Method for controlling a pump system which comprises at least two centrifugal pumps operated hydraulically in parallel, and multi-pump system

EP4684134A1Pending Publication Date: 2026-01-28KSB SE & CO KGAA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
EP2024712237
Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-03-22
Filing Date
2024-03-15
Publication Date
2026-01-28

AI Technical Summary

Technical Problem

Existing methods for controlling multi-pump systems with hydraulically operated centrifugal pumps are inefficient due to manufacturing tolerances and structural deviations, requiring complex and time-consuming adjustments to achieve optimal operation, and are not adaptable to changing system conditions.

Method used

The implementation of an efficiency optimizer subordinate to the hydraulic controller, which evaluates total power consumption and adjusts pump speeds to minimize overall power consumption while maintaining hydraulic target variables, using iterative methods and gradient approaches to determine optimal speed distributions.

Benefits of technology

This approach reduces the total power consumption of the system, increasing overall efficiency and simplifying the control process by allowing for dynamic adjustments to optimize pump performance based on real-time conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2024057038_26092024_PF_FP
    Figure EP2024057038_26092024_PF_FP
Patent Text Reader

Abstract

The invention relates to a method for controlling a pump system which comprises at least two centrifugal pumps operated hydraulically in parallel, each of which is driven by speed-controlled, electromotive drive units, wherein the centrifugal pumps are speed-controlled by means of a common hydraulic controller which, depending on a controlled variable, outputs a synchronous speed setpoint for all centrifugal pumps as a manipulated variable, wherein an efficiency optimizer determines speed offset values for each pump on the basis of the summated power inputs of at least two centrifugal pumps and adjusts the speed setpoint output by the hydraulic controller for each pump individually by means of the offset speed value for each pump, such that the summated power input of the at least two pumps is reduced, in particular minimized.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Description

[0002] Method for controlling a pump system comprising at least two hydraulically parallel operated centrifugal pumps, and multi-pump system

[0003] The invention relates to a method for controlling a pump system having at least two hydraulically parallel operated centrifugal pumps, each driven by speed-controlled, electric motor drive units, wherein the centrifugal pumps are speed-controlled by a common hydraulic controller which outputs a synchronous speed setpoint for all centrifugal pumps as a manipulated variable depending on a controlled variable.

[0004] In multi-pump systems, two or more centrifugal pumps can be operated hydraulically in parallel. As an example, reference is made to the example shown in Figure 1, which features a total of four centrifugal pumps Pu1-Pu4 connected hydraulically in parallel, which pump into a common pressure accumulator 1. The control of the centrifugal pumps Pu1-Pu4 shown is handled by a higher-level, central control unit (hydraulic controller) to adjust the multi-pump system to a predefined control variable.

[0005] The control unit not only decides on the number of active pumps; in the embodiment shown here in Figure 1, the two pumps Pu1 and Pu2 are active, while the two pumps Pu3 and Pu4 are inactive. If the pumps are equipped with a frequency converter, their speed n1-n4 can be varied. The speed is output by the control unit as a manipulated variable depending on a controlled variable. According to current knowledge, it is usual for the active pumps to operate at synchronous speed, i.e. the two active pumps Pu1, Pu2 are both driven at the synchronous speed nsoll, while the speed of the inactive pumps Pu3, Pu4 is zero.

[0006] This approach is based on the assumption that the overall efficiency of the system is lowest when the active pumps run at synchronous speed. However, this is theoretically only the case under ideal conditions, i.e., all pumps are constructed exactly identically and the respective piping from the pumps to the collector also all has the same hydraulic resistance. In practice, however, significant deviations occur due to manufacturing tolerances and structural conditions.

[0007] A first approach to solving this problem is to adjust the higher-level hydraulic controller to the specific conditions of the system, specifically taking into account the different hydraulic resistances and line lengths in the hydraulic control equations, thereby controlling the pumps with an appropriate speed distribution. However, this approach is quite complex and time-consuming, as the control system must be adjusted on-site during commissioning. Another disadvantage is that the characteristics of the system can change over time, requiring readjustment of the control system.

[0008] Therefore, an optimized and also easier to implement approach is being sought to cover such a scenario and to achieve an optimization of the control of a multi-pump system.

[0009] This object is achieved by a method having the features of claim 1. Advantageous embodiments of the method are the subject of the dependent claims. Furthermore, the object is achieved by a multi-pump system according to the features of claim 12. According to the invention, it is proposed to expand the generic system to include an efficiency optimizer. The efficiency optimizer is arranged downstream of the higher-level control unit, i.e. the hydraulic controller. The efficiency optimizer initiates a subsequent adjustment of the manipulated variable output by the hydraulic controller, i.e. the synchronous speed. For this purpose, the efficiency optimizer evaluates the summed power consumption of the active centrifugal pumps. Based on this, an adjustment of the manipulated variable of the hydraulic controller, i.e. the common setpoint speed value output by the hydraulic controller for all active pumps, is initiated. As a result, the active pumps receive pump-specific setpoint specifications.By specifically adjusting the speed values ​​for individual pumps, the efficiency optimizer optimizes the total power consumption, specifically minimizing it as much as possible. It is crucial that the subsequent adjustment of the speed value still achieves the hydraulic target value set by the hydraulic controller. If the speed is adjusted by the energy efficiency optimizer, the hydraulic controller will also adjust the synchronous speed setting to achieve the hydraulic target value. Overall, this reduces the power consumption of the entire system, increasing the overall efficiency of the system.

[0010] It can be provided that the efficiency optimizer is already provided with the total power consumption of all active centrifugal pumps, meaning that the efficiency optimizer only knows the total power consumption, but not the individual power consumption of each centrifugal pump. Alternatively, it is also possible for the efficiency optimizer to receive the corresponding power consumption of each centrifugal pump via the individual frequency converters of the system and determine the total power consumption of the system based on this.

[0011] According to a particularly simple embodiment of the method, in particular for the case where two pumps are actively operated, it is conceivable for the pump-individual speed offset values ​​to have different signs, i.e. the speed setpoint of one pump is reduced while the speed setpoint of the other centrifugal pump is increased. It is particularly simple and conceivable if, for two active centrifugal pumps, speed offset values ​​of the same amount are determined which only differ from one another in their sign. With more than two active centrifugal pumps, the amounts of the offset values ​​can differ from one another, but it is advantageous if at least two of the active centrifugal pumps have a different sign in their offset values.

[0012] The method according to the invention is preferably carried out iteratively, i.e., the efficiency optimizer determines a speed offset value. After adjusting the centrifugal pumps to the adjusted speed setpoint, the efficiency optimizer checks the updated power consumption of the entire system, particularly to determine whether it has decreased or increased due to the adjustment of the speed setpoint. For example, if the total power consumption is reduced, the increment of the individual offset values ​​is simply increased again. If the measure has led to an increase in power consumption, the signs of the respective offset values ​​are reversed instead, and the process is repeated.

[0013] Ideally, the process is carried out iteratively until an adjusted individual target speed for all pumps is found, resulting in a minimum total power consumption of the system. Preferably, a suitable search method is implemented that uses the minimum total power consumption of the active centrifugal pumps as its target variable. A search method based on the gradient approach is preferably used.

[0014] It is particularly important that the speed offset value increments, i.e., the adjustment of the target speed using the offset values, is not carried out abruptly, but rather sufficiently slowly with a comparatively small increment to maintain the stability of the control system and minimize the influence on the hydraulic controller. For example, adjusting the target speed by offset values ​​in a range of 10 rpm is conceivable. 1 up to 50 rpm' 1, preferably in the range of 10 to 30 rpm' 1 It is not necessary for the step size to be adjusted to the same value for all active centrifugal pumps. It is also conceivable that different values ​​for the step size, especially different amounts for the step size, can be configured individually for each pump. It is also conceivable that a common value for the step size is determined, but this value is weighted individually for each centrifugal pump.

[0015] Furthermore, for the stability of the control of the system, in particular of the hydraulic controller, it is necessary to specify a maximum delta speed value by which the efficiency optimizer is allowed to change the synchronous target speed value initially specified by the hydraulic controller.

[0016] The hydraulic controller regulates the system, for example, to a target flow rate or a target head. In the latter case, proportional pressure control or differential pressure control is also conceivable. Furthermore, the hydraulic controller can regulate along a control curve that outputs the setpoint for the controlled variable depending on a parameter. Theoretically, this method can be used for various hydraulic controllers that regulate the system to any hydraulic parameter.

[0017] In addition to the method according to the invention, the invention also relates to a multi-pump system with at least two hydraulically parallel-operated centrifugal pumps, each driven by speed-controlled, electric motor drive units, wherein a hydraulic controller is provided which outputs a synchronous speed setpoint for all centrifugal pumps as a manipulated variable depending on a controlled variable. According to the invention, an efficiency optimizer is used here, which is configured to determine pump-specific offset values ​​based on the summed power consumption of at least two centrifugal pumps and to adapt the output speed setpoint of the hydraulic controller for each pump individually using the pump-specific speed offset value such that the summed power consumption of the at least two pumps is reduced, in particular minimized.The at least two hydraulically operated centrifugal pumps can be of the same design, especially identical pumps. However, a characteristic feature is that the hydraulic resistance in the piping of the parallel-connected pumps differs. In particular, the hydraulic resistance differs from the pumps to the pressure accumulator, so that even with synchronous speeds of the two pumps, the overall system efficiency is generally not optimized.

[0018] In particular, the multi-pump system is configured to carry out the method according to the invention. The above statements in connection with the method according to the invention therefore apply equally to the multi-pump system.

[0019] Further advantages and features of the invention will be described below with reference to an embodiment illustrated in the figures. They show:

[0020] Figure 1 : a simplified circuit diagram of a multi-pump system,

[0021] Figure 2: a QH diagram with pump characteristic curve and parabola of the optimal efficiency

[0022] Figure 3: Principle representation of the method according to the invention,

[0023] Figure 4: Time courses of the search parameter and the target variable during the execution of the search procedure,

[0024] Figure 5: an enlarged view of the time courses of the search parameter and the target variable during the first two iterations of the search procedure,

[0025] Figure 6: Time courses of the search parameters and the target variable in an execution of the search method with several search parameters, Figure 7: the speed courses for two active pumps of a multi-pump system in the simulated execution of the method according to the invention

[0026] Figure 8: the curves of the electrical power consumption of two active pumps of a multi-pump system during the simulated execution of the method according to the invention and

[0027] Figure 9: the curve of the electrical power consumption of the entire system during the simulated execution of the method according to the invention

[0028] The starting point is the multi-pump system shown in Figure 1, with, as an example, four centrifugal pumps Pu1-Pu4 connected hydraulically in parallel. All pumps Pu1-Pu4 are equipped with a frequency converter. As in the prior art, a higher-level hydraulic controller generates a synchronous speed nsetpoint for the active pumps as a control variable to regulate the system to a setpoint, e.g., a setpoint discharge head at the pressure accumulator.

[0029] The method according to the invention now provides an extension whereby the system can determine the optimal speed distribution for two or more active pumps.

[0030] To clarify the technical background for the process implementation, the relationships are briefly explained using the Q / H diagram shown in Figure 2. It is assumed that the pumps Pu1 and Pu2 are identical, i.e. both have the same pump characteristic curve, which is shown here for the speed n and marked with the reference symbol 3. Also shown is the curve of highest efficiency, marked with the reference symbol 4. The operating points of the two pumps Pu1, Pu2 are labeled OP1 and OP2. Ideally, both operating points OP1, OP2 are as close as possible to curve 4 of highest efficiency, since this is when the overall efficiency of the system is at its highest. Due to the different piping, the two operating points OP1, OP2 are not on top of one another despite the identical speed n. In the example in Figure 2, the operating point of pump Pu2 (OP2) is closer to curve 4 of highest efficiency than the operating point OP1 of pump Pu1.Increasing the speed of pump Pu2 while simultaneously decreasing the speed of pump Pu1 results in the efficiency of pump Pu2 contributing more heavily to the overall efficiency of the system. This can improve the overall efficiency of the system. It is important that the sum of the flow rates (Q ges ) remains unchanged. This is ensured by the higher-level hydraulic controller.

[0031] According to the invention, an iterative process is implemented for the downstream adjustment of the synchronous speed output by the hydraulic controller, whereby the active pumps individually change their speeds to increase the overall efficiency of the system. Details can be found in Figure 3.

[0032] Figure 3 shows the higher-level hydraulic controller 10, which controls the system in such a way that one or more predetermined hydraulic target values ​​are achieved.

[0033] Possible target values ​​are, for example, the total head H ges of the system or the total flow rate of the system Q ges For simplicity, the following will always refer to the controlled variable "total flow rate." However, the procedure can be applied equally well if the controller were to control another hydraulic variable in the system.

[0034] When the hydraulic controller 10 regulates the system to a specific total flow rate Qsoii, it generates a common speed setpoint nsoll for both pumps Pu1, Pu2. According to the inventive method described here, an efficiency optimizer 20 is used in addition to the hydraulic controller 10. The efficiency optimizer 20 determines speed offset values ​​for both pumps Pu1, Pu2, which are added to the common setpoint speed nsoll of the hydraulic controller 10, resulting in individual speed setpoints n1, n2 for both pumps Pu1, Pu2. In the example shown here, offset values ​​(An) of the same amount are determined, but added together with different signs. This speed offset with different signs leads to a change in the electrical power consumption P1, P2 of the individual pumps Pu1 and Pu2 as well as the total electrical power consumption of the system P_total = P1 + P2.The hydraulic variables of the system, however, do not change, as the hydraulic controller 10 immediately adjusts them by adjusting the target speed nsoll. The efficiency optimizer 20 receives the power values ​​P1, P2 and the total power consumption P_total from both pumps Pu1 and Pu2. The efficiency optimizer 20 then calculates a new offset value An from the change in the total power consumption compared to the value of a previous iteration. In this way, the energy optimizer 20 iteratively searches for the point at which the total power consumption of the system is at its lowest.

[0035] The speed change should not occur abruptly for each iteration, but as slowly as possible to avoid pressure surges in the system. It is also advisable to specify 20 limits for the optimizer regarding how far it can change the speed (e.g., + / - 200 rpm).

[0036] To optimize the offset value(s) An, the efficiency optimizer 20 preferably uses a search method that has the minimum total power consumption as its target variable and is simultaneously based on a gradient approach. One possible approach would be the Kiefer-Wolfowitz method, as this approach is very memory and processor-efficient and can be easily run on the controller for a single pump. A further advantage is that this search method can also vary more than one variable. The method could thus be easily extended to any number of pumps. Furthermore, pumps with different hydraulic characteristics could also be used together in a multi-pump system.

[0037] If the process is extended to more than two pumps, it is not necessary for the speed offset An to be identical for the pumps, but simply to have a different sign. Each pump could have its own individual speed offset. The hydraulic controller ensures that the hydraulic setpoints (Q_total) are still maintained.

[0038] The following will describe in more detail the process steps for optimizing the offset values ​​An and minimizing the power consumption P_ges.

[0039] In general, the goal of the search procedure is to minimize a defined value (here, the total power consumption of the pumps). This value is called the "target variable." To achieve this goal, optimal parameters, referred to below as "search parameters," are sought. In the case of the efficiency optimizer, these are the speed values ​​n1, n2, ... of the active pumps. The number of search parameters can be unlimited. The target variable and the search parameters to be optimized must be defined before optimization. Furthermore, suitable starting values ​​for the search parameters must be specified.

[0040] The procedure for this search method can be explained using a simple theoretical example with only one search parameter:

[0041] Step 1 : Check whether increasing or decreasing the search parameters minimizes the target size (power consumption).

[0042] To do this, an offset is added to the default search parameter and the target variable is determined using simulation. The same offset is then subtracted from the search parameter and the new target variable is determined using simulation.

[0043] Step 2: Specify a new value for the search parameter

[0044] The gradient between the results of the two simulations (with negative and positive offset) is calculated. Based on this, a new value for the search parameter is calculated.

[0045] A system can be described by the following equation: y = (1.5 - a) 2 Here, y is the target variable to be minimized. The search parameter to be varied is a. A = 1 is set as the initial value. Figure 4 shows the behavior of a and y during the optimizer run. The results show that a value for the search parameter a can be found that significantly minimizes the target variable y.

[0046] To better illustrate the functionality, the first two iterations are shown enlarged in Figure 5. The process starts at point 1 (default value). An offset is subtracted from the search parameter a, and the simulation starts again (point 2). Subsequently, the same offset is added to the search parameter a, and the simulation is repeated (point 3). The simulation in this example shows that the negative offset (point 2) increased the target variable, while the positive offset (point 3) reduced it. Therefore, the search parameter for a is chosen to be higher in the next iteration (point 4).

[0047] Parameterization:

[0048] This determines the offset value (points 2 and 3). The more noisy the raw signal is, the higher the offset must be to obtain valid results. If the raw signal is not noisy, a smaller value can be selected.

[0049] Step size (for point 4):

[0050] This determines the size of the step size (point 4). If the starting values ​​are already good and optimization is only desired in the short range, a small value can be set here. If no good starting values ​​are known and the entire value range is to be searched, a large step size should be selected here.

[0051] If there is more than one search parameter, it may be necessary to select different step sizes. However, there is only one parameter for the step size. A weighting can then be defined for each search parameter. For example, if there are three search parameters (A, B, C), there are also three weighting parameters. If one of the search parameters should have a smaller step size than the others, the corresponding weighting parameter is selected to be smaller than the other weighting parameters.

[0052] Minimum parameter, Maximum parameter (For point 4):

[0053] It is often necessary to limit the range of variation of the search parameters. When, as in the present invention, controller parameters are optimized in a running system, instabilities must be avoided.

[0054] Settling time:

[0055] If a search parameter has been changed, it may take some time for the target variable to settle. This settling process must be waited for before the results are evaluated.

[0056] Measurement duration:

[0057] When the search method, as in the present invention, is run in a real system, the target variable signal is often overlaid with noise. This noise is averaged for evaluation by the optimizer. The averaging time is determined by this parameter.

[0058] The example outlined here can be extended to multiple search parameters. An example with three search parameters is outlined below:

[0059] The objective variable y is to be minimized. Here, y depends on the search parameters a, b, and c, which are to be varied, according to the following exemplary equation: y = (a+2*b-3*c -40) 2 ;

[0060] The diagrams in Figures 6a-6d show the course of a, b, c and the target variable y.

[0061] The left diagram of the respective figures 6a-6d shows the curves without noise, the right diagram shows the result of a simulation under the assumption that the raw signal y, which goes into the optimizer, is superimposed by noise.

[0062] In this example, a minimum for the target variable is also found. It is noticeable that the target variable is minimized both with and without noise, but with different values ​​for the three search parameters. This is not unusual because there are multiple solutions to the described problem and the solution path is altered by the noise. Furthermore, it is noticeable that in the approach without noise, the search parameters and the target variable move evenly toward the optimum. In the approach with noise, however, the signals sometimes change direction abruptly. Nevertheless, the search method is well suited to finding an optimum even in the presence of noise.

[0063] The inventive search method was investigated for two pumps using a simulation. The two pumps are installed in such a way that the hydraulic resistance at the outlet nozzle is different. The method was implemented as previously described with reference to Figure 3 and the following explanations of the search method.

[0064] Figure 7 shows the speeds of the two pumps Pu1, Pu2 over time. Figure 8 shows the electrical power consumption of the individual pumps Pu1, Pu2 and Figure 9 shows the electrical power consumption of the entire system P_total = P1 + P2. At the start of the search process, both speeds are identical (synchronous target speed of the hydraulic controller 10). Over time, the speed n2 of pump Pu2 increases and the speed n2 of pump Pu1 decreases. As a result, the power consumption P2 of pump Pu2 also increases and the power consumption P1 of pump Pu1 decreases. The total power P_total decreases. The operating point is therefore reached with a lower power consumption P_total.

Claims

Patent claims 1. Method for controlling a pump system which has at least two centrifugal pumps (Pu1-Pu4) operated hydraulically in parallel, which are each driven by speed-controlled, electric motor drive units, wherein the centrifugal pumps (Pu1-Pu4) are speed-controlled by a common hydraulic controller (10) which outputs a synchronous speed setpoint nsoll for all centrifugal pumps (Pu1-Pu4) as a manipulated variable as a function of a controlled variable, characterized in that an efficiency optimizer (20) determines pump-specific speed offset values ​​on the basis of the summed power consumption P_ges of at least two centrifugal pumps (Pu1-Pu4) and adapts the speed setpoint nsoll output by the hydraulic controller for each pump (Pu1-Pu4) individually by means of the pump-specific speed offset value such that the summed power consumption P_ges of the at least two pumps (Pu1-Pu4) is reduced, in particular minimized.

2. Method according to claim 1, characterized in that the efficiency optimizer (20) receives the individual power consumptions P1, P2 of the pumps (Pu1-Pu4), in particular from the frequency converters, and sums them.

3. Method according to claim 1, characterized in that the efficiency optimizer (20) receives the summed power consumption P_ges.

4. Method according to one of the preceding claims, characterized in that the determined pump-individual speed offset values ​​of the at least two pumps (Pu1-Pu4) have different signs and are summed to the speed setpoint values ​​nsetpoint of the pumps (Pu1 -Pu4).

5. Method according to one of the preceding claims, characterized in that a speed offset value An of the same amount is determined for the at least two pumps (Pu1-Pu4), wherein preferably the target speed of one pump (Pu1-Pu4) is increased by the speed offset value An and the target speed of the other pump (Pu1-Pu4) is reduced by the speed offset value An.

6. Method according to one of the preceding claims, characterized in that the method is carried out iteratively in that the efficiency optimizer (20) checks the summed power consumption P_ges after each adjustment of the target speed values ​​and iteratively determines the point at which the summed power consumption P_ges of the centrifugal pumps (Pu1-Pu4) is lowest.

7. Method according to claim 6, characterized in that the efficiency optimizer (20) carries out a search method with the summed power consumption P_ges of the centrifugal pumps (Pu1-Pu4) as the target variable, in particular a search method according to a gradient approach.

8. Method according to one of the preceding claims, characterized in that the step size of the speed offset value by which the speed setpoint may be changed per iteration lies in a definable range, in particular in a value range from 10 rpm to 50 rpm, preferably in the range between 10 rpm and 30 rpm.

9. Method according to claim 8, characterized in that the step size is defined differently for each centrifugal pump (Pu1-Pu4) or a common step size is individually weighted for the at least two centrifugal pumps (Pu1-Pu4).

10. Method according to one of the preceding claims, characterized in that a maximum delta speed value can be specified for the energy optimizer (20), which defines the maximum change value of the target speed of the pump (Pu1 -Pu4) compared to the initial target speed nsoll of the hydraulic controller (10).

11. Method according to one of the preceding claims, characterized in that the hydraulic controller (10) regulates to a desired delivery flow or a desired delivery head or along a control curve as a controlled variable.

12. A multi-pump system comprising at least two hydraulically parallel-operated centrifugal pumps (Pu1 -Pu4), each driven by speed-controlled, electric motor drive units, wherein a hydraulic controller (10) is provided which outputs a synchronous speed setpoint nsoll for all centrifugal pumps (Pu1 -Pu4) as a manipulated variable as a function of a controlled variable, characterized in that an efficiency optimizer (20) is provided which is configured to determine pump-specific speed offset values ​​on the basis of the summed power consumption P_ges of at least two centrifugal pumps (Pu1 - Pu4) and to adapt the output speed setpoint nsoll of the hydraulic controller (10) for each pump (Pu1 -Pu4) individually by means of the pump-specific speed offset value such that the summed power consumption P_ges of the at least two pumps (Pu1 -Pu4) is reduced, in particular minimized.

13. Multi-pump system according to claim 12, characterized in that the at least two hydraulically parallel operated centrifugal pumps (Pu1 -Pu4) are pumps of the same design, but the hydraulic resistance between the pumps (Pu1 -Pu4) and the pressure accumulator (1) can be different.

14. Multi-pump system according to one of claims 12 or 13, characterized in that the multi-pump system is configured to carry out the method according to one of claims 2 to 11.