Method and system for controlling or regulating technical devices

Abstract

The control or regulation variable of the technical device (30) is influenced by at least one output variable (A), wherein the at least one output variable (A) is generated from at least one input variable (E). According to the application, a predicted future value (P1) of a variable (G) detected in the technical device (30) is used as input variable (E) in order to at least partially compensate for the time delay between the detection of the variable (G) and the generation of the output variable (A) from the input variable (E). In this way, the delay in the transmission and processing of the value of the input variable (E) can be compensated for and the dead time resulting therefrom in the control or regulation of the technical device (30) can be avoided.

Description

Technical Field

[0001] This invention relates to methods and systems for controlling or regulating technical devices. The invention also relates to input units and control or regulation devices. Background Technology

[0002] In the control and regulation system of a technical device (e.g., an instrument, apparatus, machine, facility), at least one control or regulation parameter of the technical device is affected by at least one output parameter of the control or regulation device.

[0003] Control or regulation parameters can be either discrete parameters (such as signals used to turn a pump on and off) or analog parameters, that is, parameters with continuous values ​​and time (such as temperature, pressure, level, stroke, angle, and speed).

[0004] The output parameters can be analog or binary control signals, which are applied to the controlled or regulated object (technical process, control segment) via actuators (such as relay switches, contactors, solenoid valves, or motors) and thereby change the control or regulation parameters. Actuators are also called "actuators".

[0005] Here, the control or regulation device generates at least one output parameter based on at least one input parameter.

[0006] Analog input parameters are, for example, measurement signals from sensors for temperature, pressure, level, stroke, angle, or rotational speed.

[0007] Binary input parameters are, for example, feedback signals from sensors in the control or regulation path regarding the state of control parameters, such as switch position (on / off), valve position (open / closed), or motor motion state (rotating / stationary).

[0008] For example, controlling the movement of a production line or drive.

[0009] For example, adjusting the flow rate of the medium through a pipe, the liquid level in a container, or the speed or torque of an electric motor.

[0010] Such control and regulation systems are used, for example, in process management systems. Process management systems are used to automate processes in technical facilities. Automated processes can be, for example, technological or manufacturing processes, or processes for generating or distributing electrical energy.

[0011] Process management systems are typically structured hierarchically through multiple levels (see, for example, EP 3 125053B1). At the lowest level (the so-called field level), the state of the technical process is detected by field instruments configured as sensors (e.g., pressure transducers, temperature sensors, level sensors, flow sensors) or the process is selectively influenced by field instruments configured as actuators (e.g., position controllers for control valves).

[0012] In addition, there is a control and / or regulation stage with control and / or regulation devices, in which a processing unit (CPU central processing unit) as part of a programmable logic controller implements field-related control and / or regulation functions as much as possible in real time, wherein they receive the values ​​of process parameters from sensors as input parameters and issue instructions to actuators as output parameters, for example.

[0013] Above this, in the process guidance stage, higher-level control and regulation are carried out in the control computer, where an operating system consisting of one or more operator stations enables facility operators to operate and observe the process.

[0014] Data exchange between field instruments and the processing unit of a programmable logic controller (PLC) is typically performed via a digital fieldbus (e.g., PROFIBUS DP or PROFINET). Because field instruments themselves usually do not have a corresponding fieldbus connection, they are connected to the digital fieldbus via distributed peripheral stations. These peripheral stations typically include an interface module (header assembly) for connecting to the digital fieldbus and multiple peripheral components (specifically digital and analog input / output units) for connecting field instruments. Each input or output unit can have only one so-called "channel" for connecting a single field instrument. However, the input or output unit can also have multiple channels for connecting multiple field instruments.

[0015] Typically, peripheral components are located directly on-site in field instruments, while the programmable logic controller (PLC) is situated at a more central point within the facility. Future-oriented control and regulation solutions even incorporate cloud-based control or regulation devices.

[0016] In a typical process, the processing unit periodically reads and processes the values ​​of the input parameters provided by the input units of the peripheral components, and generates the values ​​of the output parameters. These values ​​of the output parameters are then written into the output unit.

[0017] However, the problem here is that the "actual" value (apparently) of the input parameters provided by the input unit at the processing time point in the processing unit is outdated and no longer current. This delay acts as (additional) dead time during control or regulation. The reasons for this delay are, for example:

[0018] • The entire signal processing process in the input unit, such as analog-to-digital conversion and filtering,

[0019] • Transmission to the fieldbus controller via a partially slow fieldbus.

[0020] • Transmission from the fieldbus controller to the processing unit, and

[0021] • The resulting delay is that the values ​​of all or at least one set of input cells must be read in before processing can begin.

[0022] This problem is particularly prevalent in process management systems for facilities with large spatial distributions of process or manufacturing technologies, especially those that can extend to several square kilometers or more, such as those in the chemical, oil and gas, metal, mining, power plant, and transportation infrastructure (airports, tunnels).

[0023] To slightly compensate for the poor (re)control and regulation quality caused by outdated values, the processing cycle is shortened instead. However, more frequent operation places a significant load on the processing unit. Furthermore, higher transmission bandwidth is required.

[0024] Another approach is so-called clock synchronization, where detection, processing, and output are strictly clock-controlled. The delay between detection and processing is not eliminated, but it is deterministic. The disadvantages here are complex scheduling and poor variability during continuous operation. Furthermore, it typically requires specially designed and therefore expensive components. Summary of the Invention

[0025] Therefore, the objective of this invention is to reduce the impact of the aforementioned delay.

[0026] This invention is based on the understanding that the time delay between the detection of input parameters and the generation of output parameters from the input parameters, especially the delay during the transmission of input parameters, cannot be eliminated and can be reduced, if necessary, at a high cost. Therefore, in order for the processing unit to process the current "actual" value of the input parameters instead of the "old" value when generating output parameters based on the input parameters, it is necessary to provide the processing unit with a pre-predicted "future" value as the input parameter. If the predicted value is in the future at a time interval that varies within the order of the described time delay, particularly the transmission time, these time delays can be at least partially, and preferably completely, compensated for.

[0027] The parameters detected in the technical device can be continuously tracked together and, with the aid of algorithmic methods, can be appended to or replaced by the current value of the parameter to generate one or more predicted future values ​​of the parameter, and used as input parameters for control or regulation.

[0028] Therefore, according to the present invention, the predicted future value of the parameter detected in the technical device is used as an input parameter in order to at least partially compensate for the time delay between detecting the parameter and generating the output parameter from the input parameter.

[0029] Therefore, the predicted value is preferably in the future at a time interval, wherein the time interval is set according to the time delay between detecting the parameter and generating the output parameter from the input parameter.

[0030] Because the accuracy of prediction decreases as the time interval increases, the time interval is preferably equal to or less than the sum of these time delays. Advantageously, an optimal balance is struck between processing the most current value possible (the time interval should be as large as possible) and the accuracy of the prediction (the time interval should be as small as possible).

[0031] Predictions can be made using methods familiar to those skilled in the art (e.g., updating change curves, artificial intelligence, pattern recognition, etc.). For example, even with constant inflow or outflow, the liquid level in a tank with an upper cylindrical shape and a lower conical shape does not change continuously, but rather exhibits discontinuity at the bends in the tank. Even in the case of irregular inflow and / or outflow, bends have an effect and ultimately construct predictable patterns using pattern recognition.

[0032] The parameters of the technical device (which are detected and whose values ​​are derived) are preferably analog sensor signals (e.g., temperature sensors, pressure sensors, level sensors, or flow sensors). However, they can also be binary sensor signals, such as switch positions (on / off), valve positions (open / closed), or motor motion states (rotating / stationary).

[0033] According to a particularly advantageous design, the method of the present invention includes the following steps:

[0034] a) Parameters of the detection technology device,

[0035] b) Obtain the (current) value of the parameter.

[0036] c) Predict at least one future value of the parameter based on at least the derived (current) value of the parameter.

[0037] d) Use at least one predicted future value of the parameter as an input parameter for generating at least one output parameter.

[0038] Predicting the future value of a parameter can be achieved solely based on its current derived value. However, for better accuracy, it is preferable to make the prediction using or based on other values, such as:

[0039] • Use another, earlier derived value for the parameter,

[0040] • Use the previously predicted values ​​of the parameters.

[0041] • Current and / or earlier values ​​derived using parameters from other detection methods.

[0042] • Use values ​​that were predicted earlier using other parameters.

[0043] If at least one output parameter is generated in the control or regulation device based on at least one input parameter, then it is preferable to perform at least steps a) and b) outside the control or regulation device.

[0044] In principle, step c) can be performed at any suitable location on the transmission path between the detection and control or regulation device of the input parameter value, or even within the control or regulation device itself.

[0045] However, according to a particularly advantageous design, all steps a) through c) are performed outside the control or regulation device, particularly in distributed peripheral stations or in field instruments, and at least one predicted future value of the parameter is transmitted to the control or regulation device. Therefore, the predicted future value of the parameter used for external control or regulation of the device is supplied to the control or regulation device.

[0046] Alternatively, the execution of steps a) to c) can be distributed across multiple components. Thus, for example, steps a) to b) can be executed in a first component outside the control or regulation device, particularly in a distributed peripheral station or in a field instrument, and step c) can be executed in a second component arranged along the transmission path from the first component to the control or regulation device, particularly in a fieldbus controller.

[0047] According to another advantageous design, the value of the parameter is obtained at a first time point, and the at least one predicted future value relates to a second time point, which is in the future relative to the first time point at a defined or definable time interval.

[0048] This time interval is then preferably set to the transmission time to a control or regulating device based on either the derived (current) value of the parameter or a predicted future value of the parameter, which generates at least one output parameter based on at least one input parameter. This allows for a very simple balancing of the constant transmission time from the parameter detection component to the control or regulating device.

[0049] Preferably, the time interval is equal to or less than the transmission time. Here, an optimal balance can be struck between the accuracy of the prediction (the larger the time interval, the less accurate the prediction) and the handling of outdated values ​​of the input parameters (the larger the time interval, the more outdated the values ​​of the input parameters).

[0050] According to another advantageous design, the value of the parameter is obtained at a first time point and multiple predicted future values ​​are generated from it, the predicted future values ​​relating to multiple future time points, each of which is in the future relative to the first time point at different time intervals.

[0051] The (current) value of the parameter, the time point at which the value of the parameter was obtained, and all predicted future values ​​of the parameter can be output to a common dataset, for example, to a control or regulation device.

[0052] Then, based on the (current) value of the parameter or the predicted future value of the parameter, the transmission time to the control or regulation device can select one or more predicted future values ​​of the parameter as input parameters, and the control or regulation device generates at least one output parameter based on at least one input parameter.

[0053] For example, a control or adjustment device can determine the time point at which the data set is received, and select one of the predicted future values ​​of the parameter for control or adjustment based on the difference between the time point at which the data is received and the time point at which the (current) value of the parameter is derived.

[0054] This design is particularly well-suited for situations where the transmission time from the detected parameter to the control or regulation device is variable. Maximum flexibility and adaptability to varying transmission times can be achieved if the time intervals between predicted future values ​​are equidistant, the time intervals are parameterizable, the number of values ​​is parameterizable, and all these values ​​are timestamped.

[0055] A system for controlling or regulating a technical device according to the invention includes a control or regulation device with at least one output parameter, which can influence control or regulation parameters of the technical device. The control or regulation device is configured to generate at least one output parameter based on at least one input parameter. The input parameter is a predicted future value of a parameter detected in the technical device, in order to at least partially compensate for the time delay between detecting the parameter and generating the output parameter from the input parameter.

[0056] The input unit according to the present invention for generating input parameters of a control or adjustment device for a technical apparatus is configured to perform the following steps:

[0057] a) Detecting parameters of the technical device, particularly analog sensor signals.

[0058] b) Obtain the (current) value of the parameter.

[0059] c) Predict at least one future value of the parameter based on at least the derived value of the parameter.

[0060] d) Output the predicted future value of the parameter.

[0061] The design schemes and advantages of the method steps a) to c) described above in conjunction with the method according to the present invention can also be used in the input unit according to the present invention.

[0062] Preferably, the input unit has an interface to a communication device and is configured to output the predicted future value of the parameter to the communication device through the interface.

[0063] The distributed peripheral stations according to the present invention include the above-described input units.

[0064] Furthermore, the field instrument according to the present invention includes the aforementioned input unit.

[0065] In a control or regulation device for a technical apparatus according to the invention, the control or regulation device generates at least one output parameter based on at least one input parameter, the input parameter being a predicted future value of a parameter detected in the technical apparatus.

[0066] According to an advantageous design, the control or regulation device has an interface to a communication device and is configured to receive a predicted future value of the parameter from the communication device via the interface. Attached Figure Description

[0067] The invention, along with other advantageous designs of the invention according to features of various embodiments, will now be described in more detail with the aid of embodiments shown in the accompanying drawings; illustrated herein:

[0068] Figure 1 This shows a portion of a process management system with peripheral components.

[0069] Figure 2 This invention presents a first design scheme for an input unit according to the present invention, which serves as an analog input unit.

[0070] Figure 3 This illustrates a second design scheme for the input unit according to the present invention, serving as an analog / binary input unit.

[0071] Figure 4 This illustrates the principle-based method flow under the condition of constant transmission time via fieldbus.

[0072] Figure 5This illustrates a principle-based method flow in the case of variable transmission time via fieldbus. Detailed Implementation

[0073] Figure 1 A portion of a process management system 20 for a technical process or process facility 30 is shown in a schematic block diagram. It includes an operator station 1 in the process guidance level 2, a programmable logic controller (SPS) 3 in the control level 4, and field instruments 21 and 22. The field instruments detect the state of the technical process or process facility 30 by means of sensors 5 on the field level 17 or influence the process or process facility 30 in a targeted manner by means of actuators 6.

[0074] Higher-level control and regulation are carried out in process guidance level 2, wherein the operating system, consisting of operator station 1 and, if necessary, additional operator stations (not shown), enables operators of the facility to operate and observe the process or process facility 30. The technical process or process facility 30 is, for example, a facility of the chemical industry, oil and gas industry, metal industry, mining, or power plant.

[0075] Sensor 5 is, for example, a sensor used for pressure, temperature, liquid level, and flow rate.

[0076] Actuator 6 is, for example, a position regulator for regulating valves.

[0077] The operator station 1 and the programmable logic controller 3 are interconnected via facility bus 7 (e.g., Ethernet).

[0078] The programmable logic controller 3, in addition to other components, has a processing unit (CPU unit) 8, which processes the control program or application program and processes the value of the input parameter E and generates the value for the output parameter A based on it.

[0079] The programmable logic controller 3 performs field-related control and regulation functions as much as possible in real time. Here, the value of the input parameter E is received by the sensor 5 via the digital fieldbus 9 (e.g., PROFIBUS DP), and the value of the output parameter A is output to one or more actuators 6 via the digital fieldbus 9. To connect to the fieldbus 9, the controller 3 has a communication component 23, which also includes a controller 24 (“Fieldbus Controller”) for the fieldbus 9.

[0080] also, Figure 1 The intelligent field instrument 22 is shown, which is directly connected to the fieldbus 9. The field instrument 22 includes a sensor 5 and an input unit 12.

[0081] Because the field instrument 21 itself does not have a fieldbus connection, the field instrument is connected to the digital fieldbus 9 via distributed peripheral stations 10.

[0082] Therefore, controller 3 is connected to peripheral station 10 and field instrument 22 via digital fieldbus 9.

[0083] Peripheral station 10 includes an interface module (header assembly) 11 for connection to digital fieldbus 9 and a plurality of peripheral components. The peripheral components can be digital and analog input and output components or units. In this embodiment, particularly present as peripheral components are input unit 12, additional input unit 13, and output unit 14 according to the invention.

[0084] The peripheral components are implemented here as single-channel components for connecting the field instruments 21 (sensor 5 or actuator 6) respectively, but they can also be implemented as multi-channel, that is, they thus include multiple channels that can connect to the field instruments 21 (with sensor 5 or actuator 6) respectively.

[0085] For the sake of simplicity, only a few field instruments 21, 22, a single distributed peripheral station 10, a single controller 3, and a single operator station 1 are shown. However, in practice, facilities typically include multiple such components.

[0086] The input units 12 and 13 detect signals from the sensor 5 (e.g., measuring current, measuring voltage) and derive the values ​​of these parameters. If these values ​​are now transmitted to the controller 3 via the fieldbus 9 and used as input parameters E in the processing unit 8, the (apparent) "actual" value of the input parameter E provided by the input units 12 and 13 at the point in time of processing in the processing unit 8 may be outdated and no longer current. This delay acts as (additional) dead time during control or regulation. The reasons for this delay are, for example, as follows:

[0087] • The entire signal processing process in input units 12 and 13, such as analog-to-digital conversion and filtering.

[0088] • Transmission from input unit 12 to interface module 11 in the internal backplane bus of peripheral station 10,

[0089] • The transmission from controller 3 to fieldbus controller 24 via a partially slow fieldbus 9.

[0090] • Transmission from fieldbus controller 24 to processing unit 8, and

[0091] The resulting delay is that all or at least one set of input units 12 and 13 are read in before processing is performed in processing unit 8.

[0092] This problem is particularly evident in large-scale process management systems 20, in facilities where process technology or manufacturing technology is widely distributed in space, especially in facilities that extend over several square kilometers, such as chemical industry, oil and gas industry, metal industry, mine, power plant, transportation infrastructure (airport, tunnel), etc.

[0093] Therefore, according to the present invention, a pre-predicted "future" value is provided to the processing unit 8 as an input parameter E. Thus, the processing unit 8 is able to actually process the current "actual" value of the input parameter E, rather than the "old" value. This compensates for the aforementioned delay. Parameters detected in the technical device (e.g., measured current, measured voltage) can be continuously tracked together, and one or more predicted future values ​​of the parameter can be generated by appending or replacing the current value of the parameter using an algorithmic method and used as input parameter E in the controller 3.

[0094] Here, the prediction of future values ​​can, in principle, be performed at any point along the transmission path from the location where the parameter value of the deriving facility 30 is obtained to the processing unit 8. In the case of the peripheral station 10, this can be performed, for example, in the input units 12, 13, at other locations in the distributed peripheral stations 10, or in the controller 3, for example, in its communication component 23, especially in the fieldbus controller 24. In the case of the intelligent field instrument 22, this can be performed, for example, in the input unit 12 or in the controller 3, for example, in its communication structure group 23, especially in the fieldbus controller 24.

[0095] In this embodiment, in the case of input unit 12, the prediction of future values ​​is performed within the input unit itself. That is, the predicted future values ​​are then transmitted to controller 3 via fieldbus 9 and used as input parameter E by processing unit 8.

[0096] In the case of input unit 13, the prediction of future values ​​is instead performed in fieldbus controller 24. That is, the (current) value of a parameter (e.g., measured current, measured voltage) is transmitted from input unit 13 to controller 3 and then the predicted future value of that parameter is generated by fieldbus controller 24, which is then used by processing unit 8 as an input parameter.

[0097] Now according to Figure 2The principle structure of input unit 12 is illustrated by way of example, which generates and outputs predicted future values. Input unit 12 is configured here as an analog input unit. This analog input unit is connected to sensor 5 (here, a measuring current generator) via line 52 through its analog input terminal 51. The measuring current generator is, for example, a measuring converter that generates a measuring current I between 4mA and 20mA proportional to the detected measuring parameter G. The measuring parameter G is a parameter of facility 30, such as temperature, pressure, flow rate, etc.

[0098] The input unit 12 includes a voltage measuring device 54, which includes a differential amplifier 55 and a post-amplified analog-to-digital converter 56. The analog-to-digital converter has its voltage measurement input 57 directly connected to the analog input 51, that is, the analog input 51 is the voltage measurement input 57.

[0099] The current measuring resistor 61 is constructed as a precision resistor and has a resistance value of, for example, 250Ω. The voltage measuring device 54 detects the voltage drop generated across the current measuring resistor 61 by the measuring current I and converts this voltage drop into a current digital measurement value 62 for the detected measurement parameter G.

[0100] The current value 62 of the measurement parameter G is received by the control unit 63 and transmitted to the prediction unit 64. The prediction unit 64 predicts at least one future value 65 of the measurement parameter G based on the current value 62 and returns the predicted future value 65 to the control unit 63. Prediction can be achieved, for example, by updating change curves, artificial intelligence, pattern recognition, etc. For example, possible change curves of the measurement parameter G can be derived from the last change curve (e.g., by averaging). Change curves of other measurement parameters can also be used to estimate future values ​​of the measurement parameter G. In the simplest case, typical change curves can be pre-designed, i.e., stored, for example, in the prediction unit 64.

[0101] The control unit 63 outputs the predicted future value 65 to the upper-level controller 3 via the communication interface 66.

[0102] When according to Figure 1 When the input unit 12 is used in the distributed peripheral station 10, the communication interface 66 is configured as a connection to the internal bus (e.g., backplane bus) of the peripheral station 10. The actual interface to the fieldbus 9 is then provided through the interface module 11.

[0103] In the case of field instrument 22, communication interface 66 is already the interface to fieldbus 9.

[0104] Figure 3The input unit 12' shown can be selectively operated as an analog input unit for measuring current or as a binary input unit for measuring voltage. (This is in contrast to the previous description.) Figure 2 Unlike input unit 12, a switching device 58 is additionally present here. Switching device 58 includes a parallel circuit consisting of a high-ohm resistor 59 and a controllable switch 60, which is connected in series with a low-ohm current-measuring resistor 61 at analog input terminal 51 or voltage-measuring input terminal 57. Analog input terminal 51 is thus switchable between a high-ohm state and a low-ohm state by means of the controllable switching device 58.

[0105] The high-ohm resistor 59 has a resistance value of, for example, 1 MΩ. When switch 60 is closed, i.e., in the low-ohm state of analog input 51, voltage measuring device 54 detects the voltage drop generated by the measuring current I across current measuring resistor 61 and sets this voltage drop to the current digital measurement value 62 for the detected measurement parameter.

[0106] During operation as a binary input unit, the controllable switch 60 is turned off, thereby forming a voltage divider between the high-ohm resistor 59 and the low-ohm current measuring resistor 61, and the voltage measuring device measures the high or low level of the voltage applied to the input terminal of the corresponding binary input unit.

[0107] During operation as an analog input unit for measuring current, the controllable switch 60 is closed, rendering the high-ohm resistor 59 ineffective, and the measuring current flows through the low-ohm current measuring resistor 61.

[0108] The future value 65 of parameter G, predicted by input units 12 and 12', is transmitted to controller 3 and used in processing unit 8 as input parameter E for generating output parameter A.

[0109] Now, Figure 4 and Figure 5 It should be explained how the predicted future value of parameter G is obtained and then used when controlling or adjusting the facility 30.

[0110] exist Figure 4 In this case, it is assumed that the transmission time T from the input unit 12 of the distributed peripheral station 10 or the field instrument 22 to the processing unit 8 is constant.

[0111] The measurement parameter G is detected at the input unit 12, and its current value M1 is obtained at a first time point t1. In the input unit 12, the predicted future value P1 of the measurement parameter at a second time point t2 is derived from the current value M1. This second time point is in the future relative to the first time point t1 at a defined time interval tx. Therefore, calculations are performed from x into the future.

[0112] Herein, M is used to represent the actual signal change curve and P is used to represent the predicted signal change curve.

[0113] The predicted future value P1 is transmitted to the controller 3 via the field bus 9. In the processing unit 8, instead of the measured value M1, the predicted future value P1 is used as an input parameter for control or regulation. Here, te represents the time point at which the value P1 is received. Me shows the value that can be measured at the receiving time point if there is no delay; this value can be used as a reference for deviation.

[0114] The time interval tx is set according to the transmission time T from the input unit 12 to the controller 3, and herein the time interval is preferably less than or equal to the transmission time T.

[0115] When tx=0, the entire system behaves as if no predicted future value is applied.

[0116] When tx=T, the processing unit 8 of the controller 3 provides a temporally correct predicted value. However, this prediction has fuzziness according to the dynamics of the input signal related to the transmission time to be bridged by prediction.

[0117] Advantageously, as shown in Figure 4 , an optimization is made between processing an older (at that time) more applicable value and a more real-time value that may deviate more strongly accordingly. Therefore, 0<tx<T is set. Therefore, it is preferable to set an optimization between the prediction accuracy (the larger the time interval, the less accurate the prediction) and the processing of the outdated value of the input parameter (the larger the time interval, the more outdated the value of the input parameter is).

[0118] In the case of Figure 5 , it is assumed that the transmission time T from the input unit 12 of the distributed peripheral station 10 or the field instrument 22 to the processing unit 8 is variable.

[0119] Herein, also on the side of the input unit 12, the current value M1 of the measurement parameter G is obtained at the first time point t1. From the current value M1 of the measurement parameter G, a plurality of predicted future values P1, P2, P3 of the measurement parameter are generated in the input unit 12, these future values relate to a plurality of future time points t2, t3, t4, which are in the future relative to the first time point t1 at different time intervals (herein at different multiples of td).

[0120] Then, the obtained current value M1 of the parameter, the time point t1 at which the value of the parameter is obtained, and all the predicted future values P1, P2, P3 of the parameter can be output to the field bus 9 in a common data set D and thereby output to the processing unit 8 in the controller 3.

[0121] Then, based on the transmission time T of the current value M1 obtained from the parameter or the predicted future values ​​P1, P2, P3 of the parameter to the processing unit 8 in the controller 3, the processing unit 8 can select one or more of the predicted future values ​​P1, P2, P3 as input parameters.

[0122] For example, processing unit 8 can determine the time point te of receiving dataset D, and select one of the predicted future values ​​of the parameter as an input parameter E for controlling or adjusting facility 30 based on the difference between the received time point te and the time point t1 at which the value of the parameter is obtained. Me again shows the value that can be measured at the receiving time point without delay, which is scanned by a "clock" due to the time interval of the predicted future values ​​P1, P2, P3.

[0123] If (as in this case) the time interval td of the predicted future values ​​is equidistant, the time interval is parameterizable, the number of values ​​is parameterizable, and all these values ​​are timestamped, then maximum flexibility and adaptability to varying transmission times can be achieved. However, a timestamp does not mean that each value must give a complete time. In the case of equidistant and known time intervals, the time of the current value (from the perspective of input unit 12) satisfies the future values ​​thereafter. The first timestamp can also be a pure count value, which is incremented by a system-wide clock signal in both input unit 12 and processing unit 8.

[0124] Therefore, the accuracy of control or regulation performed by means of controller 3 can be improved using the present invention. The accompanying reduction in the workload of processing unit 8 lowers its cost and power consumption (CO2). Here, existing fieldbuses, such as PROFINET-I / O or Profibus-DP, can be used: in Figure 4 Under these conditions, the amount of data can remain unchanged. Figure 5 In such cases, these values ​​can be transmitted using consistent data usage. The disadvantages of cost-effective, but slow, transmission paths can be partially or completely compensated for through prediction.

[0125] In particular, in cloud-based automation (i.e., cloud-based control or regulation devices), it is able to compensate for large fluctuations in transmission time.

Claims

1. A method for controlling or regulating a technical device (30), wherein, The control or adjustment parameters of the technical device (30) are influenced by at least one output parameter (A), wherein at least one output parameter (A) is generated based on at least one input parameter (E), characterized in that a predicted future value (P1) of a parameter (G) detected in the technical device is used as an input parameter (E) to at least partially compensate for the time delay between detecting the parameter (G) and generating the output parameter (A) from the input parameter (E), wherein the value of the parameter (G) is detected at a first time point (t1). (M1), and at least one predicted future value (P1) is associated with a second time point (t2), which is in the future relative to the first time point (t1) at a defined or definable time interval (tx), wherein the time interval (tx) is less than the transmission time (T) of the derived value (M1) of the parameter (G) or the predicted future value (P1) of the parameter (G) to the control or regulation device (3), which generates at least one output parameter (A) based on at least one input parameter (E).

2. The method according to claim 1, comprising the following steps: a) Detect the parameter (G) of the technical device (30). b) Obtain the value (M1) of the parameter (G). c) Based on at least one derived value (M1) of the parameter (G), predict at least one future value (P1) of the parameter. d) Use at least one predicted future value (P1) of the parameter (G) as an input parameter (E) for generating at least one of the output parameters (A).

3. The method according to claim 1, wherein, The parameter (G) of the technical device (30) described in step a) is an analog sensor signal.

4. The method according to claim 2, wherein, At least one output parameter (A) is generated in the control or regulation device (3) according to at least one of the input parameters (E), wherein steps a) and b) are performed at least outside the control or regulation device (3).

5. The method according to claim 4, wherein, All steps a) to c) are performed outside the control or regulation device (3), and wherein at least one predicted future value (P1) of the parameter (G) is transmitted to the control or regulation device (3).

6. The method according to claim 5, wherein, Perform all steps a) to c) at the distributed peripheral station (10) or at the field instrument (22).

7. The method according to claim 4, wherein, Steps a) to b) are performed in a first component outside the control or regulation device (3), and step c) is performed in a second component arranged along a transmission path from the first component to the control or regulation device (3).

8. The method according to claim 7, wherein, Steps a) to b) are performed in a distributed peripheral station (10) or in a field instrument (22), and step c) is performed in a second component arranged in a controller (24) of a fieldbus (9).

9. The method according to any one of claims 2 to 8, wherein, The value (M1) of the parameter (G) is obtained at a first time point (t1), and therein, a plurality of predicted future values ​​(P1, P2, P3) are generated, which relate to a plurality of future time points (t2, t3, t4) that are in the future relative to the first time point (t1) at different time intervals (td).

10. The method according to claim 9, wherein, Based on the value (M1) obtained from the parameter (G) or the predicted future value (P1, P2, P3) of the parameter (G), the transmission time (T) to the control or regulation device (3) selects one or more future values ​​of the predicted future value (P1, P2, P3) of the parameter (G) as input parameters (E), and the control or regulation device generates at least one output parameter (A) based on at least one of the input parameters (E).

11. A system (20) for controlling or regulating a technical device (30), the system having a control or regulation device (3) with at least one output parameter (A), the output parameter being capable of influencing the control or regulation parameters of the technical device (30), wherein, The control or regulation device (3) is configured to generate at least one output parameter (A) based on at least one input parameter (E), characterized in that the input parameter (E) is a predicted future value (P1) of a parameter (G) detected in the technical device (30) in order to at least partially compensate for the time delay between the detection of the parameter (G) and the generation of the output parameter (A) by the input parameter (E), the value (M1) of the parameter (G) is detected at a first time point (t1), and at least one predicted future value (P1) is associated with a second time point (t2), the second time point being in the future relative to the first time point (t1) by a defined or definable time interval (tx), wherein the time interval is less than the transmission time (T) of the obtained value (M1) of the parameter (G) or the predicted future value (P1) of the parameter (G) to the control or regulation device (3).

12. An input unit (12) for generating input parameters (E) for a control or adjustment device (3) of a technical device (30), characterized in that, The input unit is configured to perform the following steps: a) Detect the parameter (G) of the technical device (30). b) Obtain the value (M1) of the parameter (G). c) Based on at least one derived value (M1) of the parameter (G), predict at least one future value (P1) of the parameter (G). d) Output the predicted future value (P1) of the parameter (G). In this process, the value (M1) of the parameter (G) is detected at a first time point (t1), and at least one predicted future value (P1) is associated with a second time point (t2), which is in the future relative to the first time point (t1) at a defined or definable time interval (tx), wherein the time interval (tx) is less than the transmission time (T) of the obtained value (M1) of the parameter (G) or the predicted future value (P1) of the parameter (G) to the control or regulation device (3), which generates at least one output parameter (A) based on at least one of the input parameters (E).

13. The input unit according to claim 12, wherein, The parameter (G) of the technical device (30) described in step a) is an analog sensor signal.

14. A distributed peripheral station (10) having an input unit (12) according to claim 12.

15. A field instrument (22) having an input unit (12) according to claim 12.

16. A control or regulation device (3) for a technical apparatus (30), said control or regulation device being configured such that said control or regulation device generates at least one output parameter (A) according to at least one input parameter (E), characterized in that, The input parameter (E) is a predicted future value (P1) of a parameter (G) detected in the technical device (30), wherein the value (M1) of the parameter (G) is detected at a first time point (t1), and at least one predicted future value (P1) is associated with a second time point (t2), which is in the future relative to the first time point (t1) at a defined or definable time interval (tx), wherein the time interval (tx) is less than the transmission time (T) of the obtained value (M1) of the parameter (G) or the predicted future value (P1) of the parameter (G) to the control or regulation device (3).

Images