A PLC-based intelligent control method and system for die-casting machine production

Abstract

The present application relates to the field of metal casting technology, and specifically to a PLC-based intelligent control method and system for die-casting machine production. The method includes: constructing a standard speed curve, collecting actual speed and oil supply pressure; constructing an actual speed curve with the actual speed, and obtaining the die-casting progress based on the standard speed curve; obtaining a progress speed curve, and obtaining a non-compensated progress speed curve without speed control; obtaining the difference sequence of the two progress speed curves at different stages, thereby obtaining the stability of each stage; converting the actual speed curve into an actual progress speed curve, and obtaining the injection speed jitter frequency according to the die-casting progress and stability at the current moment; combining the two progress speed curves to obtain the phase characteristic value at the current moment; and combining the die-casting progress through a neural network to intelligently control the die-casting machine. The present application avoids the risk of errors caused by complex operations during the debugging process of technicians, and improves the quality of die-casting parts.

Description

Technical Field

[0001] The present application relates to the field of metal manufacturing technology, and in particular to a PLC-based intelligent control method and system for die-casting machine production. Background Art

[0002] Die casting is a common metal casting method, mainly used for model casting of various non-ferrous metals. It is formed by injecting molten metal into the model and cooling it. Due to its one-time molding and high work efficiency, it is widely used in the industrial field. In the die casting process, since the molten metal needs to be injected into the model to exclude air, bubbles may appear inside the molten metal during the injection process, which will lead to substandard quality of die-cast parts and waste of raw materials. Therefore, strict restrictions on the injection speed are required.

[0003] In the die casting process, a pressurizing device is usually used to push the hammer head, and the injection speed is controlled by controlling the pressure. The traditional injection speed control generates an injection speed curve according to the set parameters through the PLC setting program, so that the hammer head moves forward according to the injection speed curve to complete the injection; at the same time, the injection speed is compensated by PID control to maintain the stability of the injection speed.

[0004] However, due to the different models in the die-casting process, the designed injection speed curves are different. Therefore, each time a new model is replaced, technical personnel need to debug new PID parameters. The debugging process is complicated, the number of debugging times is large, the design is difficult, and design errors can easily lead to poor quality of die-casting parts. Summary of the invention

[0005] In order to solve the technical problem that different models need to be debugged multiple times and the design is difficult, this application provides a PLC-based die-casting machine production intelligent control method and system, and the technical solutions adopted are as follows:

[0006] In the first aspect, the present application proposes a PLC-based die-casting machine production intelligent control method, the method comprising the following steps:

[0007] The preset speed is used to build a standard speed curve. The preset speed is input into the PLC to move the hammer. The speed of the hammer when it moves is recorded as the actual speed and the oil supply pressure when the hammer moves.

[0008] Different stages of the standard speed curve are preset, the area of ​​each stage is obtained, the actual speed curve is constructed based on the actual speed at the current moment and all previous actual speeds, and the die-casting progress at the current moment is obtained based on the area ratio of the actual speed curve to the standard speed curve at different stages;

[0009] Convert the standard speed curve into a progress speed curve according to the die-casting progress at each moment; convert the actual speed curve without speed control into a non-compensated progress speed curve according to the die-casting progress; obtain the injection speed jitter sequence of each stage based on the difference between the progress speed curve and the non-compensated progress speed curve; obtain the injection speed jitter frequency according to the product of all frequencies and amplitudes in the injection speed jitter sequence of each stage, and obtain the stability of each stage according to the volatility of the injection speed jitter frequency and amplitude;

[0010] The actual speed curve is converted into the actual progress speed curve according to the die-casting progress at each moment, and the characteristic curve at the current moment is obtained according to the difference between the die-casting progress and stability at the current moment; the characteristic curve is transformed into a spectrum diagram through Fourier transform, and the injection speed jitter frequency at the current moment is obtained through the spectrum diagram; the phase characteristic value at the current moment is obtained by combining the zero point obtained by the difference between the actual progress speed curve and the progress speed curve with the die-casting progress difference at the current moment and the injection speed jitter frequency;

[0011] The die-casting machine is intelligently controlled through a neural network based on the current die-casting progress, injection speed jitter frequency, phase characteristic value and oil supply pressure.

[0012] In the above scheme, the die casting is divided into different stages according to the injection speed data in the die casting process, and the die casting progress is calculated according to the injection speed data; since the die casting progress can characterize the specific part of the die casting model filled with molten metal, and the injection resistance is similar when the same part of the die casting model is filled with molten metal, and the injection resistance is the main factor causing the jitter of the injection speed, the jitter of the injection curve under the same die casting progress is similar, so the die casting progress is used instead of time as the vertical axis of the injection speed curve, which can more accurately describe the periodic injection resistance in the repeated die casting process and obtain a better injection speed compensation effect. Further, the injection speed data is resampled according to the die casting progress, and the stage stability is obtained through the jitter of the resampled data, and different injection speed adjustment measures are adopted for different stages; the injection speed data is feature extracted according to the stage stability, and the extracted features are used as input, and the oil supply pressure is controlled by a neural network to complete the injection speed compensation of the injection speed. Since the injection speed compensation process of the present application is completed automatically, the debugging process of the traditional PID control for injection speed compensation can be omitted, thereby avoiding the risk of errors caused by the complex operations of technicians during the debugging process, improving the quality of die-casting parts, and accelerating the production efficiency of new die-casting models.

[0013] In one embodiment, the method for obtaining the die-casting progress at the current moment according to the area ratio of the actual speed curve and the standard speed curve at different stages is:

[0014] The area corresponding to the actual speed curve is recorded as , sort the areas of all stages in the standard speed curve in order to obtain an area sequence, and add up all the areas in the area sequence from the first to the last area. When the accumulated area exceeds When the accumulation stops, the serial number of the stage of the accumulated area is obtained, and the stage corresponding to the serial number is the stage at the current moment and is recorded as the current stage;

[0015] make The ratio of the area of ​​the current stage to the area of ​​all stages before the current stage is recorded as the stage progress, and the sum of the stage progress and the serial number of the previous stage of the current stage is recorded as the die-casting progress at the current moment.

[0016] In one embodiment, the method of converting the standard speed curve into the progress speed curve according to the die-casting progress at each moment; and converting the actual speed curve without speed control into the non-compensation progress speed curve according to the die-casting progress is:

[0017] The ratio of the area corresponding to the stage at each moment in the standard speed curve to the area of ​​the stage is recorded as the first ratio; the sum of the first ratio and the sequence number of the previous stage is used as the die-casting progress of the standard speed curve at each moment; the curve is redrawn with the speed in the standard speed curve as the vertical axis and the die-casting progress as the horizontal axis, and a number of sampling points are preset for the redrawn curve to obtain the progress speed curve;

[0018] Input the data in the standard speed curve, collect the hammer speed without speed control, and record the curve composed of the hammer speed in this case as the non-compensated speed curve; for each moment of the non-compensated speed curve, obtain the die-casting progress of the non-compensated speed curve at each moment according to the difference between the area corresponding to the non-compensated speed curve at each moment and the area of ​​the standard speed curve; redraw the curve with the speed in the non-compensated speed curve as the vertical axis and the die-casting progress as the horizontal axis, and preset several sampling points to obtain the non-compensated progress speed curve.

[0019] In one embodiment, the method for obtaining the injection speed jitter sequence of each stage based on the difference between the progress speed curve and the non-compensated progress speed curve is:

[0020] For the non-compensated progress speed curve and the progress speed curve, the speed sequences corresponding to all the sampling points in each stage are obtained respectively and recorded as the non-compensated speed sequence and progress speed sequence of each stage; the speed of the progress speed sequence of the same stage is subtracted from the speed of the non-compensated speed sequence at the same time, and all the differences are used to construct the injection speed jitter sequence corresponding to each stage.

[0021] In one embodiment, the method of obtaining the injection speed jitter frequency according to the product of all frequencies and amplitudes in the injection speed jitter sequence of each stage and obtaining the stability of each stage according to the volatility of the injection speed jitter frequency and amplitude is:

[0022] For any stage, the injection speed jitter sequence is converted into a spectrum diagram through Fourier algorithm; the data storage form of the spectrum diagram is a matrix with two rows and multiple columns, the first row of the same column represents the frequency size, and the second row represents the amplitude corresponding to the frequency;

[0023] Normalize all amplitudes to obtain normalized amplitudes, accumulate the product of the frequency of each column and the normalized amplitude corresponding to the frequency to obtain the jitter frequency of the injection speed in the stage; take the standard deviation of the normalized amplitudes of all frequencies as the jitter stability of the injection speed;

[0024] The stability of each stage is obtained based on the jitter stability and injection speed jitter frequency of each stage.

[0025] In one embodiment, the method for obtaining the stability of each stage according to the jitter stability and the injection speed jitter frequency is as follows:

[0026]

[0027] represents the jitter stability of the injection speed in the rth stage, Indicates the jitter frequency of the injection speed in the rth stage, represents the stability of the rth stage.

[0028] In one embodiment, the method of converting the actual speed curve into the actual progress speed curve according to the die-casting progress at each moment, and obtaining the characteristic curve at the current moment according to the difference between the die-casting progress and stability at the current moment is:

[0029] For the actual speed curve, after obtaining the die-casting progress at each moment, the actual speed curve is converted into an actual progress speed curve, the abscissa of the actual progress speed curve is the die-casting progress, and the ordinate is the injection speed, wherein the actual progress speed curve is resampled according to the horizontal axis spacing of the progress speed curve;

[0030] For the actual progress speed curve, the current stage is recorded as the target stage, and the cycle progress is obtained by subtracting the stability of the current die-casting stage from that of the target stage. The characteristic curve at the current moment is determined according to the size of the cycle progress.

[0031] In one embodiment, the method of obtaining the cycle progress by subtracting the stability of the die-casting stage at the current moment from that of the target stage and determining the characteristic curve at the current moment according to the size of the cycle progress is:

[0032] The cycle progress is obtained by subtracting the die-casting progress at the current moment from the stability of the target stage;

[0033] If the cycle progress is greater than 0, the curve from the cycle progress to the die-casting progress at the current moment in the actual progress speed curve is recorded as the characteristic curve at the current moment;

[0034] If the cycle progress is less than or equal to 0, the curve between the die-casting progress of 0 in the target stage and the die-casting progress at the current moment is recorded as the characteristic curve at the current moment.

[0035] In one embodiment, the method of obtaining the phase characteristic value at the current moment by combining the difference between the die-casting progress at the zero point and the current moment obtained by the difference between the actual progress speed curve and the progress speed curve with the injection speed jitter frequency is:

[0036] The length of the progress speed curve is adjusted so that a split progress speed curve is constructed from the first sampling point to the sampling point with the same length as the actual progress speed curve, and the jitter curve is obtained by subtracting the actual progress speed curve from the split progress speed curve;

[0037] In the jitter curve, find the first zero point before the current moment, and record the moment after this zero point as the zero-crossing moment. If the ordinate of the zero-crossing moment is a positive number, the current moment is in the first half of the cycle; if the ordinate of the zero-crossing moment is a negative number, the current moment is in the second half of the cycle; if the current moment is the zero point, there is no need to find the first zero point before it. At this time, the speed at the current moment has no jitter, and the phase characteristic value is obtained based on this;

[0038] The expression of the phase eigenvalue is:

[0039]

[0040] In the formula, is the phase eigenvalue at time t, is the die casting progress at time t, is the die-casting progress of the first zero point before time t, is the injection speed cycle, and time t is the current time.

[0041] In the second aspect, an embodiment of the present application also provides a PLC-based intelligent control system for die-casting machine production, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned PLC-based intelligent control methods for die-casting machine production are implemented.

[0042] The beneficial effects of this application are as follows:

[0043] Die casting is divided into different stages according to the injection speed data in the die casting process, and the die casting progress is calculated according to the injection speed data; since the die casting progress can characterize the specific parts of the die casting model filled with molten metal, and the injection resistance is similar when the same parts of the die casting model are filled with molten metal, and the injection resistance is the main factor causing the jitter of the injection speed, the jitter of the injection curve under the same die casting progress is similar, so the die casting progress is used instead of time as the vertical axis of the injection speed curve, which can more accurately describe the periodic injection resistance in the repeated die casting process and obtain a better injection speed compensation effect. Further, the injection speed data is resampled according to the die casting progress, and the stage stability is obtained through the jitter of the resampled data, and different injection speed adjustment measures are adopted for different stages; the injection speed data is feature extracted according to the stage stability, and the extracted features are used as input, and the oil supply pressure is controlled by a neural network to complete the injection speed compensation of the injection speed. Since the injection speed compensation process of the present application is completed automatically, the debugging process of the traditional PID control for injection speed compensation can be omitted, thereby avoiding the risk of errors caused by the complex operations of technicians during the debugging process, improving the quality of die-casting parts, and accelerating the production efficiency of new die-casting models. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A flow chart of a PLC-based intelligent control method for die-casting machine production provided in one embodiment of the present application;

[0046] Figure 2 This is the standard speed curve. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a PLC-based die-casting machine production intelligent control method and system proposed in the present application in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0049] A PLC-based die-casting machine production intelligent control method and system embodiment:

[0050] The specific scheme of a PLC-based intelligent control method and system for die-casting machine production provided by the present application is described in detail below with reference to the accompanying drawings.

[0051] See also Figure 1 , which shows a PLC-based die-casting machine production intelligent control method and system flow chart provided by an embodiment of the present application, the method comprising the following steps:

[0052] Step S001, a standard speed curve is constructed by preset speed, the preset speed is input into PLC to move the hammer head, and the speed of the hammer head when moving is recorded as the actual speed and the oil supply pressure when the hammer head moves.

[0053] In the die casting process, a pressurizing device is usually used to push the hammer head, and the injection speed is controlled by controlling the pressure. A displacement sensor is used to monitor the actual injection speed of the hammer head to obtain the actual speed of the hammer head. The displacement sensor used in this embodiment is Balluff BML-S1B0-Q53F-M410-H0-KA02.

[0054] The pressure of the injection cylinder is obtained by a pressure sensor connected to the injection cylinder of the die-casting machine, and the oil supply pressure is obtained.

[0055] In this embodiment, the collection period of the actual speed and the oil supply pressure collected by the sensor is 1 ms, which can be set by the implementer.

[0056] The injection speed at each moment is set. Different standard speed curves correspond to different molds. In this embodiment, the first five speed values ​​of the die-casting machine are 0.272m / s, 0.274m / s, 0.275m / s, 0.276m / s, and 0.277m / s.

[0057] The injection speed at all times is input into a curve and recorded as the standard speed curve. The PLC control program of the die-casting machine makes the hammer head move according to the injection speed. During the movement of the hammer head, the displacement sensor and pressure sensor are used to collect the actual speed and oil supply pressure. The horizontal axis of the standard speed curve is time, and the vertical axis is speed. The standard speed curve is as follows: Figure 2 The standard speed curve is to control the injection speed of the hammer head under ideal conditions.

[0058] Figure 2Among them, V1 is the hammer head shooting speed, T1 is the shooting time, B is the slow acceleration stage, C is the slow uniform speed stage, B and C constitute the slow shooting stage, D is the acceleration shooting stage, E is the fast shooting stage, F is the deceleration I stage, G is the deceleration II stage, and F and G constitute the deceleration stage.

[0059] At this point, the actual speed, fuel supply pressure and standard speed curve are acquired.

[0060] Step S002, preset different stages of the standard speed curve, obtain the area of ​​each stage, construct an actual speed curve with the actual speeds at the current moment and all previous ones, and obtain the die-casting progress at the current moment according to the area ratio of the actual speed curve to the standard speed curve at different stages.

[0061] Under ideal conditions, the standard speed curve can ensure stable injection speed and avoid bubbles caused by oscillation of the molten metal. However, in the actual injection process, due to the uneven internal space of the die-casting model, the oil supply pressure needs to be changed in real time to ensure the relative stability of the injection speed.

[0062] The traditional control method uses PID to control the oil supply pressure to ensure that the injection speed is relatively stable. This method is called injection speed compensation. When facing different die-casting models, the instability of the injection speed will change, and then the PID control parameters need to be re-adjusted. The debugging is difficult and prone to errors, resulting in a decrease in the quality of die-casting parts.

[0063] Therefore, the present application extracts the features that cause unstable injection speed, uses a neural network to control the oil supply pressure, avoids PID parameter debugging, and improves the quality of die-casting parts.

[0064] In a die casting process, die casting is usually divided into: slow stage, acceleration stage, fast stage, deceleration stage, pressurization node, pressure holding stage, shot termination and mold opening stage.

[0065] In different stages, the instability of the injection speed is different. If the injection speed needs to be stabilized with high precision, different stages must be distinguished. For the same die-casting model, the different stages of die-casting can be characterized by time, but the time for a die-casting of different die-casting models may be set according to the specific circumstances of the model. Therefore, this embodiment uses the die-casting progress to uniformly characterize which stage the current die-casting has entered.

[0066] In the die-casting process, when die-casting reaches a certain extent, the part filled with molten metal is the same, and the main factor causing the jitter of the injection speed is the injection resistance generated by the irregular die-casting model on the injection of metal liquid. When the injection reaches a certain part filled with molten metal, the injection resistance is similar. Therefore, the die-casting progress can more accurately describe the periodic injection resistance in the repeated die-casting process.

[0067] Since the vertical axis of the standard speed curve is the hammer moving speed and the horizontal axis is time, the area below the curve represents the forward distance of the hammer, and the forward distance of the hammer is proportional to the compression space of the injection cavity. The greater the forward distance, the greater the die casting progress. In this application, the forward distance of the hammer is used to characterize the die casting progress, where the area is the area between the straight line with the vertical coordinate of 0 and the curve.

[0068] Since the speed value of the die casting machine is a manually set value, the time of different stages of the standard speed curve can be obtained, and the stages where the injection speed jitter needs to be prevented include: slow acceleration stage, slow uniform speed stage, accelerated injection stage, fast injection stage, deceleration stage I, and deceleration stage II, a total of six stages. Therefore, the area corresponding to each stage can be obtained, and the area of ​​the rth stage is recorded as .

[0069] For the actual speed collected, the actual speed curve is obtained with speed as the vertical axis and time as the horizontal axis. For the actual speed curve corresponding to the current moment, the area corresponding to the actual speed curve is obtained and recorded as , the die-casting progress at the current moment is obtained according to the difference between the area of ​​the actual speed curve and the area of ​​the standard speed curve.

[0070] Preferably, in this embodiment, the areas of all stages are sorted in order of the stages to obtain an area sequence, and all areas in the area sequence are accumulated from the first to the last area. When the accumulated area exceeds When , the accumulation stops, and the serial number of the stage of the accumulated area is obtained. The stage corresponding to the serial number is the stage at the current moment and is recorded as the current stage. The ratio of the area of ​​the current stage to the area of ​​all stages before the current stage is recorded as the stage progress, and the sum of the stage progress and the serial number of the previous stage of the current stage is recorded as the die-casting progress at the current moment.

[0071] For example, the area sequence is {20, 37, 43, 328, 35, 18}, and its corresponding serial number is {1, 2, 3, 4, 5, 6}. The corresponding area at the current moment is 78, so the stage progress is: , the current stage is the third stage, and the sequence number of the previous stage is 2, so the die casting progress is .

[0072] At this point, the die-casting progress at the current moment is obtained.

[0073] Step S003, converting the standard speed curve into a progress speed curve according to the die-casting progress at each moment; converting the actual speed curve without speed control into a non-compensated progress speed curve according to the die-casting progress; obtaining the injection speed jitter sequence of each stage based on the difference between the progress speed curve and the non-compensated progress speed curve; obtaining the injection speed jitter frequency according to the product of all frequencies and amplitudes in the injection speed jitter sequence of each stage, and obtaining the stability of each stage according to the volatility of the injection speed jitter frequency and amplitude.

[0074] When the molten metal is injected, when it is filled into different parts of the model, different injection resistances will appear due to the width of the filling parts of the model, which will lead to unstable injection speed.

[0075] In different stages, the injection speed is different, the model filling speed is different, resulting in different changes in the injection resistance. When the injection speed is low, the injection position changes slowly, and the injection resistance caused by the width of the model is small. At this time, the jitter of the injection speed is light. On the contrary, when the injection speed is fast, the jitter of the injection speed is more serious. At the same time, the places where the molten metal is filled are different at different stages, and the jitter of the injection speed is also different. For example, in the slow stage, the air in the compression cavity is mainly discharged instead of the air inside the model. At this time, there is no irregular model wall, and the jitter of the injection speed is very small.

[0076] Therefore, the stage stability is calculated for different stages, so as to adopt different injection speed adjustment measures for different stages.

[0077] The standard speed curve is resampled, and the ratio of the area corresponding to the stage at each moment in the standard speed curve to the area of ​​the stage is recorded as the first ratio. The die-casting progress of the standard speed curve at each moment is obtained by summing the first ratio and the sequence number of the previous stage. The curve is redrawn with the speed in the standard speed curve as the vertical axis and the die-casting progress as the horizontal axis. U sampling points are evenly set for the drawn curve, which is recorded as the progress speed curve. In this embodiment, U is 2000, and the number of sampling points can be set by the implementer.

[0078] Afterwards, for all the idealized speed data, for the new die-casting model, a die-casting is performed without using injection speed compensation, and the actual speed curve of the hammer head at this time is obtained and recorded as the non-compensated speed curve. For each moment of the non-compensated speed curve, the die-casting progress of the non-compensated speed curve at each moment is obtained according to the difference between the area corresponding to the non-compensated speed curve at each moment and the area of ​​the standard speed curve.

[0079] The non-compensated speed curve is resampled, and the curve is drawn with the speed as the vertical axis and the die-casting progress as the horizontal axis. Similarly, U sampling points are set for the drawn curve, which is recorded as the non-compensated progress speed curve.

[0080] For the non-compensated progress speed curve and the progress speed curve, the speed sequences corresponding to all sampling points in each stage are obtained respectively and recorded as the non-compensated speed sequence and the progress speed sequence of each stage.

[0081] The speed of the progress speed sequence in the same stage is subtracted from the speed of the non-compensated speed sequence at the same time, and all the differences are used to construct the injection speed jitter sequence corresponding to this stage, which characterizes the change of the injection speed jitter with the injection progress in this stage.

[0082] For any stage, the injection speed jitter sequence is taken as input and calculated using the fast Fourier transform algorithm, and the output is the corresponding spectrum diagram. The data storage form of the spectrum diagram is a matrix with two rows and multiple columns. The first row of the same column represents the frequency size, and the second row represents the amplitude corresponding to the frequency. The fast Fourier transform algorithm is a well-known technology and will not be described in detail in this application.

[0083] Preferably, in this embodiment, all amplitudes are normalized to obtain the normalized amplitude, and the frequency of each column and the normalized amplitude corresponding to the frequency are multiplied and accumulated to obtain the injection speed jitter frequency of the stage. , where represents the frequency of the nth column, represents the normalized amplitude corresponding to the frequency of the nth column, N represents the number of columns in the spectrum graph, Indicates the injection speed jitter frequency of the stage.

[0084] It should be noted that the maximum and minimum value normalization is a method for normalizing the amplitude, and the implementer can use other normalization methods for processing, such as the linear normalization method.

[0085] The stability of the injection speed is obtained based on the amplitude of all frequencies.

[0086] Preferably, in this embodiment, for the spectrum diagram, the standard deviation of the normalized amplitudes of all frequencies is obtained, and the standard deviation is normalized by the maximum and minimum values ​​and recorded as the jitter stability of the injection speed.

[0087] Preferably, in another embodiment, for the spectrum diagram, the absolute value difference between the normalized amplitude of each frequency and the next normalized amplitude is calculated, and the average of all the calculated absolute value differences is used as the jitter stability of the injection speed after linear normalization.

[0088] The stability of each stage is obtained based on the jitter stability and injection speed jitter frequency of each stage.

[0089] Preferably, in this embodiment, the inverse of the product of the jitter stability of the injection speed and the jitter frequency of the injection speed is used as the stability of the stage. . represents the jitter stability of the injection speed in the rth stage, Indicates the jitter frequency of the injection speed in the rth stage, represents the stability of the rth stage.

[0090] Among them, the inverse of the jitter frequency of the injection speed is the jitter period of the injection speed, and the jitter of the injection speed is repeated in cycles; the smaller the jitter stability, the more stable the injection speed, the greater the stability of the stage, and the longer the jitter data can be used as a reference to characterize the jitter situation at the current moment.

[0091] Finally, the greater the stability of a stage, the more stable the jitter of the injection speed in that stage; further, for the jitter of the injection speed at the current moment, the data before the die-casting progress corresponding to the current moment can be used to complete the feature extraction of the injection speed at that moment.

[0092] So far, the stability of each stage has been obtained.

[0093] Step S004, converting the actual speed curve into the actual progress speed curve according to the die-casting progress at each moment, and obtaining the characteristic curve at the current moment according to the difference between the die-casting progress and stability at the current moment; obtaining a spectrum diagram from the characteristic curve through Fourier transform, and obtaining the injection speed jitter frequency at the current moment through the spectrum diagram; obtaining the phase characteristic value at the current moment by combining the difference between the zero point obtained by the difference between the actual progress speed curve and the progress speed curve and the die-casting progress at the current moment with the injection speed jitter frequency.

[0094] When die casting is in different stages, different hammer head injection speed data can be selected according to the stability of the stage to extract the injection speed jitter characteristics;

[0095] Since the stability of the stage is obtained by the inverse of the frequency, and the inverse of the frequency is the period, the stability of the stage can be regarded as the period of the stage.

[0096] For the actual speed curve, after obtaining the die-casting progress at each moment, the actual speed curve is converted into the actual progress speed curve, the horizontal axis of the actual progress speed curve is the die-casting progress, and the vertical axis is the injection speed, wherein the actual progress speed curve is resampled according to the horizontal axis spacing of the progress speed curve. For example, at this time, the die-casting progress of the last point of the actual progress speed is 3.5, and when the die-casting progress of the progress speed curve is 3.5, there are 1200 sampling points before, then the actual progress speed curve is evenly divided through 1200 sampling points.

[0097] For the actual progress speed curve, the stage at the current moment is recorded as the target stage, and the cycle progress is obtained by subtracting the stability of the die-casting progress at the current moment from that of the target stage. If the cycle progress is greater than 0, the curve from the cycle progress to the die-casting progress at the current moment in the actual progress speed curve is recorded as the characteristic curve at the current moment; if the cycle progress is less than or equal to 0, the curve from the die-casting progress of 0 in the target stage to the die-casting progress at the current moment is recorded as the characteristic curve at the current moment.

[0098] Furthermore, a Fourier transform algorithm is used to obtain a corresponding frequency spectrum of the characteristic curve at the current moment, and the injection speed jitter frequency of the frequency spectrum is calculated, and the inverse of the injection speed jitter frequency at the current moment is taken as the injection speed period at the current moment.

[0099] The length of the progress speed curve is adjusted so that a split progress speed curve is constructed from the first sampling point to the sampling point with the same length as the actual progress speed curve, and the jitter curve is obtained by subtracting the actual progress speed curve from the split progress speed curve.

[0100] In the jitter curve, find the first zero point before the current moment, and record the moment after this zero point as the zero-crossing moment. If the vertical coordinate of the zero-crossing moment is a positive number, the current moment is in the first half of the cycle; if the vertical coordinate of the zero-crossing moment is a negative number, the current moment is in the second half of the cycle; if the current moment is zero, there is no need to find the first zero point before it. At this time, it is considered that the speed at the current moment has no jitter, and the phase characteristic value is obtained based on this.

[0101] Depending on the current location, different phase eigenvalues ​​are obtained:

[0102]

[0103] In the formula, is the phase eigenvalue at time t, is the die casting progress at time t, is the die-casting progress of the first zero point before time t, is the injection speed cycle, and time t is the current time.

[0104] Among them, the zero point is used as the basis for judging the phase, the injection speed period is used as the jitter period, the approximate phase is calculated as the phase characteristic value, and the phase characteristic value at the current moment is obtained.

[0105] At this point, the injection speed jitter frequency and phase characteristic values ​​at the current moment are obtained.

[0106] Step S005, intelligently controlling the die-casting machine through a neural network based on the die-casting progress, injection speed jitter frequency, phase characteristic value and oil supply pressure at the current moment.

[0107] According to the traditional PLC-based die-casting machine that uses PID as injection speed compensation, data is collected. The die-casting progress, injection speed jitter frequency, phase characteristic value and oil supply pressure at different times are obtained based on PID control.

[0108] The die-casting progress, injection speed jitter frequency, and phase eigenvalues ​​at multiple moments are taken as the input of the fully connected network. The loss function is the mean square error of the oil supply pressure. The oil supply pressure control value at the next moment is obtained to complete the network training.

[0109] Through the above steps, the die-casting progress, injection speed jitter frequency, phase characteristic value and oil supply pressure at the current moment are obtained.

[0110] The current die-casting progress, injection speed jitter frequency, and phase eigenvalue are input into the trained network, and the output is a signal for controlling the oil pressure, so that the oil supply pressure at the next moment is controlled based on the current oil supply pressure, thus completing the intelligent control of the die-casting machine.

[0111] Based on the same inventive concept as the above method, an embodiment of the present invention also provides a PLC-based die-casting machine production intelligent control system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned PLC-based die-casting machine production intelligent control methods are implemented.

[0112] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0113] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A PLC-based intelligent control method for die-casting machine production, characterized in that: The method comprises the following steps: The preset speed is used to build a standard speed curve. The preset speed is input into the PLC to move the hammer. The speed of the hammer when it moves is recorded as the actual speed and the oil supply pressure when the hammer moves. Different stages of the standard speed curve are preset, the area of ​​each stage is obtained, the actual speed curve is constructed based on the actual speed at the current moment and all previous actual speeds, and the die-casting progress at the current moment is obtained based on the area ratio of the actual speed curve to the standard speed curve at different stages; Convert the standard speed curve into a progress speed curve according to the die-casting progress at each moment; convert the actual speed curve without speed control into a non-compensated progress speed curve according to the die-casting progress; obtain the injection speed jitter sequence of each stage based on the difference between the progress speed curve and the non-compensated progress speed curve; obtain the injection speed jitter frequency according to the product of all frequencies and amplitudes in the injection speed jitter sequence of each stage, and obtain the stability of each stage according to the volatility of the injection speed jitter frequency and amplitude; The actual speed curve is converted into the actual progress speed curve according to the die-casting progress at each moment, and the characteristic curve at the current moment is obtained according to the difference between the die-casting progress and stability at the current moment; the characteristic curve is transformed into a spectrum diagram through Fourier transform, and the injection speed jitter frequency at the current moment is obtained through the spectrum diagram; the phase characteristic value at the current moment is obtained by combining the zero point obtained by the difference between the actual progress speed curve and the progress speed curve with the die-casting progress difference at the current moment and the injection speed jitter frequency; The die-casting machine is intelligently controlled through a neural network based on the current die-casting progress, injection speed jitter frequency, phase characteristic value and oil supply pressure; The method of converting the standard speed curve into the progress speed curve according to the die-casting progress at each moment; and converting the actual speed curve without speed control into the non-compensation progress speed curve according to the die-casting progress is: The ratio of the area corresponding to the stage at each moment in the standard speed curve to the area of ​​the stage is recorded as the first ratio; the sum of the first ratio and the sequence number of the previous stage is used as the die-casting progress of the standard speed curve at each moment; the curve is redrawn with the speed in the standard speed curve as the vertical axis and the die-casting progress as the horizontal axis, and a number of sampling points are preset for the redrawn curve to obtain the progress speed curve; Input the data in the standard speed curve, collect the hammer speed without speed control, and record the curve composed of the hammer speed in this case as the non-compensated speed curve; for each moment of the non-compensated speed curve, obtain the die-casting progress of the non-compensated speed curve at each moment according to the difference between the area corresponding to the non-compensated speed curve at each moment and the area of ​​the standard speed curve; redraw the curve with the speed in the non-compensated speed curve as the vertical axis and the die-casting progress as the horizontal axis, and preset a number of sampling points to obtain the non-compensated progress speed curve; The method for obtaining the injection speed jitter sequence of each stage based on the difference between the progress speed curve and the non-compensated progress speed curve is: For the non-compensated progress speed curve and the progress speed curve, the speed sequences corresponding to all sampling points in each stage are obtained respectively and recorded as the non-compensated speed sequence and progress speed sequence of each stage; the speed of the progress speed sequence of the same stage is subtracted from the speed of the non-compensated speed sequence at the same time, and all the differences are used to construct the injection speed jitter sequence corresponding to each stage; The method of obtaining the injection speed jitter frequency according to the product of all frequencies and amplitudes in the injection speed jitter sequence of each stage and obtaining the stability of each stage according to the fluctuation of the injection speed jitter frequency and amplitude is as follows: For any stage, the injection speed jitter sequence is converted into a spectrum diagram through Fourier algorithm; the data storage form of the spectrum diagram is a matrix with two rows and multiple columns, the first row of the same column represents the frequency size, and the second row represents the amplitude corresponding to the frequency; Normalize all amplitudes to obtain normalized amplitudes, accumulate the product of the frequency of each column and the normalized amplitude corresponding to the frequency to obtain the jitter frequency of the injection speed in the stage; take the standard deviation of the normalized amplitudes of all frequencies as the jitter stability of the injection speed; The stability of each stage is obtained based on the jitter stability and injection speed jitter frequency of each stage; The method of converting the actual speed curve into the actual progress speed curve according to the die-casting progress at each moment, and obtaining the characteristic curve at the current moment according to the difference between the die-casting progress and stability at the current moment is: For the actual speed curve, after obtaining the die-casting progress at each moment, the actual speed curve is converted into an actual progress speed curve, the abscissa of the actual progress speed curve is the die-casting progress, and the ordinate is the injection speed, wherein the actual progress speed curve is resampled according to the horizontal axis spacing of the progress speed curve; For the actual progress speed curve, the current stage is recorded as the target stage, and the cycle progress is obtained by subtracting the stability between the current die-casting stage and the target stage. The characteristic curve at the current moment is determined according to the size of the cycle progress. The method of obtaining the phase characteristic value at the current moment by combining the difference between the zero point obtained by the difference between the actual progress speed curve and the progress speed curve and the die casting progress at the current moment with the injection speed jitter frequency is as follows: The length of the progress speed curve is adjusted so that a split progress speed curve is constructed from the first sampling point to the sampling point with the same length as the actual progress speed curve, and the jitter curve is obtained by subtracting the actual progress speed curve from the split progress speed curve; In the jitter curve, find the first zero point before the current moment, and record the moment after this zero point as the zero-crossing moment. If the ordinate of the zero-crossing moment is a positive number, the current moment is in the first half of the cycle; if the ordinate of the zero-crossing moment is a negative number, the current moment is in the second half of the cycle; if the current moment is the zero point, there is no need to find the first zero point before it. At this time, the speed at the current moment has no jitter, and the phase characteristic value is obtained based on this; The expression of the phase eigenvalue is: In the formula, is the phase eigenvalue at time t, is the die casting progress at time t, is the die-casting progress of the first zero point before time t, is the injection speed cycle, and time t is the current time.

2. The PLC-based intelligent control method for die-casting machine production as claimed in claim 1, characterized in that: The method for obtaining the die-casting progress at the current moment according to the area ratio of the actual speed curve and the standard speed curve at different stages is: The area corresponding to the actual speed curve is recorded as , sort the areas of all stages in the standard speed curve in order to obtain an area sequence, and add up all the areas in the area sequence from the first to the last area. When the accumulated area exceeds When the accumulation stops, the serial number of the stage of the accumulated area is obtained, and the stage corresponding to the serial number is the stage at the current moment and is recorded as the current stage; make The ratio of the area of ​​the current stage to the area of ​​all stages before the current stage is recorded as the stage progress, and the sum of the stage progress and the serial number of the previous stage of the current stage is recorded as the die-casting progress at the current moment.

3. The PLC-based intelligent control method for die-casting machine production as claimed in claim 1, characterized in that: The method for obtaining the stability of each stage according to the jitter stability and the jitter frequency of the injection speed is: represents the jitter stability of the injection speed in the rth stage, Indicates the jitter frequency of the injection speed in the rth stage, represents the stability of the rth stage.

4. The PLC-based intelligent control method for die-casting machine production as claimed in claim 1, characterized in that: The method of obtaining the cycle progress by making a difference between the stability of the die-casting stage at the current moment and the stability of the target stage, and determining the characteristic curve at the current moment according to the size of the cycle progress is as follows: The cycle progress is obtained by subtracting the die-casting progress at the current moment from the stability of the target stage; If the cycle progress is greater than 0, the curve from the cycle progress to the die-casting progress at the current moment in the actual progress speed curve is recorded as the characteristic curve at the current moment; If the cycle progress is less than or equal to 0, the curve between the die-casting progress of 0 in the target stage and the die-casting progress at the current moment is recorded as the characteristic curve at the current moment.

5. A PLC-based intelligent control system for die-casting machine production, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the intelligent control method for die-casting machine production based on PLC as described in any one of claims 1-4 are implemented.

Images