JP2505221B2 - Flow control method in transfer system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention provides a constant flow rate in a conveyor system such as a belt conveyor that continuously conveys powder particles such as tobacco, steel, cement or fertilizer. The present invention relates to a flow rate control method for controlling a transfer speed.

[Conventional technology]

FIG. 10 shows an example of a conventional transportation system such as a belt conveyor that controls the flow rate of powdered or granular material to be transported so as to be constant.

This system is for powder and granular material A supplied on the conveyor 11.
The granular material A is conveyed with a constant opening of the gate 12 for adjusting the amount of the powder, and is added by the granular material A in the active length section by the conveyor scale 13 which makes the fixed section on the conveyor 11 the active length section. While detecting the weight, the transport speed of the powder A
That is, the belt speed of the conveyor 11 is detected by the speedometer 14.

Then, the average flow rate on the conveyor 11 is obtained by the calculation unit 15 based on the detected weighting and transport speed, and the control unit 16 adjusts the speed of the drive motor 17 based on the average flow rate, and the granular material is obtained. The flow rate of A is controlled to be constant. The dead time from the conveyor scale 13 to the outlet of the conveyor 11 may be compensated by incorporating a fixed delay element.

[Problems to be solved by the invention]

According to the conventional system as described above, the average flow rate can be made constant when the flow rate fluctuation in the vicinity of the gate 12 is not so large, but the conveyor scale 13 has the resolution of the working length section or less with respect to the weight distribution. Instead, feedback control is performed based on the weighted average on a constant working length section.
Further, the dead time from the working length section to the outlet of the conveyor changes depending on the belt speed, so that there is a problem that the flow rate control accuracy cannot be improved.

[Means for solving problems]

FIG. 1 is a diagram for explaining the principle of the present invention. The weight detection means 2 detects the weight applied by the transported object in a certain section on the transport path of the conveyor 1 that continuously transports the transported object, and The transport speed of 1 is detected by the speed detecting means 3,
In the conveyor system in which the control means 4 controls the conveyance speed of the conveyor 1 based on the detected conveyance speed and weighting, the storage means 5 for storing the weight distribution of the conveyed object on the working length section of the weighting detection means in advance. By the control means 4, the conveyor speed v _k of the conveyor 1 and the weight f _k in a given section are provided.
Are sampled at a constant cycle, and the transfer characteristics depending on the transport velocity v _k-1 sampled previously, the weight distribution w _{i, k-1} stored in the storage means 5 at the time of the previous sampling _, and the position of the weight detection means 2 Based on the weighting function h _i indicating
Based on the weight element corresponding to one load time obtained by subtracting the load element from the sampled current load amount and the load element due to the transported body in the active section at the time of the previous sampling, The weight of the transported object carried into the working length section in one sampling time is calculated, the weight distribution stored in the storage means 5 is updated based on the calculated weight, and the conveyor is also calculated based on this weight distribution. The transport speed of the conveyor 1 is controlled so that the flow rate per unit time of the transported object carried out from the carry-out port of No. 1 becomes the set constant amount Q ₀ .

[Action]

Among the weights by the transported object in the active section at the current sampling, the weighting elements by the transported object in the active section at the previous sampling are the weight distribution stored in the storage means 5 and the previously sampled transport. It can be obtained by a weighting function showing the transfer characteristic depending on the speed and the position of the weight detection means 2.

Therefore, the weight of the transported object carried in the working length section during one sampling time from the previous sampling time to the current sampling time can be obtained from the difference between the current weight and the weight element and the weight function, The calculated weight is added and the weight distribution in the storage means 5 is updated.

When the weight distribution of the storage means 5 is updated for each sampling, the weight distribution on the conveyor can be obtained with the resolution according to the sampling cycle, and thus the actual flow rate on the conveyor 1 can be detected and the flow rate can be controlled. .

〔Example〕

FIG. 2 is a block diagram of the belt conveyor system in the embodiment of the present invention.

Reference numeral 1 is a conveyor for continuously transporting the powder or granules, 2 is a conveyor scale for detecting the weight of the powder or granules in a constant working length section on the conveyor 1, and the weight is detected to detect an electric signal (weight measurement signal). ) Is equipped with a load cell 2a. Also,
2b is an amplifier that amplifies the output signal of the load cell 2a, 2c is a filter that estimates the stationary position from the transient observation data of the load cell 2a and corrects the vibration component, the load cell 2a,
As shown in FIG. 9, since the weight is detected as the strain of the metal elastic body 21 and is converted into an electric signal, the mechanism of the load cell 2a itself is not oscillating, and the movement by the filter 2c is substantially caused. It is not necessary to perform the target estimation.

3 is a pulse generator for detecting the conveying speed of the conveyer 1, 4 is a weight measurement signal output from the filter 2c and a pulse signal output from the pulse generator 3, and the weight f _k and the conveying speed v _k are set at a constant cycle. This is a control calculation unit that performs sampling at T and digitally controls the conveying speed of the conveyor 1 so that a set constant flow rate is maintained, as will be described later. The control calculation unit 4 includes a memory 5 that stores the weight distribution of the powder or granular material on the conveyor 1.

6 is in accordance with the speed setting signal output by the control calculation unit 4,
It is a speed controller that controls a variable speed motor 7 that drives the conveyor 1. As the variable speed motor 7, a high-precision servomotor having a fast response speed is used, and this drive system is a proportional element.

FIG. 4 is a block diagram of the control calculation unit 4 and the memory 5.

Reference numeral 41 is a clock generation unit, 42 is a CPU, 43 is a ROM storing a control program and the like, 44 is an operation unit, and the system bus B is connected to the memory 5 as the storage unit.

The weight measurement signal obtained by the conveyor scale 2 is input to the analog input section (AI) 45, and the belt speed data by the pulse signal output from the pulse generator 3 is input from the digital input section (DI) 46.

Then, the speed setting signal obtained as described later is output from the analog output unit (AO) 47 and input to the speed controller 6. The display unit 49 displays the belt speed and the like based on the data output through the digital output unit (DO) 48.

The conveyor system in this embodiment is a conveyor 1
When starting from a state where no particles are on top, set the initial weight distribution by sequentially calculating the weight distribution of the particles to be loaded, or when restarting after pausing. When the powder particles are on the conveyor 1 but the weight distribution thereof is unknown, the average value based on the sampled weight is set as the initial weight distribution.

In addition, after the initial weight distribution is set, the weight distribution in the memory 5 is sequentially updated based on the difference in weighting for each sampling cycle.

Then, based on the weight distribution w _{i, k} of the granular material on the conveyor 1 which is sequentially recorded / updated in the memory 5, a certain range δ _x from the outlet of the conveyor is conceptually shown in FIG. The belt speed such that the flow rate per unit time becomes a preset constant value Q ₀ when the powder particles inside are discharged, Is calculated, this belt speed v ₀ is output to the speed controller 6 as a speed setting signal, and the variable speed motor 7 is controlled.

Note that the weight distribution w _{i, k} is a refractory average weight for each section whose width changes in accordance with the sampling period T and the belt speed v _k (hereinafter referred to as the weight of the sampling zone, represented by w _k ). The feature of the present invention is that the resolution of the weight distribution w _{i, k} can be arbitrarily reduced by appropriately setting the sampling period.

FIG. 7 is a diagram for explaining in principle the transfer characteristics depending on the position in the working length section [0, L _D ] of the conveyor scale 2, which is installed at the bottom of the conveyor 1 as shown in FIG. pressurized weight due to the load point W applied to the measuring carrier 2 ₁ F, when the speed of the conveyor 1 and V (constant) time variation of the weighting quantity F is as shown in FIG. (b).

Therefore, the Laplace transform F (s) of the weight F is u (t) is a unit step function.

The rate of change of point load W (t) = (dW / dt) is Therefore, by expressing the point load by a delta function, £ {W (t)} = £ {W · δ (t)} = W, and the transfer function in the active length section is Becomes

Therefore, if the belt speed V changes, the working length passing time
Since T _D changes, the transfer function G _W (s) becomes time-varying.

However, since the working length L _D (= V · T _D ) is constant, the position-dependent transfer characteristic is time-invariant. Further, since the equation (1) is derived with the belt speed V being constant, the position-dependent transfer characteristic is obtained. Is represented by the formula in which T _D in formula (1) is replaced with L _D.

Therefore, the weight W and the weight F of the load on the working length section are related by the weighting function h _(x) having the same shape as that shown in FIG. 7C, as shown in FIG. The weight distribution w _{(x, t)} and the weighted weight F (t) on the working length section are It becomes a relationship.

Hereinafter, the setting of the initial distribution as the initial state and the operation in the steady state will be described separately.

(Setting of initial distribution) Even if the conveyor 1 moves, the weight distribution of the powder and granules with respect to the conveyor 1 does not change, but the weight distribution w _{(x, t) with} respect to the working length section of the conveyor scale 2 is changed as the conveyor 1 moves. And changes in the moving direction for each sampling period.

Therefore, the weighting component W _k given to the weighting F (k · T) by the weighting w _k of each sampling section discretized for each sampling period T changes for each sampling by the weighting function h _(x) .

The weight distribution w _{(x, t)} is the weight w _k of the discrete sampling interval
The weighting function h _(x) on the working length section is calculated for each sampling section based on the belt speed v _k sequentially obtained for each sampling.

That is, as shown in FIG. 7 (c), the weigh length interval [0, L _D] in the interval of [x _1, x _2], Is calculated.

Then, when the weight at the _nth sampling is represented by f _n and the belt speed is represented by v _n , one sampling time T up to that point
The weight of the granular material loaded during the period is the weight w _n of the sampling section, and the weight f _k until the leading granular material reaches the working length end (L _D ) and the weight of the sampling section w _n Relationship is Becomes

The arithmetic control unit 4 calculates the weights w ₁ , w ₂ , w ₃ , ... Of each sampling section by the inverse calculation of the equation (2) based on the weighted weight f _n , the belt speed v _n, and the weighting function h _(x) that are sequentially sampled. Obtained and sequentially stored in the memory 5.

In this embodiment, the active length section [0,
L _D ] is subdivided into preset minute sections Δx, and the position and weighting function on the working length section are discretized by the minute sections Δx.

FIG. 8 is a diagram conceptually showing the storage state of the weight distribution obtained in the above manner in the memory 5.
Is set with a large number of storage areas corresponding to the positions discretized by the minute section Δx, and the weights w ₁ , w ₂ , ... Of each sampling section are recorded in a plurality of records corresponding to the discretized positions in the section. The same weight data is recorded in the area.

(Operation in Steady State) FIG. 6 is a diagram for explaining changes in the weight distribution when the sampling and the transfer control are sequentially performed from the state where the weight distribution is stored in the memory 5 in advance. As described above, the weight w _k of the sampling section is averaged with a discrete value for each section.

Incidentally, the position x in the active length section is set to an integer value i corresponding to the discretization of the position of the active length section by Δx as described above, and the weight distribution w _{i, k} is set to the weight distribution w at the sampling time point k.
_{It is shown as k, i} (k corresponds to t). Similarly, the value of the weighting function h _{(x) at} the position i is h _i , and the velocity v _k is
The value is converted into a length with x _k = v _k · T, and is represented by a value with Δx as a unit.

The weight f _k at the sampling time point k is the sum of the weighted component W ₁ due to the granular material discharged from the active length section until the next sampling time point k + 1 and the weighted component W ₂ due to the granular material remaining in the active length section. Become.

That is, f _k = W ₁ + W ₂ + ε (where ε is the bias error).

Similarly, the weighting f _{k + 1} at the sampling time point _{k + 1} is
It is the sum of the weighted component W ₃ due to the granular material that was in the working length section at the sampling time point k and the weighted component W ₄ due to the granular material that was carried into the working length section within one sampling time from the sampling time point k.

That is, f _{k + 1} = W ₃ + W ₄ + ε.

On the other hand, the weighted component W ₃ expressed by the equation (5) is due to the residual content of the granular material in the working length section at the sampling time k, and the weight distribution w _{i, in} the range of the equation (5) _{k + 1} is the same as the weight distribution w _{i, k in the} equation (4) and is stored in the memory 5.

Therefore, like the weighted components W ₁ and W ₂ , the weighted component W _{3 of the} equation (5) can be obtained from a known amount, and the powder particles carried into the working length section within 1 sampling time from the sampling time k The weighted component W ₄ by the field is obtained by the calculation of W ₄ = f _{k + 1} -f _k + W ₁ + W ₂ -W ₃ (7).

By taking the difference f _{k + 1} −f _k between the weights f _{k + 1} and f _k as described above, the bias error ε can be canceled.

Let w _{M be} the average value of the weight distribution of the weighted component W ₄ , Therefore, Then, the weight component W ₄ and the weighting function h _i are used to obtain the weight of the granular material carried in within one sampling time up to the sampling point k + 1, that is, the weight w _{k + 1} of the sampling section, and stored in the memory 5. It

In this way, the weight distribution in the memory 5 is updated for each sampling with the average weight of the powdered particles carried in every sampling cycle _, and based on the weight distribution w _{i, k} in the memory 5, the weight distribution shown in FIG. As described in the equation (1), the belt speed is controlled so that the flow rate of the powder or granular material discharged from the outlet of the conveyor becomes a preset constant value Q ₀ .

In the steady state above, the case where the initial weight distribution is known has been described, but when the initial weight distribution is unknown, the weight of the granular material on the active length section obtained by the sampled weight f _k is calculated. The average value <W> is set as the initial weight distribution, and the steady state operation similar to the above is performed.

Next, the stability when the above average value <W> is used as the initial weight distribution will be described.

Error of the weight distribution at each position weigh length in section, e (x) = w ( x) - <W> ...... (9) a is the interval of the error e (x) [0, L _D] The average value within is “0” and is generally considered to be a normal random variable.

Substituting equation (9) into equations (3) to (5), As well Therefore, if <W> does not change, Therefore, the right side of the above equation (10) is not always "0" as an instantaneous value, but it is "0" as an expected value.

When the weight distribution at the next time point is obtained with the uniform distribution <w> as the initial weight distribution, if the weighting function h _i is left-right symmetric, the error term of w ₄ is Becomes

In this way, the error is transmitted almost periodically but does not diverge and is stable.

 FIG. 3 is a flow chart showing the control of the embodiment.

The weight function h _i is set, the weight measurement signal indicating the belt speed and the weight is sampled, and the recording position in the memory 5 for each sampling period is calculated.

After the weighted components W ₁ and W ₂ are calculated, it is determined whether or not there is a granular material on the active length section as an initial state by the sampled weighting, and if the weighting is “0” (2 The weight function h [x ₁ , x ₂ ] which is a coefficient of the equation) and the inverse calculation of the weight distribution are performed to sequentially set the initial weight distribution.

On the other hand, if the sampled weight is not “0”, the weight distribution in the steady state is obtained based on the setting of the initial weight distribution as the average value or the difference of the weight.

Then, the belt speed to be set is obtained from the weight distribution, the belt speed is set between the preset maximum speed v _max and the stopped state, the inverse characteristic compensation of the variable speed motor 7 is performed, and the speed setting signal is transmitted. Is output.

Since the variable-speed motor 7 uses a high-precision servomotor having a proportional response and a high response speed, the time constant of the first-order lag during inverse characteristic compensation is set to "0".

In the above embodiment, the case where the conveyor scale 2 has one weighing carrier 2 ₁ has been described. However, even when a plurality of weighing carriers are used, the same processing is performed only by changing the weighting function h (x). It can be performed. When a plurality of weighing carriers are used in this way, the working length section becomes long, and the control span is widened and the control accuracy is lowered in the conventional control method. However, according to the present invention, regardless of the length of the working length section. Control accuracy can be maintained.

Further, by keeping the weight distribution of the section from the end of the working length section to the conveyor outlet on the memory 5, the dead time in the section can be substantially eliminated.

〔The invention's effect〕

As described above, according to the present invention, in the conveyer system that continuously conveys the object to be conveyed, the conveyance speed and the weighted weight in the constant section are sampled at a constant cycle, and the current period is in the working length section. The load element obtained by subtracting the load element from the sampled current load amount by obtaining the load element due to the transferred body in the working length section at the time of the previous sampling, out of the load amount by the transferred body, and the one sampling time. The weight distribution stored in the storage means is recorded and updated based on the weighting function corresponding to the weighting function, which is obtained based on the corresponding weighting function during one sampling time. It is possible to obtain the weight distribution on the conveyor with high resolution, and by this, the actual flow rate can be detected and highly accurate flow rate control can be performed.

Further, by appropriately setting the sampling cycle, it becomes possible to control the flow rate variation in the working length section in an arbitrarily shortened section, and it is possible to improve the control accuracy.

[Brief description of drawings]

FIG. 1 is a diagram for explaining the principle of the present invention, FIG. 2 is a block diagram of a belt conveyor system in an embodiment of the present invention, FIG. 3 is a flow chart showing control of the embodiment, and FIG. 4 is an arithmetic control unit in the embodiment. And a memory block diagram, FIG. 5 is a diagram for explaining the relationship between the weight distribution, the flow rate at the conveyor outlet, and the belt speed, FIG. 6 is a diagram for explaining the change in the weight distribution in a steady state, and FIG. 7 is a conveyor scale. FIG. 8 is a diagram for explaining the transmission characteristics in the working length section in principle, FIG. 8 is a diagram conceptually showing the storage state of the weight distribution in the memory of the embodiment, and FIG. 9 is a load cell in the conveyor scale of the embodiment. FIG. 10 and FIG. 10 are views showing an example of a conventional transfer system. 1 ... Conveyor, 2 ... Weight detection means, 3 ... Speed detection means, 4 ... Control means, 5 ... Storage means.

Claims (1)

(57) [Claims] 1. A load amount detecting means (2) detects a load amount of a conveyed body in a certain section on a conveying path of a conveyer (1) which continuously conveys the conveyed body, and a conveying speed of the conveyer. In a transport system in which the control means (4) controls the transport speed of the conveyor based on the transport speed and the load amount detected by the speed detection means (3), the transported object on the working length section of the load amount detection means is previously received. A storage means (5) for storing the weight distribution of the body is provided, the conveyor speed and the load amount in the working length section are sampled at a constant cycle by the control means, and the previously sampled conveyor speed and the storage means are stored. Based on the stored weight distribution at the time of the previous sampling and the weighting function indicating the transfer characteristic depending on the position of the load amount detecting means, the load amount by the transported object in the active length section at the current sampling Among them, determine the load element by the conveyance object that was in weigh length interval in the previous sampling, the load element and one from the current load amount sampled is obtained by subtracting the load element
Based on the weighting function corresponding to the sampling time,
The weight of the transported object carried into the working length section in one sampling time is calculated, the weight distribution stored in the storage means is updated based on the calculated weight, and the weight distribution of the conveyor is calculated based on this weight distribution. A flow rate control method in a transport system, characterized in that the transport speed of a conveyor is controlled so that the flow rate per unit time of a transported object carried out from a carry-out port becomes a set constant amount.