Pump system and method for determining the flow in a pump system

Abstract

A pump system has at least one fluid container (2) which comprises an inlet (4) and an outlet (6), at least one pump (8) arranged in the inlet (4) or the outlet (6), and a control device (16) which includes a flow evaluation device for determining a flow through the fluid container (2) of the pump system. The flow evaluation device is configured such that the flow evaluation device uses a system model for determining the flow. The system model includes at least two different sub-models, a sub-model which describes the inflow behavior of the fluid container (2) and a sub-model which describes the outflow behavior of the fluid container (2). A corresponding pump flow evaluation method is provided.

Claims

1. A pump system comprising: at least one fluid container comprising an inlet and an outlet; at least one pump arranged in the inlet or in the outlet; and a control device comprising a flow evaluation device for determining a flow through the fluid container of the pump system, the flow evaluation device being configured such that the evaluation device uses a system model for determining flow, wherein: the system model is comprised of at least two different sub-models, comprising a sub-model which describes inflow behavior of the fluid container and a sub-model which describes outflow behavior of the fluid container; the control device comprises: a memory configured to store data acquired in the pump system, and a parameter evaluation device configured to determine parameters of the at least two sub-models on the basis of the stored data; and the parameter evaluation device is configured to determine parameters of the first sub-model simultaneously with a determination of parameters of the second sub-model.

2. A pump system according to claim 1, wherein the two different sub-models are of a different nature from each other.

3. A pump system according to claim 1, wherein the parameter evaluation device is configured to determine parameters of the at least two sub-models by way of error minimization between an estimated output variable which is determined by the sub-models, and a corresponding output variable which has been measured or calculated in the system by way of the use of the least-mean-squares method.

4. A pump system according to claim 1, wherein the fluid container is provided with a level sensor which detects the fluid level in the inside of the fluid container, wherein the detected fluid level is stored as part of the acquired data.

5. A pump system according to claim 1, wherein the at least one pump comprises an electrical drive motor and a power detection device which detects the current electrical power of the drive motor, wherein the electric power is stored as part of the acquired data.

6. A pump system according to claim 1, wherein a pressure sensor is arranged at the outlet side of the pump, said pressure sensor detecting the outlet pressure of the pump, wherein the outlet pressure is stored as part of the acquired data.

7. A pump system according to claim 1, wherein the control device is configured such that the control device detects a number of active pumps in the pump system or a rotation speed of the at least one pump, wherein a number of the active pumps or a rotation speed or both a number of the active pumps and a rotation speed is stored as part of the acquired data.

8. A pump system according to claim 1, wherein the first sub-model is a function of time and of at least one determined model parameter.

9. A pump system according to claim 1, wherein the second sub-model is a function of data which is acquired in the system, and of at least at least one determined model parameter.

10. A pump system according to claim 1, wherein the control device is configured such that the second sub-model is used in order to determine pump flow.

11. A pump system according to claim 1, wherein: the fluid container is a pump sump, and the pump is arranged in an outlet to pump fluid out of the pump sump; and the first sub-model describes the inflow into the pump sump and the second sub-model describes behavior of the pump.

12. A pump system according to claim 1, wherein: the pump is arranged in the inlet to fill the fluid container; and the first sub-model describes outflow out of the fluid container and the second sub-model describes behavior of the pump.

13. A pump flow evaluation method for determining outlet flow of a pump system, the method comprising: providing the pump system, which pump system comprises at least one pump and a fluid container comprising an inlet and an outlet; providing a control device comprising a flow evaluation device for determining a flow through the fluid container of the pump system, the flow evaluation device being configured such that the evaluation device uses a system model and a memory configured to store data acquired in the pump system; acquiring data from the pump system and storing the acquired data in the memory; with the control device determining outlet flow by the system model which is comprised of at least two sub-models, comprised of a sub-model which describes inflow behavior of the fluid container and of a sub-model which describes outflow behavior of the fluid container; providing the control device with a parameter evaluation device configured to determine parameters of the sub-models on the basis of the stored data; and with the parameter evaluation device, determining parameters of the first sub-model simultaneously with determining parameters of the second sub-model.

14. A pump flow evaluation method according to claim 13, wherein the at least two sub-models are of a different nature to the extent that they have a different behavior as a reaction to a change of input values.

15. A pump flow evaluation method according to claim 13, wherein during a continuous operation of the pump system the parameter evaluation device continuously determines and adapts parameters of the sub-models by continuously determining parameters of the first sub-model simultaneously with continuously determining parameters of the second sub-model, based on previously measured data stored in the data memory, which is provided with said step of acquiring data from the pump system during said continuous operation of the pump system.

16. A pump system according to claim 1, wherein the parameter evaluation device is configured to, during a continuous operation of the pump system, continuously determine and adapt parameters of the sub-models by continuously determining parameters of the first sub-model simultaneously with continuously determining parameters of the second sub-model, based on previously measured data stored in the data memory, which data is acquired during said continuous operation of the pump system.

17. A pump system comprising: at least one fluid container comprising an inlet and an outlet; at least one pump arranged in the inlet or in the outlet; and a control device comprising a flow evaluation device for determining a flow through the fluid container of the pump system, the flow evaluation device being configured such that the evaluation device uses a system model for determining flow, wherein: the system model is comprised of at least two different sub-models, comprising a sub-model which describes inflow behavior of the fluid container and a sub-model which describes outflow behavior of the fluid container; the control device comprises: a memory configured to store data acquired in the pump system, and a parameter evaluation device configured to determine parameters of the at least two sub-models on the basis of the stored data; the parameter evaluation device is configured to continuously determine parameters of the first sub-model and to continuously determine parameters of the second sub-model during a continuous operation of the pump system, based on previously measured data stored in the data memory, which data is from said continuous operation of the pump system; and the continuous determination of parameters of the first sub-model occurs simultaneously with the continuous determination of parameters of the second sub-model.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the drawings:

(2) FIG. 1 is a schematic view showing a pump system according to the invention, in the form of a pump sump with a pump which is arranged therein;

(3) FIG. 2 is a schematic view showing a flow evaluation device according to the invention;

(4) FIG. 3 is a diagram showing the course of the fluid level in dependence on the inlet flow and outlet flow in the pump sump according to FIG. 1;

(5) FIG. 4 is a representation according to FIG. 3, but with a fluctuating inlet flow;

(6) FIG. 5 is a view with several diagrams providing the evaluation of the inlet flow and of the pump flow on the basis of a system model;

(7) FIG. 6 is a schematic view showing a pump system which is suitable for the supply of water; and

(8) FIG. 7 is a view with diagrams which represent the fluid level in the fluid container according to FIG. 6, in dependence on the inflow and the outflow.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

(9) According to a first possible embodiment of the invention, the pump system according to the invention can be configured as a system for delivering waste water. FIG. 1 shows a corresponding arrangement. A pump sump 2 forms a fluid container, which is provided with an inlet 4 and with an outlet 6. The inlet 4 is situated on the upper side and the outlet 6 on the lower side of the pump sump 2. A pump 8 which delivers water or liquid out of the pump sump 2 is arranged on the outlet 6. In this example, the pump 8 is arranged outside the pump sump 2. A submersible pump however could also be applied. There are several relevant variables in such a pump system, specifically the inlet flow q.sub.in and the outlet flow q.sub.p, wherein the outlet flow q.sub.p corresponds to the delivery flow of the pump 8. Two pressure sensors 10 and 12 are provided in the system. The first pressure sensor 10 forms a level sensor. The pressure sensor 10 is arranged in the inside of the pump sump 2, for detecting the hydrostatic pressure at the base of the pump stump 2. The level h of the fluid in the inside of the pump sump 2 can be determined from the hydrostatic pressure in the known manner. The pressure sensor 12 is arranged in the outlet 6 or the outlet conduit downstream of the pump 8, i.e. on the delivery side of the pump 8 and detects the outlet pressure p.sub.out of the pump 8. A further variable which is relevant to the system is the electrical power consumption P of the electrical drive motor 14 of the pump 8. The pump 8 preferably comprises n integrated control device 16 which receives sensor signals from the pressure sensors 10 and 12 and also detects the electric power consumption P of the drive motor 14. The control device 16 can additionally control the electric drive motor 14 of the pump 8. The control device 16 moreover comprises a flow evaluation device and by way of this is in the position of determining the flows q.sub.in and q.sub.p. Alternatively this control device 16 can be an external device especially if there are two or more pumps in the system.

(10) This flow evaluation device 18 is represented schematically in FIG. 2. The flow evaluation device uses a system model 20 simulating the pump system, for the computation or for the approximate evaluation of the flows q.sub.p and q.sub.in. The system model 20 consists of two sub-models 22 and 24, whose function is described in more detail further below. The flow evaluation device 18 further comprises a data acquisition module 26 which continuously acquires or detects measured system parameters, in this example the height h of the fluid level in the pump sump 2 which is computed on the basis of the signal of the pressure sensor 10 in the control device 16, the differential pressure p across the pump 8, i.e. the pressure difference between the pressure sensors 10 and 12, the electrical power consumption P as well as a switching signal s which indicates as to whether the electrical drive motor 14 is switched on or off. This data which is continuously detected by the data acquisition module 26 is continuously stored in a memory means in the form of a data memory 28. Thereby, the newer data can regularly overwrite older data. A parameter evaluation device 30 which is likewise part of the flow evaluation device, in the manner described below determines the model parameters or parameters 32 for the sub-models 22 and 24, on the basis of the data stored in the data memory 28. On the basis of the sub-models 22 and 24 formed in such a manner, these determine the flows q.sub.p and q.sub.in on the basis of supplied, current data measured in the system.

(11) The use of the system model 20 for determining the flow through the pump system which corresponds to the outlet flow q.sub.p of the pump 8, avoids having to measure the flow in a direct manner. The efficiency of the pump changes relatively rapidly since the pump 8 is subjected to a wear and a contamination, so that a flow evaluation is not possible based solely on the electrical variables of the drive motor 14 and the measured pressures. A change of the level h over time t is also not a reliable variable for the outlet flow q.sub.p, if the inlet flow q.sub.in simultaneously changes. This is explained by way of FIGS. 3 and 4.

(12) The behavior of the system over time t is represented in FIG. 3 in three diagrams. The lower curve in FIG. 3 shows the outlet flow q.sub.p over time t. The middle curve shows the inlet flow q.sub.in over time t and the upper curve shows the height h of the fluid level in the pump sump 2 over time t. It can be seen that the inlet flow q.sub.in is constant. The pump 8 is switched on at the point in time T.sub.1 and is switched off again at the point in time T.sub.3. The level h rises up to the point in time T.sub.1 due to the constant inlet flow q.sub.in. The inlet flow q.sub.in in this time interval is proportional to the increase of the level h. The level h drops again on switching on the pump 8 at the point in time T.sub.1, wherein, as is represented in the equations in FIG. 3, the change of the level h over time t is proportional to the difference of the inlet flow q.sub.in and the outlet flow q.sub.p. If the inlet flow q.sub.in is constant, as is represented in FIG. 3, then the outlet flow q.sub.p can be determined from the difference of the level change in the case of a switched-on and switched-off pump 8, without further ado. This is no longer possible of the inlet flow q.sub.in changes, as is represented in FIG. 4.

(13) The three curves in FIG. 4 correspond to the curves in FIG. 3. In contrast to FIG. 3, the inlet flow q.sub.in is not constant in the operating condition according to FIG. 4, but increases at the point in time T.sub.0 and reduces at the point in time T.sub.2, as is represented in FIG. 4. As can be recognized in the upper curve, the speed at which the level h in the pump sump 2 rises, increases with the increase of the inlet flow q.sub.in at the point in time T.sub.0 in the interval t.sub.4. Accordingly, the speed of the dropping of the level h is lower in the time intervals t.sub.5, t.sub.6 and t.sub.7 between the points in time T.sub.1 and T.sub.2, than in the comparable interval in the operating condition according to FIG. 3. The speed at which the level h in the pump sump 2 drops, increases again in the interval t.sub.8, with the reduction of the inlet flow q.sub.in at the point in time T.sub.2. It is to be recognized that one cannot deduce the outlet flow q.sub.p solely from the speed at which the level h changes, if the inlet flow q.sub.in changes, since the temporal change of the level h is always proportional to the difference between the inlet flow q.sub.in and the outlet flow q.sub.p.

(14) The system model 20 is applied in the manner described hereinafter, in order to be able to determine the outlet flow q.sub.p also in such operating conditions. The system model 20 consists of the two sub-models 22 and 24. What is essential to the system is the fact that the sub-models 22 and 24 are of a different nature. The sub-model 22 describes the inflow behavior, i.e. the inflow q.sub.in, whereas the sub-model 24 describes or represents the outflow behavior in the form of the outlet flow q.sub.p. The first model 22 is thereby dependent on a parameter and time t, i.e. q.sub.in=f ( t). The second sub-model 24 is of a different nature and is dependent on a parameter , the switch-on signal s, the electrical power P and the differential pressure p between the pressure sensors 10 and 12, i.e. q.sub.p=g(, s, P, , p).

(15) The following equation results due to the fact that the inlet flow q.sub.in and the outlet flow q.sub.p, as represented by way of FIGS. 3 and 4, are dependent on the change of the height h in the pump sump 2:

(16) $A\left(h\left(t\right)\right)\frac{h\left(t+t\right)-h\left(t\right)}{t}=f\left(,t\right)-g\left(,s\left(t\right),P\left(t\right),p\left(t\right)\right)$
In this formula, h corresponds to the level of the fluid level in the pump sump 2, t to the time, t a time interval and A(h) to the cross-sectional area of the pump sump 2, wherein the cross-sectional area can be a function of the height h, if the pump sump 2 does not have a cross section which is constant over the height. The cross-sectional area A(h) of the pump sump 2 is assumed as being known in the subsequent consideration.

(17) The following model can be applied as a first sub-model 22 representing the feed or inflow behavior:
f(,t)=.sub.0+.sub.1 atan(.sub.2t+.sub.3)
The following model can be applied for example as a second sub-model 24 which represents the outflow behavior:
g(,s,P,p)=.sub.0s+.sub.1sP+.sub.2s.sub.p
The two sub-models 22 and 24, apart from the different input variables, comprise different model parameters or parameters .sub.0, .sub.1, .sub.2, .sub.3 and .sub.0, .sub.1, .sub.2 respectively, which are defined by the parameter evaluation device 30.

(18) It is to be understood that the previously mentioned models are only examples. Differently formed models which are different with regard to their nature can also be applied as a sub-model 22 as well as sub-model 24. Thus for example the sub-model 24 which represents the outflow behavior can also be simplified:
g(,s,P)=.sub.0s+.sub.1sP
Such a model representing the pump 8 is advantageous for example, since the outlet pressure p.sub.out is not necessary as an input variable for this. Such a pressure detection is not common in many waste-water installations. Inasmuch as this is concerned, the design of the installation is simplified here. Another simplification of the model could be as follows:
g(,s,p)=.sub.0s+.sub.1sp
Such a sub-model 24 representing the outflow behavior or the behavior of the pump 8 has the advantage that no electrical variable of the pump 8 needs to be detected, In contrast, this model is merely based on the model parameters .sub.0 and .sub.1, the switch signal s, the differential pressure p between the pressure sensors 10 and 12, as well as time t.
A more extended version of the sub-model 24 with speed information is:

(19) $g\left(,\mathrm{sP},p,\right)={}_0\frac{s}{}+{}_1\frac{\mathrm{sP}}{{}^2}+{}_2\frac{sp}{}+{}_{3}s$
This version is preferably used if the pump is controlled by a frequency converter as the speed information can be received from the frequency converter.

(20) In a case in which more than one pump is to be arranged in a pump sump 2, which often occurs, there are two different approaches to accordingly represent this by models. On the one hand it is possible to form a model in each case for each pump and to take into account a switch-on, which is to say start/stop signal s for each of the two pumps 8. Two second sub-models 24 would then result in such a case, for example in the form:
q.sub.1=g.sub.1(.sub.1,s.sub.1,P.sub.1,p),q.sub.2=g.sub.2(.sub.2,s.sub.2,P.sub.2,p),
wherein there are two parameter sets .sub.1 and .sub.2, as well as switch-on signals S.sub.1, S.sub.2, one for each pump in each case.

(21) On the other hand, it is possible to use a model or sub-model which simulates both pumps, if both pumps or several pumps 8 are simultaneously switched-on, inasmuch as it is the case of pumps 8 of the same type. In this case, the variable s would not be a pure start/stop signal representing the switching-on and switching-off of the pump, but a signal which simultaneously represents how many pumps 8 are simultaneously switched on. In this case the electrical power P would represent the average power of one pump, I. e. the sum of the power from all active pumps divided with the number of active pumps.

(22) The parameters and of the sub-models 22 and 24 on operation of the pump system are continuously determined and adapted, by the parameter evaluation device 30 on the basis of the previously measured data stored in the data memory 28. The thus adapted parameters 32 (, ) then form the basis for determining or identifying the inlet flow q.sub.in and the outlet flow q.sub.p. A continuously running adaptation or optimization of the models is thus effected, so that these sub-models 24 and 24 simulate or represent the system as exactly as possible.

(23) The manner of functioning of the models is explained further by way of FIG. 5. The upper curve in FIG. 5 shows the change of the height h over time t. The second curve shows the electrical power consumption P of the drive motor 14 over time t, and the switch signal s which represents the switched-on condition of the drive motor 14. One can recognize that the pump 8 is switched on between the point in time T.sub.1 at about 13 seconds and the point in time T.sub.3 at about 42 seconds. The third curve in FIG. 5 shows the pressure signal which results therefrom, for the differential pressure p between the two pressure sensors 10 and 12. The pressure difference p increases on switching on the pump 8.

(24) The fourth curve in FIG. 5 parallel to this shows the pump sump flow q.sub.pit, i.e. the flow q.sub.pit which leads to the rising and dropping of the level h in the pump sump 2. The pump sump flow q.sub.pit is the difference between the inlet flow q.sub.in and the outlet flow q.sub.p. The actual pump sump flow q.sub.pit which results from the measurement of the height h and the known cross section A(h) of the pump sump 2 is represented in the fourth curve in FIG. 5 as an unbroken line. The dashed line shows the estimated pump sump flow q.sub.pit,est which is determined on the basis of the sub-models 22 and 24.

(25) The pump sump flow q.sub.int,est which is thus determined by the models is based on the inlet flow q.sub.in,est and the outlet flow q.sub.p,est which are determined by the sub-models 22 and 24 and are represented in the lower curve in FIG. 5. One can recognize that the models represent the actually measured pump sump flow q.sub.int in an accurate manner.

(26) The pump sump flow q.sub.int results according to the following equation:

(27) $q_{\mathrm{pit}}\left(t\right)=A\left(h\left(t\right)\right)\frac{h\left(t+t\right)-h\left(t\right)}{t}=q_{\mathrm{in}}\left(t\right)-q_p\left(t\right)$
The variables used in the equation correspond to the variables which are described above.

(28) The deduction of the parameters and for the sub-models 22 and 24 can be effected by way of comparing the result variables determined by the models, with actually measured, corresponding variables, for example whilst using the least-mean-squares method. As explained above, the pump sump flow q.sub.int can be computed in dependence on the time t according to the previous equation from the level h which is measured in the pump sump 2, i.e. this variable can actually be measured and compared to the pump sump flow q.sub.pit,est which is determined on the basis of the models. With regard to the pump sump flow q.sub.pit,est which is determined on the basis of the models, it is the case of an estimated pump sump flow which results from the sub-models 22 and 24 which are described above. A method which minimizes the prediction errors is applied. The model parameters .sub.0, .sub.1, .sub.2, .sub.3 as well as .sub.1 and .sub.2 are accordingly adapted for this.

(29) The aim of the flow evaluation device 18, with the embodiment example according to FIG. 1 is to determine the outlet flow q.sub.p. Actually, only the second sub-model 24 is necessary for this. However, in order to be able to determine its parameter , the first sub-model 22 representing the inlet flow q.sub.in is necessary, in order in the described manner to determine the model parameters and by way of comparing the estimated pump sump flow q.sub.pit,est with the measured pump sump flow q.sub.pit. The sub-model 22 in this case thus forms an auxiliary model which is used for parameter evaluation.

(30) FIG. 6 shows a second embodiment example of the invention. The embodiment example shows a pump system for the water supply. A tank 2 is provided here as a fluid container, wherein the pump 8 is arranged in its inlet 4. The tank 2 lies above the pump 8, so that this pump pumps water or fluid into the tank 2 in an inlet flow q.sub.p. The fluid can escape out of the tank 2 via the outlet 6 on account of gravity. An outlet flow q.sub.out is therefore formed. A level sensor 34 which for example can likewise be configured as a pressure sensor, as is described by way of FIG. 1, is provided for determining the fluid level or height h in the tank 2. A pressure sensor 36 which detects the outlet pressure pp of the pump 8 is arranged here at the exit side of the pump 8. In this case too, a first sub-model 22 which represents the inlet flow, in this case the pump flow q.sub.p and a second sub-model 24 which represents the outlet flow q.sub.out can also be applied in the corresponding manner in the control device for determining the flow. Here too, the flow in the tank q.sub.pit is dependent on the difference of the flows q.sub.p and q.sub.out. Inasmuch as this is concerned, the parameter evaluation of the parameters and for both sub-models 22 and 24 can be effected in a manner which corresponds to that which was previously described. A Fourier series for example can be applied as a model for the outlet flow q.sub.out
q.sub.out=f(,t)=.sub.0+.sub.1 cos(t)+.sub.2 sin(t)+ . . .
or an approximation polynomial
q.sub.out=f(,t)=.sub.0+.sub.1t+.sub.2t.sup.2+ . . .
The pump flow q.sub.p which corresponds to the inlet flow, as in the first embodiment example, can be approximated as a function g (, s, P, p). Thereby, p is the differential pressure across the pump 8 and the switching variable s represents the number of active pumps in the case that several pumps are arranged in parallel, as has been described beforehand.

(31) The differential pressure can be calculated as a delivered pump pressure when the pump is operating subtracted from a system pressure, where the system pressure is evaluated and stored when the pump is stopped e. g. both the delivered pump pressure and the system pressure can be measured with one pressure sensor pp 36.

(32) FIG. 7 shows the course of the height h over time and accordingly the change of the outlet flow q.sub.out and of the pump flow q.sub.p, wherein the pump 8 is switched on at the point in time T.sub.1 and is switched off at the point in time T.sub.3.

(33) As is schematically represented in FIG. 2, a continuous adaptation of the sub-models 22 and 24 can be simultaneously effected with the flow evaluation also with the second embodiment example according to FIGS. 6 and 7.

(34) While specific embodiments of the invention have been shown and described in detail to illustrate the application of the principles of the invention, it will be understood that the invention may be embodied otherwise without departing from such principles.

APPENDIX

List of Reference Characters

(35) 2 pump sump 2 tank 4, 4 inlet 6, 6 outlet 8, 8 pump 10 pressure sensor, level sensor 12 pressure sensor 14 drive motor 16 control device 18 flow evaluation device 20 system model 22, 24 sub-models 26 data acquisition module 28 data memory, memory means 30 parameter evaluation device 32 parameter 34 level sensor 36 pressure sensor q.sub.in inlet flow of the pump sump q.sub.out outlet flow q.sub.p outlet flow of the pump sump, pump flow q.sub.pit pump sump flow, q.sub.pit=q.sub.pq.sub.out or q.sub.pit=q.sub.inq.sub.p h level P electric power P.sub.out outlet pressure of the pump s switch-on signal which represents the number of pumps which are switched on p differential pressure across the pump t time t t time intervals E(, t) mathematic model which describes the inflow behavior parameter vector which comprises the parameters of the model for the inflow behavior .sub.i parameter g (, s, P, p) mathematic model which describes the pump flow or the outflow behavior parameter vector which contains the parameters of the model describing the pump flow i parameter