JP2008273634A - Constant amount supply system for powder and granular material - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a constant amount supply system for powder and granular material capable of carrying out accurately stable constant amount supply, according to a past operation result, regardless of the presence/absence of the generation of disturbance. <P>SOLUTION: The constant amount supply system for the powder and granular material comprises a model producing portion 14 producing a model of a feeder 1 according to a relationship between operation amount and a discharging flow rate about each load obtained by past operation; an operation amount estimating/computing portion 15 estimating the operation amount from the model produced by the model producing portion 14 and predetermined final target flow rate and target flow rate convergence time so as to make the discharging flow rate ouputted from the model materialize a reference track; and an operation amount correcting portion 17 correcting the estimated operation amount estimated by the operation amount estimating/computing portion 15 so as to make a deviation between the discharging flow rate detected from a flow rate detecting portion 11 and a flow rate value on the reference track at the time detecting the discharging flow rate become generally zero. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

The present invention relates to a powder and substance quantitative supply system capable of quickly and accurately cutting a powder and granule.

  2. Description of the Related Art Conventionally, for example, when discharging granular materials such as drugs and pigments at a constant flow rate in a granular material discharging apparatus using a vibration feeder, very accurate discharging is required in units of milligrams.

  Here, a general powder discharge device has a vibration feeder (generally excited by electromagnetic force) attached to a hopper that is filled with a material (powder) inside, and the hopper and vibration feeder are connected to each other. And installed on a load measuring device (for example, a load cell) such as an electronic balance. And when the powder feeder is controlled from the hopper by controlling the vibratory feeder, and the powder load is reduced from the load of the powder measured by the load measuring device to the preset discharge amount. By repeating the operation of stopping the vibration feeder, a predetermined amount of powder particles are discharged.

  As described in Patent Document 1, etc., as a control method of the supply amount of the granular material, that is, a control method of the vibration feeder, a set flow rate (also referred to as a target flow rate) is compared with a flow rate that actually flows. In general, feedback control is performed so that they always coincide with each other, and PID control including proportional control (P), integral control (I), and differential control (D) is generally used. In addition, as described in Patent Document 2, the relationship between the load and the flow velocity is stored in advance for each of a plurality of different vibration frequencies, and each load is maintained so as to maintain a constant flow velocity based on this relationship. There is also a method of sequentially selecting vibration frequencies corresponding to the set flow velocity.

  However, the quantitative supply system employing such a control method has the following technical problems.

JP 2003-266164 A Japanese Patent Publication No. 3-78329

  The feedback control disclosed in Patent Document 1 and the like, particularly PID control, requires skill to set the control parameters L (dead time), T (first-order lag time constant), and K (process gain). If the settings are not optimized, precise quantitative feeding cannot be performed. In addition, even when a disturbance occurs, the actual flow rate is fed back and reflected in the parameter setting.

  The control disclosed in Patent Document 2 has a problem in that when the vibration frequency is switched according to the load, the flow rate changes suddenly and the flow rate is disturbed, so that precise control cannot be performed. In order to minimize the disturbance of the flow rate, it is only necessary to memorize the relationship between the load and the flow velocity for as many vibration frequencies as possible at the time of the test run. It takes time and is not efficient.

The present invention has been made in view of such conventional problems, and the object of the present invention is to provide a highly accurate and stable quantitative supply based on past operation results regardless of whether or not a disturbance has occurred. It is in providing the fixed-quantity supply system of the granular material which can perform.

  In order to achieve the above-mentioned object, a quantitative supply system of granular material according to the present invention includes a hopper that is filled with granular material, a feeder that is attached to the hopper and that discharges the granular material, and a drive that drives the feeder. An operation amount is supplied to the drive unit so that the discharge flow rate of the granular material is substantially constant, a load detecting unit that is provided at a lower portion of the hopper, and that measures a load of the granular material in the hopper In a quantitative supply system comprising a control unit and a flow rate detection unit for detecting the discharge flow rate, the control unit is based on a relationship between an operation amount for each load and a discharge flow rate obtained by a past operation. A model creation unit that creates a feeder model, a model created by the model creation unit, and a target reference trajectory created based on a preset final target flow rate and a target flow rate convergence time. The operation amount prediction calculation unit that predicts the operation amount, the discharge flow rate detected from the flow rate detection unit, and the discharge flow rate are detected so that the discharge flow rate output from the control unit realizes the reference trajectory. A target reference trajectory control method is provided by including an operation amount correction unit that corrects an operation amount predicted by the operation amount prediction calculation unit so that a deviation from a flow rate value on a reference trajectory in time is substantially zero. Realized.

  According to the quantitative supply system having such a configuration, the user obtains the target flow rate by determining the relationship between the flow rate for each load and the manipulated variable based on the past operation results and modeling the feeder that is the control target. Then, the operation amount can be predicted only by setting the target flow rate convergence time. Furthermore, after the actual operation is started, based on the deviation between the actual flow rate and the reference trajectory, the predicted operation amount is corrected so that the actual flow rate is closer to the reference trajectory. Compared with the conventional PID control method, The followability to the reference trajectory is improved.

  Further, prior to the actual operation of discharging the granular material, the quantitative supply system sets a plurality of load ratios with a state where the granular material is fully loaded in the hopper as a load of 100%, and the load is sequentially increased from 100%. In the process of lowering, a test operation control unit that performs a test operation by supplying a predetermined operation amount to the drive unit at each load ratio is provided, and the model creation unit is configured based on the test operation result. May be created.

  According to such a configuration, the storage of operation data for model creation is a process of gradually reducing the granular material from when it is fully loaded, without having to replenish and test the granular material many times. It is efficient because it can be done at once.

  The control unit calculates the operation amount when replenishing the granular material in the hopper with reference to a history of the operation amount from when the granular material is loaded one turn before replenishment. You may make it provide a part.

In addition, the control unit further supplies an operation amount prediction calculation unit based on the target reference trajectory control method and an operation amount when replenishing the granular material in the hopper from when the granular material is loaded one turn before. You may make it provide the operation amount switch judgment part which switches the feedforward calculating part calculated with reference to the history of the operation amount until it changes at least according to the load measurement value of the granular material in the said hopper.

According to such a configuration, for example, at the time of body material replenishment in which the powder flow load that replenishes the granular material in the hopper greatly fluctuates, by performing a feedforward calculation that does not depend on the detected flow rate, By appropriately combining feedback and feedforward, the operation amount and the flow rate are not disturbed even when the raw material is replenished or the disturbance is generated, and the stable amount can be constantly supplied.

According to the quantitative supply system of the present invention, the user obtains the target flow rate by determining the relationship between the flow rate for each load and the manipulated variable based on the past operation results and modeling the feeder that is the control target. The operation amount can be predicted simply by setting the target flow rate convergence time. Furthermore, after the actual operation is started, based on the deviation between the actual flow rate and the reference trajectory, the predicted operation amount is corrected so that the actual flow rate is closer to the reference trajectory. Compared with the conventional PID control method, The followability to the reference trajectory is improved.

  DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the accompanying drawings. As shown in FIG. 1, the quantitative supply system 1 of the present embodiment includes a control target (weight feeder 1), a control unit 10 that controls the control target, and various storage units 20 that can be accessed from the control unit 10. Consists of

  As shown in FIG. 1, a weight-type feeder 1 that is a control target of the present embodiment is a hopper 2 that is mainly filled with raw materials (hereinafter referred to as powder particles in the embodiment), and a vibration feeder 3 that is attached to the hopper 2. An electromagnetic drive unit 4 (not shown) that drives the vibration feeder 3 and a load cell 5 that is provided below the hopper 2 and measures the load of the granular material in the hopper 2. In the weight type feeder 1, the operation amount MV output from the control unit 10 is input to the electromagnetic drive unit 4 to vibrate the vibration feeder 3, and the powder and granular materials in the hopper 2 are moved through the vibration feeder 3. It is discharged and supplied at a substantially constant flow rate. The output of the load cell 5 is converted into a flow rate (amount of discharged particulate matter per unit time) and input to the control unit 10 as a control amount (flow rate) PV. Then, based on the control amount PV, the operation amount MV is calculated in real time so that the flow rate of the granular material becomes substantially constant, and is supplied to the electromagnetic drive unit 4 again.

  In the following, an example in which the controlled object is the weight type feeder 1 will be described. However, the quantitative supply system of the present invention is not limited to the weight type feeder 1 of the present embodiment, but other feeders such as a capacity type feeder. It can also be applied to a feeder. The hopper may have any shape. The feeder is not limited to a vibratory feeder, and the driving method is not limited to an electromagnetic driving method, and may be a belt type, a motor screw rotating type, an auger type of a helical rod, a disk type for scraping powder with a rotating disk, or the like. As the load detection device, a load detection device other than the load cell 5 may be adopted. The load detection method may be any of electromagnetic balance type, capacity type and the like. Further, the conversion from the output value of the load cell 5 to the flow velocity may be performed by the control unit 10, the DA converter for converting the digital control amount PV into an analog signal, or the output value of the analog load cell 5. The AD converter for converting into a digital signal may be provided in the vibration feeder or may be provided in the control unit 10. In addition, the control method of this embodiment is a loss-in-weight method that controls the supply amount by measuring the weight of the entire hopper and vibration feeder with a load detection device and calculating the weight reduction rate due to material discharge. This control method may be used.

  The control unit 10 is a means for inputting the control amount PV from the weight type feeder 1, calculating the operation amount MV based on the input control amount PV, and supplying the output to the weight type feeder 1. 2, the flow rate detection unit 11, the load detection unit 12, the test operation control unit 13, the model creation unit 14, the operation amount prediction calculation unit 15, the actual operation control unit 16, the operation amount correction unit 17, and the operation amount switching. A determination unit 18 and an FF calculation unit 19 are provided.

  The flow rate detection unit 11 is a means for detecting the control amount PV, here, the flow rate PV of the granular material (the amount of discharged granular material per unit time) from the load cell 5.

  The load detection unit 12 is means for detecting the weight (hereinafter referred to as load) of the granular material remaining in the hopper 2 from the load cell 5.

  Prior to the actual operation of discharging the granular material, the trial operation control unit 13 sets the state in which the granular material is fully loaded in the hopper 2 as 100% load, for example, loads 90, 70, 50, 30% and a plurality of load ratios. In the process of setting and lowering the load sequentially from 100%, a predetermined manipulated variable MV (here, voltage input to the electromagnetic drive unit 4) is output in each load ratio state, and the weight feeder 1 is a means for performing a test run.

  As a result of the trial operation, the flow rate PV obtained from the flow rate detection unit 11, the load obtained from the load detection unit 12, and the operation amount MV supplied to the heavyweight feeder 1 by the trial operation control unit 13 are synchronized with each other in time. And stored in the operation data storage unit 23. That is, the operation data storage unit 23 stores the relationship between the flow rate PV and the manipulated variable MV for each of a plurality of types of loads. The operation data can be stored efficiently because it is not necessary to supply and test the powder particles many times, and can be performed once in the process of gradually decreasing the powder particles from the full load.

  The model creation unit 14 is a control target for each load based on the correlation between the flow rate PV for each load stored in the operation data storage unit 23 and the manipulated variable MV during past operation (trial operation or actual operation). This is a means for modeling the weight type feeder 1, that is, for obtaining a transfer function of the weight type feeder 1.

  Generally, there is a correlation between the operation amount MV input to the control target and the control amount PV (in this embodiment, the flow rate PV) indicating the state of the control target as a result of the input of the operation amount MV. There is a model of this correlation that is graphically or mathematically referred to as a model to be controlled. If the controlled object is modeled, the operation amount MV to be input to the controlled object can be predicted before the actual operation.

In this embodiment, the transfer function G _p (S) of the weight type feeder 1 is represented by G _p (S) = K * e ^−LS / (1 + T · S) by the step response method, and the process parameter of the transfer function The process gain K, waste time L, and first-order lag time constant T are determined based on the relationship graph between PV and MV shown in FIG. Further, K, L, and T determined for each load are linearly interpolated as shown in FIG. 5, expressed as a function of the load, and stored in the model / SP storage unit 21.

  Further, the model creation unit 14 determines that the final target flow rate is the target flow rate based on the process parameters stored in the model / SP storage unit 21 and the preset “final target flow rate” and “target flow rate convergence time”. A smooth trajectory of the target flow rate until convergence by the convergence time, that is, a target reference trajectory SP (S) (relationship between target flow rate and time) is created, and this target reference trajectory SP (S) is used as a model / SP storage unit. 21. In addition, about the preparation method of reference track | orbit SP (S), since a well-known method can be used, the detail is abbreviate | omitted.

Based on the reference trajectory SP (S) stored in the model / SP storage unit 21, the operation amount prediction calculation unit 15 generates a trajectory (operation amount MV and operation amount MV) for realizing the reference trajectory SP (S). This is a means for predicting and calculating the time relationship) in the operation amount storage unit 22. Note that the reference trajectory is obtained by substituting the final target flow rate and the target flow rate convergence time into the transfer function G _p (S) previously obtained, and the manipulated variable MV (S) and the flow rate PV (S). In the meantime, the relationship PV (S) = G _p (S) * MV (S) is established, so that the manipulated variable MV (S) has a flow rate PV (S) as an inverse function of the transfer function G _p (S). Is obtained by multiplying FIG. 6 is an example of the reference trajectory SP (S).

Since the process parameters K, L, and T of the transfer function G _p (S) are functions of the load, the load value at this time (actual operation start) The weight of the granular material filled in the hopper 2), and the y-axis of the reference trajectory SP (S), that is, the target flow rate integrated from S = 0 to t for each unit time. The result of subtraction from the load value when S = 0 is substituted into the load that is a variable of each process parameter to obtain a transfer function G _p (S = t) for each time, and the manipulated variable MV at time t is Desired. FIG. 6 is a graph showing the relationship between the operation amount MV and time for realizing the reference trajectory SP (S) set here.

As described above, the transfer function G _p (S) is not uniquely defined, but is a function that varies depending on the load, and the weight type feeder 1 to be controlled is accurately modeled. It is possible to accurately predict the operation amount MV for each.

  The actual operation control unit 16 is a means for reading the operation amount MV stored in the operation amount storage unit 22 and supplying the operation amount MV to the weight type feeder 1 to perform the actual operation of discharging the granular material. The actual operation control unit 16 discharges the granular material based on the operation amount MV predicted by the operation amount prediction calculation unit 15 in the initial operation. Further, the actual operation control unit 16 stores the relationship between the operation amount MV actually supplied to the weight type feeder 1 and time in the operation amount storage unit 22.

  The operation amount switching determination unit 18 calculates an operation amount MV that has been predicted and calculated based on the actual flow rate PV of the granular material detected by the flow rate detection unit 11 during the actual operation of discharging the granular material. This is means for determining whether to correct by the correction unit 17 or to switch to the one calculated by the feedforward calculation unit 19 (hereinafter referred to as FF calculation unit).

  Specifically, with the passage of time of the granular material discharge operation, the granular material in the hopper 2 starts to decrease, the load becomes a predetermined value or less, and the time change of the operation amount MV becomes a predetermined value. It is determined that raw material supply is necessary, and the raw material is supplied to the hopper 2. At the same time, the operation amount switching determination unit 18 instructs the FF operation unit 19 to calculate the operation amount MV. Then, during the raw material replenishment, the operation amount MV calculated by the FF calculation unit 19 is supplied from the actual operation control unit 16 to the weight type feeder 1. In addition, when the flow rate PV of the granular material is other than the predetermined value, the operation amount switching determination unit 18 causes the operation amount correction unit 17 to perform a correction calculation of the operation amount MV, and during that time, the operation amount correction unit 17 corrects the operation amount. The calculated operation amount MV is supplied from the actual operation control unit 16 to the weight type feeder 1.

  Specifically, the operation amount correction unit 17 compares the current time t with the flow rate PV detected from the flow rate detection unit 11 at time t and the value of the reference trajectory SP at the time t when the flow rate PV is detected (target). The operation amount MV at the next time t + 1 that has been predicted and calculated is corrected so that the deviation (PV-SP) is close to 0 and stored in the operation amount storage unit 22. This is means for updating the manipulated variable MV. When correction is performed, the actual load detected from the load detection unit 12 is reflected in the process parameters K, L, and T of the reference trajectory SP stored in the model / SP storage unit 21 to manipulate the amount of operation MV. Perform the correction. When the operation amount correction unit 17 performs the correction, the operation amount prediction calculation unit 15 corrects the operation amount MV using the inverse function as in the case of using the inverse function of the transfer function.

  Specifically, when it is determined that raw material replenishment is necessary, the FF calculation unit 19 sets the operation amount MV at the time of replenishing the raw material to the operation amount MV from when the powder is fully loaded one turn before the replenishment. With reference to the history, the operation amount MV in the operation amount storage unit 22 is updated. Since the load greatly changes when the raw material is replenished, the relationship data between the past load value and the operation amount with respect to the target flow rate is read from the operation amount storage unit 22 and extrapolated in time from the change characteristic of the load value detected by the load detector. The operation amount for the load value is obtained by the FF calculation unit 19.

  According to the quantitative supply system described above, the user obtains the relationship between the flow rate PV for each load and the manipulated variable MV based on the past operation results, and performs modeling of the weight-type feeder 1 that is the control target, thereby allowing the user to By simply setting the target flow rate and the target flow rate convergence time, the manipulated variable MV can be predicted. Further, after the actual operation is started, the predicted operation amount MV is corrected as needed so that the actual flow rate PV is closer to the reference track SP based on the deviation between the actual flow rate PV and the reference track SP. Compared with the method, the followability to the reference trajectory SP is improved.

  In addition, the quantitative supply system of the present invention exerts a suppression effect against disturbance. In the case of conventional PID control, the target value SP is a fixed value, and SP-PV is multiplied by a predetermined coefficient to calculate a PID constant. Therefore, when a disturbance occurs, SP-PV remains as it is as the next manipulated variable MV. This is reflected and causes overshoot and manipulated variable disturbance. However, in the quantitative supply system of the present invention, the manipulated variable MV is predicted based on a reference trajectory SP created in advance, and correction is performed on the basis of the predicted manipulated variable MV, so that disturbance (vibration) is generated. Even if it occurs, the manipulated variable MV is not significantly disturbed. Further, since the change in the manipulated variable MV does not become steep and vibrations and overshoots are suppressed, compared with the conventional PID control method, the life and durability are determined by the drive current (in this embodiment, the electromagnetic The life and durability of the drive unit 4) and the vibration feeder 3 are greatly improved.

In particular, when performing electromagnetic driving as a driving device, compared with the conventional PID control method,
Reducing the vibration of the manipulated variable MV by this target reference trajectory control method greatly improves the life and durability of the drive device (the electromagnetic drive unit 4 in this embodiment) and the vibration feeder 3.

  The overall operation of the quantitative supply system will be described below with reference to the system configuration diagrams of FIGS. 1 and 2 and the flowchart of FIG.

  First, the trial operation control unit 13 supplies the operation amount MV to the weight type feeder 1 that is the control target, and performs the trial operation of the weight type feeder 1 (S100). The test operation conditions of this example are as shown in FIG. The granular material is loaded in the hopper 2 in advance, the illustrated operation amount MV is supplied, and the granular material is discharged until the load reaches 90% of the full load.

  When the load reaches 90% of the full load, the trial operation control unit 13 temporarily stops outputting the operation amount MV, and then resumes outputting the operation amount MV until the load reaches 70%. During this time, the trial operation control unit 13 stores the output manipulated variable MV and the flow rate PV from the flow rate detection unit 11 in time synchronization with each other ((a) in FIG. 3).

  When the load reaches 70% of the full load, the trial operation control unit 13 temporarily stops outputting the operation amount MV, and then resumes outputting the operation amount MV until the load reaches 50%. During this time, the trial operation control unit 13 stores the output manipulated variable MV and the flow rate PV from the flow rate detection unit 11 in time synchronization with each other ((A) in FIG. 3). Similarly, the trial operation control unit 13 controls the operation amount MV, and the relationship between the operation amount MV and the flow rate PV ((c) in FIG. 3) until the load becomes 50% to 30%, and the load is 30%. Also, the relationship between the manipulated variable MV and the flow rate PV (from (D) in FIG. 3) until approximately 20% is stored in the operation data storage unit 23. Thus, the relationship between the manipulated variable MV and the flow rate PV is stored for each load (in this embodiment, four points of 90%, 70%, 50%, and 30%).

  In the present embodiment, the manipulated variable MV during the trial operation is desirably a voltage that is frequently output during actual operation. In this embodiment, the manipulated variable MV outputs a constant value regardless of the load, but does not necessarily have to be a constant value.

  In this way, the test run data can be stored efficiently without the need to replenish the powder many times and test, and can be performed once in the process of gradually reducing the powder from the full load. It is.

  Note that the load setting need not be the above four points, and the relationship between the manipulated variable MV and the flow rate PV may be stored for two or more loads. For this purpose, it is desirable to store the relationship for three or more loads. Also, the load setting may be made finer between 100% and 0%.

  The model creation unit 14 creates, for each load, a model of the weight-type feeder 1 that is a control target based on the relationship between the operation amount MV and the flow rate PV for each of a plurality of loads stored in the operation data storage unit 23 (S110). ).

In this embodiment, the transfer function G _p (S) of the weight feeder 1 is expressed by G _p (S) = K * e ^−LS / (1 + T · S) by the step response method. The weight type feeder 1 is modeled by obtaining the process gain K, the waste time L, and the first-order lag time constant T, which are the process parameters. When the load is 90%, the relationship between the manipulated variable MV and the flow rate PV in FIG. 3A is shown. When the load is 70%, FIG. 3A is shown. When the load is 50%, FIG. For%, (D) in FIG. 3 is applied to the graph in FIG. 3 to calculate K, L, and T, respectively.

  The model creation unit 14 further linearly interpolates K, L, and T calculated for each load, expresses them as a function of the load, and stores them in the model / SP storage unit 21.

  Based on the reference trajectory SP (S) stored in the model / SP storage unit 21, the operation amount prediction calculation unit 15 provides a trajectory (operation amount MV and time) of the operation amount MV for realizing the reference trajectory SP (S). ) Is predicted and stored in the operation amount storage unit 22 (S120).

The target reference trajectory is obtained by determining the parameters of the transfer function G _p (S) obtained previously from the final target flow rate and the target flow rate convergence time. The manipulated variable MV (S) and the flow rate PV (S) Since the relationship PV (S) = G _p (S) * MV (S) is established between and the operation amount MV (S), the flow rate PV (S) becomes the inverse function of the transfer function G _p (S). Obtained by multiplying S). FIG. 6 is an example of the target reference trajectory SP (S) until the weight type feeder 1 of the present embodiment reaches the preset final target flow rate within the target flow rate convergence time.

Further, since the process parameters K, L, and T of the transfer function G _p (S) are functions of loads as shown in FIG. 5, the operation amount prediction calculation unit 15 performs the operation when S = 0. After setting the load value at this time (the weight of the granular material filled in the hopper 2 at the start of actual operation) as the load of 100%, the y-axis of the reference trajectory SP (S), that is, every sampling time, that is, The result of subtracting the target flow rate integrated from S = 0 to t from the load value at S = 0 is substituted for the load that is a variable of each process parameter, and the transfer function G _{p at} an arbitrary time t (S = t) is obtained, and the manipulated variable MV (S = t) at time t is obtained. The obtained operation amount MV is stored in the operation amount storage unit 22 in association with the time t.

  When the time t comes, the actual operation control unit 16 reads the operation amount MV at the time t stored in the operation amount storage unit 22 and supplies the operation amount MV to the weight type feeder 1 to perform the actual operation of discharging the particulate matter. Perform (S130). The actual operation control unit 16 discharges the granular material based on the operation amount MV predicted by the operation amount prediction calculation unit 15 in the initial operation. Further, the actual operation control unit 16 stores the relationship between the operation amount MV actually supplied to the weight type feeder 1 and time in the operation amount storage unit 22.

  The operation amount switching determination unit 18 uses the operation amount MV that is predicted and calculated based on the actual flow rate PV of the granular material detected by the flow rate detection unit 11 during the actual operation of discharging the granular material. It is determined whether the correction is performed by 17 or the one calculated by the FF calculation unit 19 is switched (S140).

  Specifically, with the passage of time of the granular material discharge operation, the granular material in the hopper 2 starts to decrease, the load becomes a predetermined value or less, and the time change of the operation amount MV becomes a predetermined value. It is determined that raw material supply is necessary, and the raw material is supplied to the hopper 2. At the same time, the operation amount switching determination unit 18 instructs the FF calculation unit 19 to calculate the operation amount MV, and the calculated operation amount MV is stored and updated in the operation amount storage unit 22 (S160). While the raw material is being replenished, the operation amount MV calculated by the FF operation unit 19 is read from the operation amount storage unit 22 and supplied from the actual operation control unit 16 to the weight type feeder 1.

  In addition, when the discharge flow rate of the granular material is other than the predetermined value, the operation amount switching determination unit 18 causes the operation amount correction unit 17 to perform the correction calculation of the operation amount MV, and the calculated operation amount MV is the operation amount. The storage unit 22 is updated. Meanwhile, the operation amount MV corrected and calculated by the operation amount correction unit 17 is read from the operation amount storage unit 22 and supplied from the actual operation control unit 16 to the weight type feeder 1 (S150).

  Specifically, when the current time is t, based on the deviation between the flow rate PV detected from the flow rate detection unit 11 at time t and the value of the reference trajectory SP at time t, the deviation approaches 0. The operation amount MV predicted at the next time t + 1 is corrected and stored in the operation amount storage unit 22. Note that a method for correcting the manipulated variable MV so that SP-PV = 0 is well-known, and a description thereof will be omitted here.

  The inclination of the reference trajectory SP (or the target flow rate convergence time) is corrected at any time according to the deviation between the flow rate PV at time t and the flow rate PV at time t-1, that is, the degree of change in flow rate. Also good. When the reference trajectory SP is corrected, the operation amount prediction calculation unit 15 recalculates the operation amount MV after the time t, and updates the value stored in the operation amount storage unit 22. Accordingly, the operation amount correction unit 17 also corrects the operation amount MV at the next time t + 1 based on the corrected value at time t on the reference trajectory SP.

  In addition, when it is determined in the operation amount FF calculation (S160) at the time of material replenishment that the material replenishment is necessary, the operation amount MV at the time of material replenishment is the operation amount from the time when the powder is fully loaded to the time of replenishment. The operation amount MV in the operation amount storage unit 22 is updated with reference to the MV history. Since the load greatly changes when the raw material is replenished, the relationship data between the past load value and the operation amount with respect to the target flow rate is read from the operation amount storage unit 22 and extrapolated in time from the change characteristic of the load value detected by the load detector. The operation amount is corrected by obtaining the operation amount for the load value by the FF calculation unit 19.

  The above steps from S130 to S160 are repeated until the discharge of the granular material is completed.

  FIG. 7A shows a load, an operation amount MV, and a target flow rate when the quantitative supply system of the present embodiment performs quantitative discharge of the granular material so as to approach a predetermined target flow rate (actual). The relationship of the flow rate PV is shown. FIG. 7B shows the load, the manipulated variable MV, and the target flow rate when the quantitative supply device using conventional PID control discharges the granular material so as to approach the predetermined target flow rate. And the (actual) flow rate PV.

  In both cases, the controlled object is the same, but the control method is different. In addition, although the final target flow rate is set to be the same, in FIG. 7A, the control target is modeled by trial operation of the heavy weight feeder 1 that is the control target, and the operation amount MV is based on a preset reference trajectory. After the actual operation is started, the operation amount MV is sequentially corrected by the operation amount correction unit 17 or the FF calculation unit 19. In FIG. 7B, the process parameters for PID control are determined by using an approximate expression proposed by Zigler Nichols, a general PID control parameter determination method.

  Comparing the graphs of FIG. 7A and FIG. 7B, in FIG. 7B, the detected flow rate PV is directly reflected in the determination of the process parameters for PID control after the start of operation, and the target can be achieved in a short time. Since feedback control is performed so as to approach the flow rate, the parameters are likely to fluctuate in the initial stage of operation, and it can be seen that overshoot occurs in both the manipulated variable MV and the flow rate PV. In addition, when the discharge of the granular material is temporarily stopped or when the target flow rate is switched, the detected flow rate PV is directly reflected in the determination of the process parameter, so that overshoot occurs. Such an overshoot is a factor that reduces the accuracy of quantitative supply.

  On the other hand, in the case of the quantitative supply system of the present embodiment, as shown in FIG. 7A, the manipulated variable MV is predicted according to the model to be controlled and the reference trajectory. Is not directly reflected in the manipulated variable MV, and it can be seen that the rise of the manipulated variable MV in the initial operation is smooth. In addition, the particulate matter is not discharged based on the predicted operation amount MV, but based on the actual flow rate PV, the predicted operation amount MV is closer to the target flow rate (reference trajectory). Therefore, the operation amount is less disturbed and the followability to the reference trajectory is good.

  As described above, the fixed amount supply system of the present embodiment can not only perform the fixed amount supply with high accuracy, but also the operation amount is less disturbed, so that the life and durability of the electromagnetic drive unit 4 and the vibration feeder 3 are improved. Also contribute to.

  FIG. 8 shows a case where the quantitative supply system of the granular material is performed over a long period of time by the quantitative supply system of the present embodiment, and the replenishment operation of the granular material in the hopper 2 is repeated four times in the process. It is a graph which shows the relationship between load, operation amount MV, and flow volume PV.

  Based on the load detected from the load detection unit 12 and the time variation of the operation amount MV output from the actual operation control unit 16, the operation amount switching determination unit 18 provides the operation amount correction unit 17 with a correction unit for the operation amount MV. It is determined whether or not the FF calculation unit 19 calculates the operation amount MV.

  Normally, the powder is supplied as the operation starts, and the load decreases, so the frictional resistance between the particles decreases, and the operation amount gradually decreases with this. is there. However, when the load is reduced to a certain amount, the discharge amount of the granular material per unit time is also reduced, so that the operation amount does not decrease in proportion to the load and settles to a predetermined value. In the present embodiment, this time is determined to be the timing when the replenishment of the raw material is necessary, and the raw material is supplied to the hopper 2, and the operation amount switching determination unit 18 causes the FF calculation unit 19 to calculate the operation amount MV.

  The FF calculation unit 19 calculates the manipulated variable MV based on the history of the manipulated variable MV one turn before, instead of sequentially feeding back the flow rate PV detected from the flow rate detecting unit 11 to correct the manipulated variable MV. . That is, when the raw material is replenished, a feedforward calculation independent of the detected flow rate is performed.

  As described above, the quantitative supply system of the present embodiment performs a feedforward calculation of the manipulated variable MV when the raw material is replenished by appropriately combining feedback and feedforward, and is predicted except when the raw material is replenished. By performing a feedback calculation that sequentially corrects the manipulated variable MV, the manipulated variable and flow rate are not disturbed even when materials are replenished or disturbances occur, and stable and stable supply is always performed. I can do it. According to the graph of FIG. 9, it can be seen that there is no disturbance in the flow rate even when the raw material is replenished.

  In this embodiment, after about 5 seconds from the start of raw material replenishment, the operation amount switching determination unit 18 switches the operation amount calculation from the previous FF operation unit 19 to the operation in the operation amount correction unit 17. However, the switching timing from the FF calculation unit 19 to the operation amount correction unit 17 is not limited to this, and varies depending on the capacity of the hopper 2, the target flow rate, and the like. Further, at the time of switching, since the quantitative supply is resumed based on the predicted operation amount MV, the operation amount and the flow rate are not disturbed as in the PID control, and the switching is performed smoothly.

  FIG. 9A is a graph showing the relationship between the load, the manipulated variable MV, and the flow rate PV when a disturbance is applied to the vibration feeder 3 during the quantitative supply operation of the quantitative supply system of the present embodiment. FIG. 7B is a graph showing the relationship between the load, the operation amount MV, and the flow rate PV when a disturbance is applied to the vibration feeder 3 during the constant amount supply operation of the same constant amount supply device as in FIG. The applied disturbances are the same, and are due to the method of tapping the vibration feeder 3 by hand.

Comparing FIG. 9A and FIG. 9B, in the case of the quantitative supply by the conventional PID control method, the change in the flow rate PV is directly reflected in the determination of the process parameter. The amount will vary greatly. On the other hand, in the case of the quantitative supply system of the present embodiment, the basic operation is performed based on the predicted operation amount MV. Therefore, even if a disturbance occurs, a rapid change in the flow rate is not reflected in the operation amount MV as it is. It can be said that the operation amount is not disturbed and the system is strong against disturbance.

  As described above, the embodiment of the quantitative supply system has been described. However, the quantitative supply system of the present invention is not limited to having all of the configuration requirements described in the above embodiment, and various changes and modifications are possible. is there. Further, it goes without saying that such changes and modifications belong to the scope of the claims of the present invention.

  For example, in the above-described embodiment, the trial operation control unit 13 supplies the operation amount MV to the weight type feeder 1 to be controlled, performs the trial operation of the weight type feeder 1, and the operation amount MV obtained as a result Although the case where the model creation unit 14 creates a model to be controlled based on the relationship with the flow rate PV has been described, the modeling needs to be performed based on the operation amount MV and the flow rate PV obtained by the trial operation. Instead, modeling may be performed based on the past actual driving history. In that case, the trial run control unit 13 becomes unnecessary.

In addition, even after the model to be controlled is once created by the model creation unit 14, the flow rate PV stored in the operation data storage unit 23 every predetermined time during actual operation, and the relationship between the operation amount MV and the load at that time The model may be recreated and finely adjusted, and the accuracy of the model, and hence the accuracy of quantitative supply, will improve each time the operation is repeated.

It is a figure which shows schematic structure of a weight type feeder, and a connection structure with a control part and a memory | storage part. It is a figure which shows the detailed structure of a control part and a memory | storage part. It is an example of test run data. It is a figure which shows how to obtain | require the process parameter of a step response transfer function. It is a graph which shows an example of the relationship between a process parameter and a load. It is a time graph of a reference orbit and an operation amount. It is a figure which shows the result of having performed fixed supply by the fixed supply system of this invention, and the result of having performed fixed supply by the conventional PID control. It is a figure which shows the relationship between supply of a granular material, and replenishment. It is a figure which shows the result of having supplied fixed quantity by the fixed supply system of this invention at the time of disturbance generation, and the result of having supplied fixed quantity by the conventional PID control. It is a flowchart which shows the whole operation | movement of a fixed quantity supply system.

Explanation of symbols

1: Weight type feeder 2: Hopper 3: Vibration feeder 4: Electromagnetic drive unit 5: Load cell 10: Control unit 11: Flow rate detection unit 12: Load detection unit 13: Test run control unit 14: Model creation unit 15: Manipulation amount prediction calculation Unit 16: Actual operation control unit 17: Operation amount correction unit 18: Operation amount switching determination unit 19: FF calculation unit 20: Storage unit 21: Model / SP storage unit 22: Operation amount storage unit 23: Operation data storage unit

Claims (4)

A hopper that is filled with powder, a feeder that is attached to the hopper and discharges the powder, a drive unit that drives the feeder, and a load of the powder in the hopper that is provided below the hopper. A quantitative supply system comprising a load detection unit for measuring, a control unit for supplying an operation amount to the drive unit so that the discharge flow rate of the granular material is substantially constant, and a flow rate detection unit for detecting the discharge flow rate In
The controller is
A model creation unit that creates a model of the feeder based on a relationship between an operation amount for each load and a discharge flow rate obtained by past operation;
Based on the model created by the model creation unit and a target reference trajectory created based on a preset final target flow rate and target flow rate convergence time, the exhaust flow rate output from the model is the reference trajectory. An operation amount prediction calculation unit that predicts the operation amount so as to realize
The manipulated variable predicted by the manipulated variable prediction calculation unit so that the deviation between the discharge flow rate detected from the flow rate detection unit and the flow rate value on the reference trajectory at the time when the discharge flow rate is detected is substantially zero. The fixed quantity supply system characterized by the target reference orbit control system provided with the operation amount correction | amendment part which correct | amends.
The quantitative supply system includes:
Prior to the actual operation of discharging the particulate matter, in the process of setting a plurality of load ratios with the state that the particulate matter is fully loaded in the hopper as a load of 100%, and decreasing the load sequentially from the load of 100%, In each load ratio, a test operation control unit that performs a test operation by supplying a predetermined operation amount to the drive unit,
The quantitative supply system according to claim 1, wherein the model creation unit creates a model of the feeder based on the test operation result.
The controller is
A feedforward calculation unit that calculates an operation amount when supplying powder in the hopper with reference to a history of operation amount from when the powder is loaded one turn before supply is provided. The fixed-quantity supply system of Claim 1 or Claim 2.
The control unit further includes

The operation amount prediction calculation unit based on the target reference trajectory control method, and the operation amount when supplying the granular material in the hopper, the history of the operation amount from when the granular material is loaded one turn before replenishment. The operation amount switching judgment part which switches the feedforward calculating part calculated with reference according to the load measurement value of the granular material in the said hopper at least is provided. The Claim 1 or Claim 2 characterized by the above-mentioned. Fixed quantity supply system.

Images