data based on the change of the pressure data in the interval between the current moment and the critical point moment, so that the number of neighborhood data points at each moment can be adjusted based on the abnormally fluctuating pressure data, and the abnormal injection pressure can be accurately obtained.

[0053] It should be particularly noted that the pressure data time series is obtained in real time. Therefore, at each latest moment, there is only the pressure data corresponding to the historical moment. Therefore, the critical point closest to the latest moment can be uniquely obtained, which is the critical point closest to the latest moment among the historical moments. The critical point is the initial moment of each stage. The interval between each moment and the closest critical point includes the current moment and the moment corresponding to the closest critical point. The distance between the current moment and the closest critical point is the distance between the current moment and the initial moment of the stage where the current moment is located.

[0054] Exemplarily, in the embodiments of the present invention, the degree of change and fluctuation of the injection pressure at each moment is calculated. Specifically, refer to the following relational expression: ; is the degree of change and fluctuation of the injection pressure at the i-th moment, is the number of moments in the interval between the i-th moment and the closest critical point, 、 respectively represent the moment indexes in the interval between the i-th moment and the closest critical point, is the injection pressure at the -th moment in the interval between the i-th moment and the closest critical point, is the injection pressure at the -th moment in the interval between the i-th moment and the closest critical point, is the linear normalization function.

[0055] In the above formula, and can have the same value.

[0056] The greater the degree of change and fluctuation of the injection pressure at the i-th moment, the higher the possibility that problems such as metal residue, injection punch wear, and hydraulic pipeline leakage cause abnormal changes in the injection pressure, resulting in lower stability and higher volatility of the pressure data change in the same stage.

[0057] represents the average difference of the injection pressures between the moments in the interval between the i-th moment and the closest critical point. The larger this value is, the greater the difference in the pressure data in the stage where the i-th moment is located, the greater the degree of change and fluctuation of the injection pressure at the i-th moment, and the greater the possibility of abnormal pressure data.

[0058] After obtaining the degree of change and fluctuation of the injection pressure corresponding to each moment based on the above steps, the size of the neighborhood data points at each moment can be adjusted according to the degree of change and fluctuation of the injection pressure, so as to accurately obtain the real-time monitoring results of the die-casting equipment parameters.

[0059] S4: Adjust the preset number of nearest neighbors according to the change amount of the degree of change and fluctuation of the injection pressure between each moment and its previous moment to obtain the number of nearest neighbor data at each moment; use the number of nearest neighbor data at each moment in the SOS algorithm to calculate the anomaly probability at each moment, so as to obtain the real-time monitoring results of the die-casting equipment parameters.

[0060] It should be noted that after dividing the real-time pressure data into stages based on the above steps, the influence of the pressure data change in different stages on the pressure data change at the current moment can be avoided, so as to accurately obtain the degree of change and fluctuation of the injection pressure corresponding to each moment. The greater the degree of change and fluctuation of the injection pressure, the greater the possibility of abnormal pressure data.

[0061] Based on this, the embodiment of the present invention can adaptively adjust the number of nearest neighbor data in combination with the change volatility of the pressure data in the same stage at the current moment. If the volatility is greater, it means that the possibility of abnormal data in the pressure data in the same stage at the current moment is greater. Therefore, it is necessary to increase the reference data to provide more normal data to evaluate the correlation between the pressure data at each moment, thereby improving the anomaly recognition accuracy.

[0062] Exemplarily, in the embodiment of the present invention, when adjusting the preset number of nearest neighbors according to the change amount of the degree of change and fluctuation of the injection pressure between each moment and its previous moment to obtain the number of nearest neighbor data at each moment, the product of the ratio of the degree of change and fluctuation of the injection pressure at this moment and its previous moment and the preset number of nearest neighbors at this moment can be recorded as the adjustment factor at this moment; if the adjustment factor at this moment is greater than the length of the interval where this moment is located, then use the length of the interval where this moment is located as the number of nearest neighbor data at this moment; otherwise, round up the adjustment factor at this moment to be used as the number of nearest neighbor data at this moment.

[0063] Among them, when setting the preset number of nearest neighbors at each moment, the rounded-up value of 30% of the length of the stage where this moment is located can be used as the preset number of nearest neighbors at this moment; the preset number of nearest neighbors at each moment can be specifically set according to actual needs, and the embodiment of the present invention does not limit this too much here.

[0064] It can be understood that the above-mentioned interval where this moment is located is the interval between this moment and the nearest critical point.

[0065] Exemplarily, in the embodiment of the present invention, to calculate the number of nearest neighbor data at the i-th moment, the following relational expression can be specifically referred to: ; is the number of neighboring data at the i-th moment, is the preset number of neighboring data at the i-th moment, is the degree of fluctuation of the injection pressure change at the i-th moment, is the degree of fluctuation of the injection pressure change at the moment, is the number of moments in the interval where the i-th moment is located, represents the ceiling symbol.

[0066] In the above formula, represents the adjustment factor at the i-th moment. If is greater than 1, it indicates that the degree of fluctuation of the injection pressure change at the i-th moment is greater than that at the moment. The injection pressure data at the i-th moment is more likely to have abnormal changes. At this time, it is necessary to increase the number of its neighboring data to include more normal reference data to evaluate its relevance, so as to make the abnormal evaluation at the current moment more reliable. If is not greater than 1, it indicates that the possibility of abnormal changes in the injection pressure data at the i-th moment is relatively low. At this time, the number of neighborhood data at the current moment can be reduced, thereby reducing the data processing volume.

[0067] It can be understood that if the possibility of abnormal changes in the injection pressure data at the i-th moment is very low, even if the number of neighboring data obtained based on the above steps is very small, the correlation degree at the current moment can be accurately obtained, so as to accurately obtain the real-time monitoring result of the pressure parameter.

[0068] Exemplarily, in the embodiment of the present invention, in the SOS algorithm, the abnormal probability of each moment is calculated using the number of neighboring data of each moment, including: constructing a dissimilarity matrix through the Euclidean distance between each moment and the neighboring data; based on the dissimilarity matrix, calculating the correlation matrix between moments, and normalizing the correlation matrix into a probability distribution to obtain the abnormal probability of each moment.

[0069] Among them, when calculating the correlation matrix between moments based on the dissimilarity matrix, the distance in the dissimilarity matrix can be converted into a correlation value through a conversion function such as a Gaussian function, so as to obtain the correlation matrix. The specific steps of calculating the abnormal probability of each moment using the number of neighboring data of each moment in the SOS algorithm can be implemented by the prior art, and the embodiments of the present invention will not be elaborated herein.

[0070] Exemplarily, in the embodiments of the present invention, the number of neighboring data at each moment is used in the SOS algorithm to calculate the anomaly probability at each moment, so as to obtain the real-time monitoring result of the die-casting equipment parameters, including: if the anomaly probability at any moment is greater than the anomaly probability threshold, the real-time monitoring result of the die-casting equipment pressure parameter corresponding to this moment is abnormal; otherwise, the real-time monitoring result of the die-casting equipment pressure parameter corresponding to this moment is normal.

[0071] Among them, the anomaly probability threshold can be set to 0.7; the anomaly probability threshold can be specifically set according to actual needs, and the embodiments of the present invention do not limit it too much here.

[0072] It can be understood that if the real-time monitoring result of the die-casting equipment pressure parameter is abnormal, the quality of the corresponding casting may have problems. In this case, an anomaly prompt can be sent in time, and the abnormal casting can be intercepted to avoid being mixed with other castings.

[0073] Exemplarily, in the embodiments of the present invention, after obtaining the real-time monitoring result of the die-casting equipment parameters, it further includes: in response to the real-time monitoring result of the die-casting equipment pressure parameter being abnormal, sending an anomaly prompt and intercepting the casting corresponding to this moment.

[0074] Among them, the anomaly prompt method can be to give a voice prompt of the production batch of the casting and the position of the casting on the production line; the anomaly prompt method can be specifically set according to actual needs, and the embodiments of the present invention do not limit it too much here.

[0075] It can be seen that in the embodiments of the present invention, when obtaining the parameter monitoring result of the die-casting equipment, the critical point in the die-casting equipment pressure data time series can be determined by using the speed change condition of the die-casting equipment; the difference mean value of the injection pressures between each moment in the interval between any moment and the nearest critical point is used to obtain the injection pressure change fluctuation degree at this moment; the product of the ratio of the injection pressure change fluctuation degrees at this moment and the previous moment and the preset number of neighbors at this moment is recorded as the adjustment factor at this moment; if the adjustment factor at this moment is greater than the length of the interval where this moment is located, the length of the interval where this moment is located is used as the number of neighboring data at this moment; otherwise, the adjustment factor at this moment is rounded up to be used as the number of neighboring data at this moment; the number of neighboring data at each moment is used in the SOS algorithm to calculate the anomaly probability at each moment, so as to obtain the real-time monitoring result of the die-casting equipment parameters, effectively improving the accuracy of the parameter monitoring of the die-casting equipment.

[0076] The embodiments of the present invention also disclose a real-time monitoring system for die-casting equipment parameters, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time monitoring method for die-casting equipment parameters provided by the present invention is implemented.

[0077] The above system further includes other components well-known to those skilled in the art, such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0078] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0079] The above are all preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A real-time monitoring method for die-casting equipment parameters, characterized in that: include: Determine the critical point in the time series of pressure data of the die-casting equipment by using the speed change of the die-casting equipment; The fluctuation degree of the injection pressure change at any moment is obtained by taking the average difference of the injection pressure between any moment and each moment in the interval closest to the critical point; The product of the ratio of the injection pressure fluctuation degree at this moment to that at the previous moment and the preset number of neighbors at this moment is recorded as the adjustment factor at this moment; If the adjustment factor at the moment is greater than the length of the interval at the moment, the length of the interval at the moment is used as the number of neighboring data at the moment; otherwise, the adjustment factor at the moment is rounded up and used as the number of neighboring data at the moment; In the SOS algorithm, the number of neighboring data at each moment is used to calculate the abnormal probability at each moment to obtain the real-time monitoring results of the die-casting equipment parameters.

2. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The method of using the speed change of the die-casting equipment to determine the critical point in the pressure data time series of the die-casting equipment also includes: The injection pressure and speed values ​​of the die-casting equipment are acquired in real time at each moment, and a pressure data time series sequence and a speed data time series sequence of the die-casting equipment are obtained.

3. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The method of determining the critical point in the pressure data time series of the die-casting equipment by using the speed change of the die-casting equipment comprises: Calculate the speed mutation degree at the i-th moment : ； , , are respectively the i-th moment and the Time, Speed ​​value at the moment; By comparing the speed mutation degree between each moment and the moment before that moment, the moment corresponding to the critical point is determined.

4. A method for real-time monitoring of die-casting equipment parameters according to claim 3, characterized in that: Determining the moment corresponding to the critical point by comparing the speed mutation degree between each moment and the moment before the moment includes: If the degree of velocity mutation at a moment is greater than the degree of velocity mutation at the previous moment, then this moment is the moment corresponding to the critical point in the pressure data time series.

5. The method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The time difference between each moment and the moment corresponding to each critical point is taken as the distance between each moment and each critical point.

6. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The variation fluctuation degree of the injection pressure at any moment is obtained by taking the average of the differences between the injection pressure at any moment and each moment in the interval closest to the critical point, including: ； is the fluctuation degree of injection pressure at the i-th moment, is the number of moments in the interval between the i-th moment and the closest critical point, , are respectively the i-th moment and the closest critical point interval. moment, The injection pressure at a certain moment, is a linear normalization function.

7. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The SOS algorithm uses the number of neighbor data at each moment to calculate the abnormal probability at each moment, including: The dissimilarity matrix is ​​constructed through the Euclidean distance between each moment and the neighboring data; based on the dissimilarity matrix, the correlation matrix between the moments is calculated, and the correlation matrix is ​​normalized into a probability distribution to obtain the abnormal probability of each moment.

8. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The SOS algorithm uses the number of neighboring data at each moment to calculate the abnormal probability at each moment to obtain the real-time monitoring results of the die-casting equipment parameters, including: If the abnormal probability at any moment is greater than the abnormal probability threshold, the real-time monitoring result of the pressure parameter of the die-casting equipment corresponding to that moment is abnormal; otherwise, the real-time monitoring result of the pressure parameter of the die-casting equipment corresponding to that moment is normal.

9. A method for real-time monitoring of die-casting equipment parameters according to claim 1, characterized in that: The real-time monitoring results of the die-casting equipment parameters are obtained, and then the following steps are further included: In response to the real-time monitoring result of the pressure parameter of the die-casting equipment being abnormal, an abnormal prompt is issued and the casting corresponding to that moment is intercepted.

10. A real-time monitoring system for die-casting equipment parameters, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time monitoring method for parameters of a die-casting equipment according to any one of claims 1-9 is implemented.

Images