Control device and procedure for improving the dynamics of a control system

Abstract

The invention relates to a control device (1) comprising a controller (2) for determining a manipulated variable (3) from a reference variable (4) and a measured controlled variable (5), a controlled system (7) containing the controlled variable (6) on which the controller (2) acts via the manipulated variable (3), and a measuring device (8) for providing the measured controlled variable (5) from the controlled variable (6) contained in the controlled system (7). The control device (1) includes a delay element (9) for the manipulated variable (3) and a subtractor (10) which, by comparing the time-delayed manipulated variable (3) with the measured controlled variable (5), provides a control error (11), which is multiplied by a compensation factor (12) and added to the measured controlled variable (5) before the latter is supplied to the controller (2). The invention further relates to a method for improving the dynamics of a control system.

Description

The invention relates to a regulating device and a method for improving the dynamics of a regulation.Cycle time and measuring transducer speed play a decisive role in the regulation of dynamic processes. Implementing computations within an appropriate cycle in the controller is technically straightforward. Typically, transmitters are available at speeds on the order of about 100 ms. However, in highly dynamic methods, these speeds prove to be inadequate. Only limited alternatives for detecting actual values from an electrical network are present in the controller. When a highly dynamic machine is involved, it can easily oscillate and lead to a destabilization of the entire system.An exemplary system includes a combination of diesel generators and battery systems in which the batteries are coupled to the supply grids via converters and have a control system that operates in the microsecond range.This highly dynamic battery system response speed far exceeds that of the actual measurements of a conventional controller. When using conventional methods, the regulation of the converters is adapted to the performance of the diesel generators and the transmission by the measurement converters used.This then has the consequence that the advantage of the highly dynamic converter technique is not utilized.It is known from island networks of the marine system that calculations are indeed processed in the fast control cycle; with the slow actual values, the regulation was made so slow that extreme overshoots could not occur. The dynamics of the aggregate to be regulated were thus artificially slowed down.It is an object of the invention to provide an improved control device. A further object of the invention is to specify a corresponding method for improving the dynamics of a regulation.The invention achieves the object directed to a control device in that, in the case of such a control device, a controller for determining a manipulated variable from a reference variable and a measured controlled variable, a controlled system which contains the controlled variable and on which the controller acts via the manipulated variable, and a measuring device for providing the measured controlled variable from the controlled variable contained in the controlled system, a delay element for the manipulated variable and a subtractor are provided, wherein the subtractor provides a controlled difference by comparing the temporally delayed manipulated variable with the measured controlled variable, which controlled difference is multiplied by a compensation factor and added to the measured controlled variable before said controlled variable is fed to the controller.The control device according to the invention compensates for the missing dynamics in the measurement. A dead time in the measurement leads to delays in the detection of changes in the actual value. A controller that solves the dead time problem can detect the actual value more quickly and accurately. This allows the control system to respond more quickly to changes and reduce the control deviation. Dead times may also result in overshoot when the control system responds to changes in the actual value. An effective solution to the dead time problem helps minimize this overshoot. This also results in the stability of the control loop being able to be improved. The control system can take the required compensation measures to minimize the effects of dead time and ensure stable performance. Overall, the invention contributes to better control.In an advantageous embodiment of the invention, the compensation factor has at least one constant portion. This constant portion provides stability in the control. If the compensation factor consists only of variable or adaptive components, this can lead to unstable control systems, in particular if unexpected or non-linear effects occur. A constant fraction acts as a base or reference point to ensure the stability of the control system. A constant proportion in the correction factor can be based on prior knowledge or expert knowledge. Corresponding indications are also found, for example, in the data sheet of the measurement transmitter used in the regulation. A constant proportion in the correction factor can also simplify the implementation of the control system. In contrast to adaptive or variable correction factors, a constant portion does not require a continuous estimation or calculation.It is expedient here if the constant component can be stored in the regulating device. The advantages of keeping information over a period of time and being accessible for future uses are obvious. By storing values, data can be retrieved and used at a later time. This makes it possible to analyze information, recognize trends, identify patterns, and make decisions based on past data. This may be helpful in error correction. Further, by storing values, the continuity of the data is maintained even when a system or application is restarted or updated.It is furthermore expedient if the guide variable is feedable and adjustable from the outside. As a result, a control can be flexibly adapted to different requirements and operating conditions. This makes it possible to adapt the behaviour of the system according to the current requirements. Furthermore, the external adjustment of the command variable allows the system to be optimized and its performance to be improved. By varying the command variable, different operating states can be tested and analyzed in order to find the best control approach. Furthermore, in many applications, it may happen that the system is exposed to disturbances or undesired influences. By setting the reference variable from the outside, these disturbances can be detected and compensated. The control can be adjusted accordingly to minimize the effects of the disturbances and keep the system on course. An adjustable command also enables the system to be quickly adjusted to changed requirements or goals. For example, if operating conditions change or new targets are established, the command may be adjusted to achieve the desired results. Finally, in some applications, it may be desirable for human operators or users to have the ability to adjust the command size. This allows direct interaction between the human and the system and allows intuitive control.It is advantageous if the compensation factor results from a product of the constant component and a nonlinear component. This regularly leads to a better control performance compared to purely linear controls. The constant portion can serve as a baseline and ensure stable control, while the nonlinear portion enables fine tuning and compensation of nonlinear effects. This can result in more precise, faster and robust control.It is furthermore advantageous if the nonlinear component has a dependence on a change in the manipulated variable over time. The compensation factor can thereby adapt to changing operating conditions or system dynamics. This allows the control system to adjust for changes and to provide continuous and optimum control. The non-linear portion may be adjusted in real time to account for current system conditions and improve control accordingly.The object directed to a method is achieved by a method for improving the dynamics of a control, in which a controller determines a manipulated variable from a reference variable and a measured controlled variable and acts on a controlled system with this manipulated variable, and a controlled variable of the controlled system is measured by means of a measuring device in order to obtain a measured controlled variable, wherein, according to the invention, the manipulated variable is determined with a time delay, a controlled difference between the temporally delayed manipulated variable and the measured controlled variable is determined, multiplied by a compensation factor and added to the measured controlled variable before the measured controlled variable is fed to the controller.A constant portion of the compensation factor is advantageously determined. It is expedient here if the constant component of the compensation factor determined is stored.It is furthermore advantageous if the reference variable is supplied from outside the controller and can be adjusted.Finally, it is advantageous if a non-linear portion of the compensation factor is determined as a function of a change in the manipulated variable over time.Systems to be controlled have sometimes shown a better behavior than the measured actual value, for example due to a time delay in the measurement value acquisition. In such cases, the actual actual value is closer to the setpoint value than the measured one, and a part of the detected control difference is therefore already corrected. With a corresponding selection of the compensation factors, the control dynamics (compared to the uncompensated system) can be increased with the aid of the invention without oscillations occurring.The invention uses an algorithm which is built into the controlled system and reduces the delay of the actual value by an "acceptance" and thereby reduces the control difference. The known dynamics of the unit are thereby calculated into the actual value and the control can thus be set correspondingly aggressively. If, contrary to expectation, the aggregate to be regulated does not follow, this can be included again in the regulation path with a control algorithm.The compensation of the delay makes it possible to apply the controlled systems also for dynamic controlled systems, so that this dynamic is not decelerated by the slowest component in the controlled system. Up to now, the control has been optimized according to these slowest components.The invention is explained in more detail by way of example with reference to the drawings. They show schematically and not to scale: FIG. 1 shows a control device according to the prior art, FIG. 2 shows a control device according to the invention, and FIG. 3 shows a further control device according to the invention.FIG. 1 shows schematically and by way of example a control device 1 according to the prior art. The control device 1 comprises a controller 2 for determining a control variable 3 from a reference variable 4 that can be supplied and adjusted from the outside by the operator or another technical system and a measured control variable 5. The control device 1 further comprises a controlled system 7 containing the controlled variable 6, on which the controller 2 acts via the manipulated variable 3, and a measuring device 8 for providing the measured controlled variable 5 from the controlled variable 6 contained in the controlled system 7. If the reference variable 4 does not change over time, the terms "setpoint value" and "reference variable" are used interchangeably. The control principle is the reference actual value comparison of the reference variable 4 with the negatively fed back measured control variable 5. Delays result in the measured or determined control variable 5, i.e. the measured actual value, by above all non-ideal measurement value acquisition. In particular in the case of routes with integrative behavior ("memory"), there is a high risk of oscillations. In order to suppress such oscillations, the control must be greatly attenuated, i.e., the dynamics must be reduced.The measuring device 8 causes a delay in the signal processing. This may result in the control becoming unstable. In particular in fast control loops, the delay can lead to undesired oscillations or even to a system failure. The delay in signal processing can reduce control accuracy. The control may not react fast enough to changes in the controlled variable or may not track it precisely enough. If the signals are processed in a delayed manner, the system may be more sensitive to interference or noise. The delay may result in the system not responding fast enough to interference or in interference being amplified. The delay in signal processing can impair the overall performance of the control. The system may not achieve the desired performance or efficiency.The invention provides a suitable control device 1 or a corresponding method for improving the dynamics of a control in order to minimize these possible consequences. FIG. 2 shows an embodiment of a control device 1 according to the invention. In comparison with the example of FIG. 1 according to the prior art, the control device 1 according to the invention further comprises a delay element 9 for the manipulated variable 3, a subtractor 10 which, by comparing the temporally delayed manipulated variable 3 with the measured controlled variable 5, provides a control difference 11 which is multiplied by a compensation factor 12 and added to the measured controlled variable 5 before the measured controlled variable 5 is fed to the controller 2. The compensation factor 12 has a constant portion 13 and is typically storable in the control device 1.FIG. 3 shows a further exemplary embodiment of an inventive control device 1. in contrast to the exemplary embodiment of FIG. 2, the compensation factor 12 results from a product of the constant component 13 and a nonlinear component 14, wherein the nonlinear component 14 has a dependence on a change 15 of the manipulated variable 3 over time.

Claims

Regulating device (1) having a regulator (2) for determining a manipulated variable (3) from a reference variable (4) and a measured controlled variable (5), a controlled system (7) which contains the controlled variable (6) and on which the regulator (2) acts via the manipulated variable (3), and a measuring device (8) for providing the measured controlled variable (5) from the controlled variable (6) contained in the controlled system (7), characterized bya delay element (9) for the manipulated variable (3), a subtractor (10) which, by comparing the temporally delayed manipulated variable (3) with the measured controlled variable (5), provides a controlled difference (11) which is multiplied by a compensation factor (12) and added to the measured controlled variable (5) before the latter is fed to the regulator (2).Regulating device (1) according to Claim 1, wherein the compensation factor (12) has at least one constant portion (13).Control device (1) according to Claim 2, wherein the constant proportion (13) can be stored in the control device (1).Control device (1) according to one of claims 2 or 3, wherein the reference variable (4) is feedable and adjustable from the outside.Control device (1) according to Claim 2, wherein the compensation factor (12) results from a product of the constant component (13) and a nonlinear component (14).Control device (1) according to Claim 3, wherein the non-linear component (14) has a dependence on a change (15) in the manipulated variable (3) over time.Method for improving the dynamics of a control, in which a controller (2) determines a manipulated variable (3) from a reference variable (4) and a measured controlled variable (5) and acts on a controlled system (7) with this manipulated variable (3), and a controlled variable (6) of the controlled system (7) is measured by means of a measuring device (8) in order to obtain a measured controlled variable (5), characterized in that the manipulated variable (3) is determined with a time delay, a controlled difference (11) between the temporally delayed manipulated variable (3) and the measured controlled variable (5) is determined, multiplied by a compensation factor (12) and added to the measured controlled variable (5) before it is fed to the controller (2).Method according to Claim 7, wherein a constant proportion (13) of the compensation factor (12) is determined.Method according to Claim 8, wherein the constant component (13) of the compensation factor (12) determined is stored.Method according to one of Claims 7 to 9, wherein the reference variable (4) is supplied and adjustable from outside the controller (2).Method according to one of Claims 7 to 10, wherein a nonlinear portion (14) of the compensation factor (12) is determined as a function of a change (15) in the manipulated variable (3) over time.

Images