DD258673B5 - Method for the digital control of the liquid feed - Google Patents

Description

W ₂ (t) = W _z -Aw / a (1-e " ^Qt ) + W ₀

is determined, where are

W _z = A _w t = the amount of liquid flowing at a constant speed; W ₀ = initial moisture;

α = time constant of the mixing plant;

t = the time elapsed from the end of the dry mixing phase.

For this 1 page drawing

The invention relates to a method for digital control of the liquid supply in the treatment of medium-viscosity mixtures by a mixing process with the aim of setting a desired consistency or viscosity, especially fresh concrete in stationary concrete mixing plants using the measurement of liquid supply and the effective power of the drive motor Mixing plant for the mixing process.

In order to achieve reproducible quality properties of the concrete in the usual batch-wise production of fresh concrete, it is necessary to comply exactly with the quantity ratios and properties of cement, water, aggregates and other additives required by the concrete recipe. The exact analysis of all components would therefore be required before each mixture, but in practice for time and economic reasons, not feasible.

The water-cement ratio and the consistency of the fresh concrete have a special influence on the concrete quality. Both are decisively influenced by the intrinsic moisture content of the aggregates, which, however, are greatly variable in practice due to storage options, reservoir conditions, weather conditions, etc. So that the quality of the hardened concrete is always secured, the risk of fresh concrete production resulting from unknown intrinsic moisture of the aggregates is hitherto reduced by increased cement addition. An obvious approach to determine the intrinsic moisture content of the aggregates is the use of moisture probes, z. B. microwave or neutron probes. With these probes, the moisture content of the fine aggregates is measured directly before each mix, and these values are used to calculate a corrected amount of water. Disadvantages of these probes are the high cost and the susceptibility due to the

complicated construction. The inhomogeneity of the sample, which may contain additional foreign matter, is the reason that the sole application of Moisture probes has not led to practically effective systems. Therefore, methods have been developed in which the measurement of the intrinsic humidity and the measurement of the consistency are combined. Thus, it is known to calculate the necessary amount of water with the measured initial moisture, to supply a major part of the amount of water and at a time during the mixing phase at which a given consistency value occurs, to add the remainder as a post-dosing (DD-WP 141129).

The consistency measurement is usually carried out indirectly via the measurement of the electrical resistance of the mixed material (DE-OS 1 784920) or on the active power determination of the drive motor of the mixing drum. The effective power of the motor is the frictional resistance of the mixed material, this value corresponds to the consistency, proportional (DE-OS 1 683778). The measurement of the active power as a function of time during the mixing phase results in a characteristic curve which has three essential sections in accordance with the current consistency of the mixed material

1. constant course during the dry mixing phase

2. Rise, maximum and drop during the time of water supply and submergence

3. constant final value when all the water supplied is mixed.

In DE-OS 2 855 324 a method is described which utilizes this curve. For each recipe, consistency setpoint values are determined at ideal intervals for equidistant intervals, converted into active power values and stored in a table as setpoint values. By comparison with corresponding measured values, the required amount of residual water is determined.

Disadvantages of this method are that a large number of measured values must be determined and stored for each recipe. In addition, first, the quantitative curve and the temporal position of the maximum curve of the initial moisture, grading curve and grain geometry of the aggregates depend and secondly by harmonious and stochastic disturbances of the mixing plant, the measured values are distorted. This complicates the calculation of the required amount of residual water. An exact determination of the amount of residual water is absolutely necessary because unusable concrete is produced by an excessively high addition of water.

In DE-AS 2432609 it is proposed to reduce a portion of the additional water quantity which is reduced by the greatest measurement error, which is referred to as the main part, to the mixture with the involvement of an arithmetic unit and a quantity meter in a large amount per unit of time and then, with the intervention of a second arithmetic unit, the remaining additional water while monitoring the power consumption of the mixer motor, add it to the mixture. Such a procedure leads to uneconomical mixing times, with the risk of segregation occurring and does not exclude possible incorrect dosing of the addition of water at too high intrinsic moisture content of the aggregates.

To determine the initial moisture content of the mixture, which results from the intrinsic moisture content of the additives, it is known from the description of the prior art in DE-OS 2339864 to observe the torque requirement of the mixer during the dry mixing phase and from there to achieve the final consistency to determine the amount of water to be added. Disadvantages of this solution are the long mixing times that occur and the risk of water overdosing given high intrinsic moisture content of the additives.

Other methods attempt to provide a defined initial moisture, e.g. By saturating the additives with water and then controlled partial removal of the water (DE-OS 2756039) or by supplying the water in vapor form (DE-OS 1 950 910) to achieve. These methods require additional equipment and time required per mixture.

From Nikolai N. Danilov, technology of industrial prefabrication of concrete and reinforced concrete elements; Verlag für Bauwesen Berlin, 1976, pages 40 to 43 discloses a method for metering the material components, in particular liquid components in fresh concrete production. The Anmachwassermenge is provided there by a metering device which can be preset from a control panel according to a predetermined dose. This metering device described there operates on the volumetric principle for liquids, whereby a high degree of accuracy of the dosage is achieved while simplifying the mechanical construction. As essential there is found that the amount of water to be added to the mixture must be determined taking into account the aggregate moisture, which often changes under production conditions. According to Danilov, the intrinsic moisture content of the aggregates is possible either by direct moisture determination in the aggregate working silo or by indirect moisture determination using the fresh concrete moisture in the mixing machine. In indirect moisture determination, the specified processing value of the fresh concrete is controlled and kept at the required level by controlling the water supply. The measurement is based on the principle that with uniform composition of the solid components of the mixture, the power that the motor of the concrete mixer needs when mixing, directly on the consistency of the mixture depends. For given materials, the power consumption of the motor depends on the total amount of water in the mixture. By means of predetermined curves of the course of the consistency characteristic values, a necessary interruption of the water supply to the concrete mixer should be able to be undertaken precisely at the moment at which the mixer motor takes up the power permissible for the given characteristic values. In the local solution, however, there is the problem that the amount of water supplied requires a certain time for mixing, which may take place while running consistency measurement switching off the water supply too late. Since replenishment with solid components is not possible in industrial mixing plants, the mixture in question is spoiled or does not have the required final consistency. If, on the other hand, one extends the mixing times while at the same time reducing the amount of water supplied, on the one hand a mixing-out effect occurs and, on the other hand, there is an excessively long residence time of the mixture in the mixing plant, so that the productivity of systems equipped in this way is restricted.

The invention is therefore based on the object of specifying a method for digital control of the liquid supply in the treatment of medium viscosity mixtures by a mixing process with the aim of setting a desired consistency or viscosity, especially fresh concrete in stationary concrete mixing plants, using the measurement the liquid supply and the active power of the drive motor of the mixing system an expected final consistency value of the mixture can be determined adaptively in time so that the required to achieve the desired consistency or viscosity amount of liquid can be predetermined without too long treatment or mixing times occur and whereby an overdose of solid additives can be avoided.

The object of the invention is achieved by the features of claim 1, wherein advantageous developments and refinements of the invention are shown in the subclaims.

It was found that the consistency of the mix is determined only by the currently mixed amount of water. Since the principle directly not measurable state variable-mixed amount of water-can be described by the measurable state variable amount of water - and here according to the invention an indicator function is utilized, the implementation of the consistency control method described below is possible.

Furthermore, a relationship was found between the course of the active power as a function of the mixed-in amount of water and used to implement the control method.

Before starting the actual mixing, starting from the representative performance values of the mixer motor obtained in a known manner in the implementation of a so-called teaching mixture, which represent a measure of the consistency of the mixed material, for a corresponding recipe or a certain degree of filling at the time of termination of the dry mixing phase, which is referred to below as t ₂ and at the time of reaching the target consistency, which is referred to below as t ₆ , determined power values and stored for further processing. These values only have to be determined once and can be fixed as system parameters. The power value determined at time t ₂ is referred to as the final dry mixing power value in the following description of the method and the power value determined at time t ₆ as the power end value. Accordingly, at the beginning of the mixing process, a final power value and dry mix end value are located in the storage tank of the mixing plant. From the beginning of the dry mixing process until the end of the mixing, power values are detected, for example likewise by determining the active power of the drive motor of the mixing plant, at equidistant time intervals. The actual dry mixing takes place in a known manner. During the dry mixing phase, according to the invention, the power (measured) values obtained at equidistant time intervals are processed using a discrete 1st order Kalman filter so that a filtered dry mix power value, ie a value close to the actual power value at the end of the dry mix phase, can be derived early.

From the determination of the filtered dry mixing power value and the comparison with the stored dry mixed service end value, incorrect metering, ie feed errors and gross deviations from the permissible target intrinsic moisture content of the additives, can be recognized immediately. At the beginning of the water supply at the time t ₂ , the filtered dry mixing power value is used as the initial value of a first component of a three-component state vector, the other components being freely selectable within the system considered. According to the above-mentioned knowledge by means of the detected power (-meß) values, a transfer of the measurable state variable - supplied amount of water - the non-measurable state variable - under mixed amount of water - which ultimately determines the consistency of the mix, within the period from the beginning of the water supply until the end of the water supply. These values now available represent the relationship between the power of the engine of the mixer and the actual mixed amount of water. Due to the further cognition of the invention mentioned above, ie the representable relationship between the course of the power and the mixed amount of water, the control system is completely writable. By means of a third-order discrete Cayman filter, control components are provided as estimated values, ie it is determined which values are to be expected at the time of the next power measurements. These control components correlate with the expected final performance value. From the comparison of the expected final power value with the stored final power value, the control variable for ending or continuing the water supply can be derived with high accuracy. If an identical recipe is processed in the next mixing process, it is expedient to record the ascertained control components, ie the estimated final value, at the time of the total submixing of the water, to transfer this value to the initial value and to use it as the starting component, ie as an initial estimate value in the control process. Such a procedure ensures a low settling time of the overall control process and positively influences the mixing process in terms of time and quality.

The invention will be explained in more detail using an exemplary embodiment with the aid of a figure. The figure 1 a to e shows the time course of the active power of the motor of a mixing plant with superimposed interference signals during the mixing of a conventional recipe. The mean effective power of the drive motor is advantageously determined by measuring current and voltage at equidistant intervals and calculating the product sums. The abscissa of Fig. 1, the time stamps t, to t ₆ are given, which correspond to the essential steps of the method according to the invention. Before starting the mixture, it is necessary to prepare a calibration mixture for each concrete recipe with expediently known quantity ratios and properties of all components used and to fix the power value N ₂ at the time mark t ₂ and the power value N ₆ as system and recipe parameters at the time mark te ,

In a known manner, the mixing drum in the time phase from t ₀ to t, with the dry components of the concrete and a small amount of water that determines the initial moisture W ₀ determined with the unknown intrinsic moisture content of the aggregates. In the dry mixing phase ti to t ₂ , the dry components are mixed. The consistency correlating active power N is constant for this purpose. The numerical size of these constants varies depending on the initial moisture W ₀ and on the weighing inaccuracies and the system configuration. This constant, which is directly difficult to measure as a result of the superimposed perturbations, is provided by means of the inventive use of a discrete Kaiman filter of the first order as an immunity-free value. With the help of the simple model N _k ^, = N _k and the recorded measured values superimposed with disturbances of the variance R, the estimated value N _{k +} , the active power and the error covariance matrix P _k ^ are determined. The values N _k are shown in FIG. 1 b. Already from the time ti., On, the estimated value N _k _ τ of the filter with the determined plant and recipe parameters N _{2 is} compared. In case of gross deviation, either a feed error may have occurred or an unusually high intrinsic moisture content of the starting materials may be present. In both cases, corresponding reactions take place. The power value N ₀ determined at time t _{2 is} correlated with the initial humidity W ₀ and is used as the initial value for the determination of the further power values N during the following phases. With the start of the feed and sub-mixing phase from t ₂ to t ₄ , the amount of water required according to the recipe flows into the mixing drum at a constant flow rate A _w (FIG. 1 c). According to the finding that only the current mixed amount of water W _{2 determines} the consistency, the mixed amount of water with the relationship

W ₂ (t) = A _w t ^-A _{w / a} (1-e " ^aI ) + W ₀ ,

where A _w t = W _z = amount of liquid supplied, indirectly determined and where α represents the time constant of the mixing plant in the considered mixing process. According to the relationship found, the dependence of the active power N on the mixed-in amount of water W _{2 is described} by a second-order polynomial as follows

Since the constant A of the polynomial is also unknown, the relationship found is represented in discrete form using a 3-component extended state vector, taking into account that the power values N now depend on W ₂ N = N (W ₂ ):

Dl 1D2 01 D, 00 1.

N _k with N = (N, N, A)

D = W ₂ , _k ., - W ₂ ,

Here, D is the non-equidistant step between two power values, since as a result of the transformation the power value (N [t] -> [N] W ₂ ) non-equidistance in W ₂ occurs.

With the measurement equation y _k = (100) N _k + V _k Cayman relationship for a 3rd order filter can be placed. This leads to the optimal prognosis of a state vector N _k . In a further method step, with the predicted state vector N _k and the total water quantity W _G predetermined by the recipe, the final power value is determined according to the relationship

Nend = N _k + N _k <W _G - W ₁₁₁₎ + V ₂ Ä _k (W _M - C _{2, k)} ²

with W _2k = mixed amount of liquid or water;

A _k = constant due to the design of the mixing plant; certainly.

The course of the predicted final value N _end during the wet mixing phase is shown in FIG. 1 d.

It has been found that at the latest at time t ₃ , which is still within the water intake phase, the predicted end value coincides with the actual later occurring later, with sufficient accuracy. Therefore, at this time, a comparison is made between the target final value N ₆ and the predicted end value N _end . The difference between the two serves as a criterion for controlling the water supply valve. If the difference between the setpoint end value and the predicted end value at the time of the comparison is greater than zero, the water inlet valve is closed before the recipe water quantity is reached; in the opposite case, the valve must be closed later.

From the absolute value of the difference, the duration of the changed valve timing is determined. After closing the water inlet valve, the rest of the water is mixed in during the period U to Ц , whereby, according to the relationship found, the following relationship now applies:

W ₂ (t) = W _G (1-e " ⁰ ') + W ₀ .

In order to control the mixing process, the state vector N _{k is} furthermore determined using the method step described. At the time t _s all the water is mixed in and it can be carried out again a comparison of N _end with N ₆ for quality control.

If several batches of the same formulation are mixed successively, the state vector N _{16 can be used} to determine the components of the state vector N ₀ . According to the invention, the component N _{0 is determined} during the dry mixing phase. Component A is a constant and can be applied directly. The component N ₀ becomes due to the found relationship with the relationship

N ₀ = N ₁₅ + A (W ₀ - W _G )

certainly.

At the same time, the end values of the components of the error covariance matrix of the Kalman filter are appropriately converted and used as initial values at time t ₂ . FIG. 1 e shows the course of the predicted end values N _end during a further mixture, when the above learning process took place in the period between t ₅ and t ₆ .

Claims (3)

claims: 1. A method for digital control of the liquid supply in the treatment of medium-viscosity mixtures by a mixing process with the aim of setting a desired consistency or viscosity, especially fresh concrete in stationary concrete mixing plants using the measurement of liquid supply and the active power of the drive motor of the mixing plant for the mixing process, wherein once at the beginning of the mixing process for a corresponding degree of filling of the mixing plant or for a given recipe by means of a teaching mixture a power value upon reaching the desired consistency or viscosity, determined as the final value and fixed as a plant parameter, wherein during the in in the period known from t, until t ₂ Wirkleistungsmeßwerte be determined and smoothed so that a free from interference, based on the current mixture dry mixing performance value at the end of the dry is prepared with the beginning of the water supply, from the time t ₂ to the currently supplied amount of water W _z = A _w t, from the measured power change taking into account the initial moisture N _{0 of} the current batch and the currently supplied water quantity W ₂ , is continuously closed on the currently mixed amount of water W ₂ (t), from the difference between the according to the recipe to obtain the desired consistency supplied amount of liquid W _G to the determined, currently mixed water amount W ₂ (t) predicted the expected final power value N _end is derived and from the performed in a known manner this expected final power value N _end with the desired consistency corresponding active power end of the teaching mixture, a control signal for the termination or continuation of the fluid supply. 2. A method for digital control of the liquid supply according to claim 1, characterized in that in the case of exceeding a mixture tolerance value of the comparison of the dry mixing power value with a fixed fixed as a plant parameter dry mix performance value ansich known measures to compensate for incorrect dosages and feed errors and deviations of the allowable nominal moisture content of the batch in the sense of a disaster regime at the beginning of the mixing process to be initiated. 3. A method for digital control of the liquid supply according to claim 1, characterized in that the actually mixed amount of liquid