JPH0596591A - Controller for injection molding machine - Google Patents

Abstract

PURPOSE:To reduce the adjusting operation of an operator largely by providing a fuzzy arithmetic section with a waveform database, a rule database and a gravity- center arithmetic section arithmetically operating the center of gravity of the membership functions of each rule. CONSTITUTION:The controller of an injection molding machine has a fuzzy arithmetic section 20 receiving the command values and detecting values of each operation condition, computing correction values at every operation condition, adding the correction values to the corresponding command values and outputting the added values as adjusting values and a control section 30 receiving the adjusting values and detecting values of each operation condition and computing manipulated variables at every operation condition. The fuzzy arithmetic section 20 receives a waveform database, in which the optimum-state transition waveform of a molten resin is stored at every molded form, a rule database, in which the heuristics of an operator is stored at every operation condition, and data from the waveform data-base and the rule database, obtains the membership functions of rules corresponding at every operation condition, and arithmetically operates the center of gravity of the synthetic membership function of the maximum values of the membership functions by agravity-center arithmetic section.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a controller for an injection molding machine, and more particularly to a controller utilizing fuzzy inference.

[0002]

2. Description of the Related Art Generally, injection molding is carried out by the steps of plasticizing resin, injecting, holding pressure, and cooling. In order to make the weight, size, photoelasticity, etc. of the molded product obtained through these processes uniform, it is necessary to improve the reproducibility of the resin state transition (transition of resin specific volume, temperature, pressure). Will be needed. In addition,
The specific volume is the reciprocal of the density of the molded product, and the holding pressure is the resin pressure inside the mold controlled in the holding step.

By the way, it is known that the relationship between the specific volume of resin in the molten state-the temperature-the pressure is represented by the following equation of state of Spencer. (P + π) (V−ω) = R _m T where P is the resin pressure, V is the specific volume, T is the resin temperature, π,
ω and R _m are constants determined according to the material.

This relationship is expressed in the three-dimensional plane of P, V, and T as shown in FIG. 4, which is shown by the solid line in FIG. 4 in one molding cycle of plasticization → injection → holding pressure → cooling described above. Draw a trace. It is desirable that this locus be the same even when the injection molding operation is repeated.

[0005]

However, in reality, since the resin temperature fluctuates from the start to the end of the molding cycle, it deviates from the locus shown by the solid line in FIG. Such a trajectory is drawn, and as a result, the quality of the molded product varies. This is because when the resin temperature changes, the specific volume in the reservoir in the heating cylinder immediately before the start of injection changes, and the initial condition (initial state) of the resin changes when filling the mold.

Conventionally, an operator estimates the flow state at the time of plasticizing the resin from the molding conditions such as the type of resin, the type of molded product, the mold temperature and the empirical rule, and the temperature of the heating cylinder, the screw rotation speed, the back pressure. Roughly set the operating conditions. After that, the operator gradually adjusts each operating condition while performing the molding to perform the molding. For this reason, the operator of the injection molding machine is not only required to have the skill, but also the adjustment work requires a lot of labor because the respective operating conditions are mutually related. An object of the present invention is to provide a control device for an injection molding machine that can significantly reduce the operator's adjustment work by incorporating the empirical rule of the operator into the control by utilizing fuzzy reasoning.

[0007]

SUMMARY OF THE INVENTION The present invention relates to an injection molding machine control device for performing injection molding based on a command value and a detected value of operating conditions such as back pressure, screw rotation speed, cylinder temperature and the like. A fuzzy calculator that receives a command value and a detected value of a condition, calculates a correction value for each operating condition, adds the correction value to the corresponding command value and outputs the adjusted value, and receives the adjusted value and the detected value of each operating condition. The fuzzy operation unit includes a waveform database storing the optimum state transition waveform of the molten resin for each molded product, and an operator's empirical rule for each operating condition. And a data base from which the waveform database and the rule database are stored to obtain a membership function of the rule corresponding to each operating condition, and Characterized in that it comprises a gravity calculation unit for calculating the center of gravity of the synthetic membership function synthesized by taking the maximum value of the membership function.

[0008]

Embodiments of the present invention will be described with reference to the drawings. Referring to FIG. 1, the control system of the injection molding machine 10 includes a fuzzy calculation unit 20, a control unit 30, and a molding product type designation unit 4
It consists of 0 and. The injection molding machine 10 includes a heating cylinder 100.
A screw 101 is rotatably and reciprocally disposed inside the screw 101. The screw 101 is driven by an injection cylinder 102 so as to reciprocate, and is rotationally driven by a rotary drive system 103. By rotating and reciprocating the screw 101, the resin in the heating cylinder 100 is filled in the cavity 105 of the mold 104 and pressure-molded.

In this embodiment, the case where the temperature of the heating cylinder 100, the back pressure in the injection cylinder 102, and the rotational speed of the rotary drive system 103, that is, the screw rotational speed are used as operating conditions will be described. Therefore, a detector (not shown) is provided in each of the heating cylinder 100, the injection cylinder 102, and the rotary drive system 103, and the temperature detection value T
D, the back pressure detection value PD, and the rotation speed detection value RD are respectively
It is given to the fuzzy operation unit 20 and the control unit 30.

The fuzzy calculator 20 is also supplied with a temperature command value TI of the heating cylinder 100, a back pressure command value PI of the injection cylinder 102, and a screw rotation speed command value RI. In response to the detected value and the command value of each operating condition, the fuzzy operation unit 20 calculates a correction value for each operating condition, and
Adding each correction value to the corresponding command value, the temperature adjustment value TC,
It is output as a back pressure adjustment value PC and a rotation speed adjustment value RC.

The control unit 30 receives the detected value and the adjusted value of each operating condition to calculate the operating amount for each operating condition, and the temperature operating amount TO of the heating cylinder 100, the back pressure operating amount PO of the injection cylinder 102, The rotation speed manipulated variable RO of the screw is output. The heating cylinder 100, the injection cylinder 102, and the rotary drive system 103 each operate based on the corresponding operation amount. The molding product type designation unit 4O is for designating a predetermined type of molded product.

Next, the fuzzy calculator 20 will be described with reference to FIG. The fuzzy operation unit 20 includes a waveform database 21 that stores optimum state transition waveforms of molten resin (see FIG. 4) for each molded product, a rule database 22 that stores an empirical rule of an operator for each operating condition, and a waveform database 21. , Receives the data from the rule database 22, obtains the membership function of the rule corresponding to each operating condition, calculates the center of gravity of the synthesized membership function obtained by taking the maximum value of the membership function of each rule, The center of gravity for outputting the temperature correction value TC 'of the heating cylinder 100, the back pressure correction value PC' of the injection cylinder 102, and the screw rotation speed correction value RC 'while considering the deviation from the optimum state transition waveform from the waveform database 21. Computing unit 23
Including and Each correction value is added to the corresponding command value,
The addition result is output as a temperature adjustment value TC, a back pressure adjustment value PC, and a rotation speed adjustment value RC.

The operator adjusts the molding conditions by empirically estimating the flow state of the resin during plasticization from the characteristics of the resin, the structure of the molded product, and other environmental conditions.
In the process of making such adjustments, the operator has effective conditioning rules for obtaining good molded products. For example, "If the resin temperature is a little higher and the screw rotation speed is a little higher (the flowability of the resin is good, but the time constant is large because the adjustment by temperature is large, do not change the temperature) and reduce the back pressure a little. Use an effective empirical rule depending on the situation, such as (preventing resin backflow).

As shown in the following example, the rule database 22 is described as a language value that a person usually uses, such as "if ~ is ~, and ~ is ~", and is stored for each molded product. I'll do it. In addition, a language value such as "high" is previously stored as a membership function (MSF).
Describe it as the degree of certainty.

(Example) Rule 1 "If the temperature detection value TD is a little high and the rotation speed detection value R is
If D is a little late, lower back pressure correction value PC '. Rule 2 "If rotational speed detection value RD is a little faster and back pressure detection value P
If D is also a little higher, lower the back pressure correction value PC '. "

The fuzzy arithmetic unit 20 stores the experience rules of such experienced persons as rules in the rule database 22 for each molded product, and automatically adjusts each operating condition by fuzzy inference. Taking rule 1 as an example, a calculation process of fuzzy inference (a center of gravity calculation process) will be described. FIG. 3 shows the configuration in the case of rule 1. The operator first describes a language expression such as "cylinder temperature is a little high", "screw rotation speed is a little slow", and "back pressure correction amount is a little small" by a function called membership function. In the membership function, the horizontal axis represents the value of the input amount, and the vertical axis represents the probability that the input will occur in the range of 0 to 1. In the example, the trigonometric function is described, but the function can be described as any other function.

When the detected temperature value is "220 ° C." and the screw rotation speed is "200 rpm" as inputs, the respective probabilities are "0.5" and "0.3", and the minimum value () MIN), and as a certainty "0.
3 "is obtained. The minimum (MIN) of this value and the membership function of the back pressure correction amount is taken, and the shaded part is taken. Rule 2
Also in, the membership function of the back pressure correction amount is obtained by the same procedure. Next, the maximum (MAX) of the membership functions in the shaded areas of Rule 1 and Rule 2 is taken to obtain a composite membership function, the center of gravity of this composite membership function is determined, and this is used as the back pressure correction value PC '. There is no limit to the number of rules, and other correction values (rotational speed correction value, temperature correction value) are obtained by the same procedure. As described above, the fuzzy operation unit 20 optimally and quickly corrects the command value of each operating condition for the deviation with respect to the ideal state transition waveform (PVT waveform) of each molded product by utilizing the know-how of an experienced person. To do.

Needless to say, when the type of molded product changes, the contents output from the rule database 22 and the contents output from the waveform database 21 are changed accordingly.

[0019]

As described above, according to the present invention, the fuzzy inference is utilized to incorporate the empirical rule of the operator into the control, so that the operator's sequential adjustment work of molding conditions can be significantly reduced. It will be possible.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an embodiment of the present invention.

FIG. 2 is a configuration diagram of a fuzzy operation unit shown in FIG.

FIG. 3 is a diagram showing an example of a membership function used in the fuzzy operation unit of the present invention.

FIG. 4 is a diagram showing a locus of state transition of molten resin in a molding cycle of an injection molding machine.

[Explanation of symbols]

 100 Heating Cylinder 101 Screw 102 Injection Cylinder 103 Rotational Drive System 104 Mold 105 Cavity

Claims (1)

[Claims] 1. A control device for an injection molding machine that performs injection molding based on command values and detected values of operating conditions such as back pressure, screw rotation speed, cylinder temperature, etc. A fuzzy operation unit that receives and calculates a correction value for each operating condition and adds it to the corresponding command value and outputs it as an adjustment value; and an operation amount for each operating condition by receiving the adjustment value and the detection value of each operating condition. And a control unit for calculating,
The fuzzy operation unit, a waveform database that stores the optimum state transition waveform of the molten resin for each molded product, a rule database that accumulates an empirical rule of the operator for each operating condition, the waveform database, from the rule database A center of gravity computing unit for receiving data, obtaining a membership function of a rule corresponding to each operating condition, and computing a center of gravity of a synthesized membership function obtained by taking the maximum value of the membership function of each rule. A control device for an injection molding machine, which is characterized in that