Diagnostic method and device for monitoring the functioning of a control loop

0001The invention relates to a diagnostic device for monitoring the operation of a control loop of an automation system according to the preamble of claim 1, a corresponding diagnostic method according to the preamble of claim 6 and a function block for its execution by a processor of an automation device according to the preamble of claim.

0002Maintenance and servicing of automation equipment can be improved if the correct functioning of units or components is monitored. If functionality fails, action can be taken at the right point in the system with maintenance, repair or fault clearance measures.

0003From DE 100 07 972 A1 a diagnostic tool for use in a process control system is known which detects problems or identifies poorly functioning devices and circuits. A process control system is described that includes a centrally located process controller communicatively coupled to at least one operator workstation and one or more plant devices via analog, digital, or combined analog digital bus lines. The plant equipment, which includes, for example, valves, valve positioning, switches and transducers (z. B. Temperature, pressure or flow transmitters) perform functions within the process, such. B. Opening or closing valves or measuring process parameters. The process controller receives signals indicating the process measurements obtained by the equipment and / or other information associated with the equipment. It uses these signals to implement a control routine and generates control signals which are sent to the abutments to control the flow of the process. Information from the plant devices and the controller is made available to one or more application programs that are executed by the operator workstation to enable an operator to perform a desired function on the process, such as viewing the current process state, Modify the operation of the process, etc. For better structuring of the control task, function blocks are used, each receiving inputs from other function blocks and / or outputting output signals to other function blocks. They each perform a specific part of the process control operation, for example, measuring or detecting a process parameter, controlling a device or the task of a PID controller. The well-known diagnostic tool automatically collects and stores data associated with the various functional blocks of devices and circuits within the process control system, and processes that data to determine which components of the process control system have problems that may result in reduced system performance. For problem detection, the diagnostic tool uses a scatter indication based on, for example, the differences between a historical statistical state of a device over a predetermined period of time, such as the average position of the valve of a valve device and the current state of the device, in this example the current one Position of the valve, can be calculated. As another example, for an application to a control function block, a scatter indication is described as a stochastic feature based on a deviation between a process parameter input of the function block and a setpoint indicated to the function block for that parameter. A disadvantage of the known diagnostic tool with evaluation of a stochastic feature of process parameters is that a statically stationary process is required. It is not possible to diagnose the transient behavior, for example during compensation processes. For here each setpoint step of a controller violates the requirements of a stationary process to a considerable extent and leads to erroneous statements in the evaluation of such stochastic features.

0004From US 6 192 321 B1 a diagnostic method for a process controller is known, which controls the sequence of a process in a control loop. During process operation, the diagnostic device monitors whether the process is in a steady state. If this is the case, a control signal of a controller is replaced by a test signal by the diagnostic device, which is thus supplied to the actuator instead of the control signal. The test signal stimulates the actuator to perform a deterministic sequence of operations designed to minimize impact on the process. The diagnostic device detects the response signals that result in the process after excitation by the test signal. By suitable choice of the test signal, the width of a dead zone, a response time, a gain or an overshoot can be calculated by analyzing the response signals. However, this diagnostic method has the disadvantage that the normal process of the process is disturbed, although this disorder is small or only of short duration. In addition, not the entire control loop is monitored by the diagnosis, but only the part to which the control signal of the controller is given. Another disadvantage is the fact that no diagnosis of a control loop in the stationary state is feasible without disturbing the process flow in any way.

0005The invention has for its object to provide a diagnostic device and a diagnostic method with which the operation of control circuits in a process engineering system can be monitored almost permanently and automatically.

0006To solve this problem, the new diagnostic device of the type mentioned in the characterizing part of claim 1 features. In the dependent claims further developments of the invention, in claim 6, a diagnostic method and in claim 7, a function block for performing the method described.

0007The invention has the advantage that it allows quasi-permanent monitoring of control loops and all involved in the respective control loop components of a process plant during plant operation, ie both in the steady state and during balancing operations. It can therefore be intervened faster with loss of performance of individual control loops in the right place of the system with troubleshooting or improvement measures. Advantageously, in addition to the diagnosis of a control loop in the stationary state, a diagnosis of the transient response, for example, at setpoint jumps, possible. Misdiagnoses that could occur in previous diagnostic facilities with evaluation of the variance of different process parameters in the case of setpoint changes are also largely avoided. Another advantage is the fact that in the sense of a non-invasive diagnosis, only the data resulting from regular process operation are used. The process is thus in no way influenced by the diagnosis. In this case, the control loop in an automation system represents a functional level on which imminent errors can be detected particularly quickly and reliably. The control loop has a clearly defined task and a defined behavior after commissioning. Deviations from this nominal behavior can be recognized by a signal processing without additional sensors and usually allow a localization of sources of error within this control loop except for the individual components transmitter, controller, actuator and controlled system.

0008In the case of a substantially stationary state, it is advantageous to determine the variance of the sequence of actual value data as a stochastic feature and to compare it with an associated reference value, which was predetermined and stored on the same control loop in a well-working state. This leads to an easy detectability of faults in system components that are in the control loop. In addition, a better informative value of the diagnosis is achieved since a reference value measured on the real control loop is used for comparison.

0009If, in the case of an excitation of the control loop essentially according to a desired value step, in particular for an exact setpoint step, the relative overshoot and / or the transient response, ie the quotient of rise time and transient time, are determined as deterministic features, this has the advantage that both Features allow very direct and diagnostically well evaluable conclusions about the operation of the control loop and the control quality. Both features are dimensionless and independent of the particular application in which the control loop is operated. For predetermining a reference value with which a value of the respective feature determined during a diagnosis during normal operation is therefore not required, application-specific calculations or empirical measurements are required in advance.

0010Advantageously, a variance increase due to decay processes of the variables of the control loop is excluded after prior excitation by a setpoint change as the cause of a misdiagnosis, if the evaluation device is first reactivated by the selection device to determine a stochastic feature, if, after the start of the setpoint jump, a prescribable multiple of the settling time of the control loop has elapsed at substantially constant setpoint data. Thereafter, the variance of the process variable again corresponds to that which prevails in the steady state.

0011The diagnostic device can be advantageously designed as a software function block that can be interconnected in a graphical user interface of an engineering system with function blocks of the control loop and for operating the diagnostic device in an automation device. This function block can, for example, be implemented as a PCS7 function block for SIMATIC controllers from Siemens AG and provided with a so-called faceplate for realizing a human-machine interface on an operating and monitoring device of the automation system.

0012With reference to the drawings, in which an embodiment of the invention is shown, the invention and refinements and advantages are explained in more detail below.

0013Show it:<dl id="dl0001"><dt>FIG. 1</dt><dd>a block diagram of an automation system,</dd><dt>FIG. 2</dt><dd>a schematic diagram of a diagnostic function block and</dd><dt>FIG. 3</dt><dd>a time diagram of a step response to clarify deterministic features.</dd></dl>

0014In Figure 1, the basic structure of an automation system is shown. At a process 1 process variables are measured and read in by an analog input 2. To control the process 1, an analog output 3 outputs control signals to the process 1 as a function of the measured process variables. The analog input 2 and the analog output 3 are components of a distributed peripheral 4. Via a field bus 5, the distributed peripheral 4 and other, for clarity, not shown in the figure 1 field devices connected to a programmable logic controller 6. During operation of the automation system runs in the programmable logic controller 6 on a processor, not shown, from a control program containing, for example, the function of a controller and a diagnosis of a control loop. The control program is using an engineering system 7, z. B. SIMATIC PCS7, designed. As a hardware basis can serve this example, a personal computer. FIG. 1 also shows a graphical function block diagram 8 intended to illustrate, by way of example, the design of a control loop with a monitoring device. Such a function block diagram can be edited graphically on the engineering system 7. In a control circuit 9, a controller 11 is disposed behind a subtractor 10, which determines a manipulated variable u from the deviation between a desired value SP and a controlled variable y, that is to say as a function of a control deviation e. This is given to an analog output module 12, the output of which affects a process 13. With an analog input module 14, the controlled variable y is measured and fed back to the subtractor 10. The control circuit 9 is assigned a monitoring module 15, which evaluates the desired value SP, the manipulated variable u and the controlled variable y. In addition to the process variables of the control circuit 9 shown, the diagnostic module 15 by evaluating other process variables - as symbolized by 3 points on the left input side of the block 15 - monitor other, not shown in the figure control loops. The output of the diagnostic results via a faceplate, which may be arranged in a control and monitoring workstation 16, which is connected to the same data transmission network 17, in which the engineering system 7 and the programmable logic controller 6 are located. Alternatively, each individual controller function block can receive its own diagnostics block as an add-on. Overviews of the state of all control loops are then created using the means of the process control system, in which the information of all diagnostic function blocks is merged. After the design of the control software, this is loaded into the programmable logic controller 6 and runs there during the operation of the automation system there according to the function blocks used and their connections.

0015Figure 2 shows the basic structure of a diagnostic function block. Input signals are the process variables to be monitored, in the embodiment shown, the desired value SP, the controlled variable y and the manipulated variable u. In addition, further process variables can be monitored and fed to the diagnostic function block 15. In a data memory 20 in each case a sequence of setpoint and actual value data of the control loop, that is, a portion of the time course of the setpoint SP and a portion of the time course of the control variable y, stored. A selector 21 analyzes the state of the associated control loop on the basis of the setpoint data. In the case of a substantially stationary state, the selection device 21 activates an evaluation device 22 for determining at least one stochastic feature, for example, B. the variance, the monitored process variables. The evaluation device 22 compares values ​​of the variance, which are determined during operation of the automation system, with respectively associated, predetermined reference values. If the respective reference value is exceeded, this may indicate an error in the control loop. If the selection device 21 determines an excitation of the control circuit substantially in accordance with a predetermined, in particular step-shaped, course of the setpoint, it activates an evaluation device 23 for determining at least one deterministic feature and compares this particular value of the feature with an associated, predetermined reference value. Preferably, when excited by a setpoint jump, the transient response of the step response and / or the relative overshoot are evaluated as deterministic features. In addition to the two states of the control loop mentioned above, there may be other states in which no evaluation of the process variables is performed. The state of the control loop determined by the selection device 21 and the value of a stochastic feature determined in the evaluation device 22 and / or the value of a deterministic feature determined in the evaluation device 23 are stored in a result memory 24 of the function block 15. The output of the values ​​and graphic processing is carried out with the help of a faceplate on the display unit of a HMI device, which also performs a trend analysis and displays the results.

0016The diagnostic function block 15 thus evaluates both stochastic features of process variables and also uses deterministic features for the diagnosis which are relevant for the assessment of the control quality of the control loop. Depending on the operating status of the control loop, the appropriate features are automatically selected. If a setpoint jump is detected on the control loop by evaluating the profile of the setpoint SP, two deterministic features are determined based on the time profile of the step response of the control variable y of the closed control loop: the relative overshoot range and the transient response ratio. The relative overshoot is related to the level of the setpoint jump. The transient response is calculated as the quotient of rise time and transient time. Both features allow for very direct and diagnostically well evaluable conclusions about the control quality. If a specifiable multiple of the rise time, for example, the four-fold rise time passed and the step response is settled, the control loop is again in a steady state and the evaluation can be switched back to stochastic features mean and variance. Preferably, instead of the theoretical minimum variance of the control loop, a variance measured real on this individual control loop in the good state is used as a reference variance in order to improve the informative value of a so-called control performance index.

0017Functions for monitoring control circuits can be located at different levels within the system architecture of an automation system or a process control system. Aspects for such decisions are on the one hand the need for storage space and computing time and on the other the relevant time horizon but also aspects such as engineering effort as well as operator control and monitoring. In a diagnostic function block, which can be regarded as an additional block for a controller function block, the functions can be accommodated with advantage to current data or access data from a limited time window of length n with n &lt;100. As far as possible, a recursive calculation for the evaluation of the data is to be preferred, as it is practically possible to dispense with data storage.

0018In the evaluation in function module 15, a distinction is made between deterministic and stochastic features of the process variables. In normal operation, stochastic features are constantly evaluated. However, as soon as a setpoint step change is detected with the selection device 21, for example, the evaluation device 23 is activated, which evaluates the setpoint value jump with respect to deterministic features. At the diagnostic function block 15, a corresponding display bit is set. Once the transient is completed and a period of time has passed, the deterministic features are output and the display bit is reset.

0019Based on the time profiles of the setpoint SP, the controlled variable y and the manipulated variable u, a control deviation e and time averages of these variables are calculated. A recursive calculation formula is applicable. An average deviation e at a constant setpoint is already an indication of problems in the control loop if the controller used is a controller with an I component. If a stationary operating point (u₀, y₀), the average current process gain kg can be estimated from the calculated values ​​if it is assumed that only mean-free interferences intervene:<maths id="math0001" num=""><math display="block"><mrow><msub><mrow><mtext>k</mtext></mrow><mrow><mtext>G</mtext></mrow></msub><mtext> = </mtext><mfrac><mrow><mover accent="true"><mrow><mtext>y</mtext></mrow><mo>¯</mo></mover><msub><mrow><mtext> - y</mtext></mrow><mrow><mtext>0</mtext></mrow></msub><mtext></mtext></mrow><mrow><mover accent="true"><mrow><mtext>u</mtext></mrow><mo>¯</mo></mover><msub><mrow><mtext> - u</mtext></mrow><mrow><mtext>0</mtext></mrow></msub></mrow></mfrac></mrow></math><img file="EP1528447A1_D0001.tif" /></maths> With <maths id="math0002" num=""><math display="inline"><mrow><mover accent="true"><mrow><mtext>y</mtext></mrow><mo>¯</mo></mover></mrow></math><img file="EP1528447A1_D0002.tif" /></maths> - Mean value of the controlled variable y and    <maths id="math0003" num=""><math display="inline"><mrow><mover accent="true"><mrow><mtext>u</mtext></mrow><mo>¯</mo></mover></mrow></math><img file="EP1528447A1_D0003.tif" /></maths> - Mean value of the manipulated variable u.

0020Depending on the application, typical stationary operating points are often known in advance, eg. For flow control: y₀ = 0 for u₀ = 0, ie the valve is closed, or at a temperature control: y₀ = y_amb for u₀ = 0, that is ambient temperature y_amb when the heating is switched off. Often, stationary operating points can also be determined during initial startup. The variance as a centered moment requires in terms of the approach the calculation of differences between each current value of a process variable and its mean:<maths id="math0004" num=""><img file="EP1528447A1_D0004.tif" /></maths>

0021The scatter<maths id="math0005" num=""><math display="block"><mrow><msub><mrow><mtext>σ</mtext></mrow><mrow><mtext>y</mtext></mrow></msub><mtext> = </mtext><msqrt><msubsup><mrow><mtext>σ</mtext></mrow><mrow><mtext>y</mtext></mrow><mrow><mtext>2</mtext></mrow></msubsup></msqrt></mrow></math><img file="EP1528447A1_D0005.tif" /></maths> as the root of the variance can be interpreted more vividly, since it has the same physical unit as the measured value.

0022A Control Performance Index CPI is calculated as the quotient of a reference variance of the controlled variable and its current variance:<maths id="math0006" num=""><math display="block"><mrow><mtext>CPI = </mtext><mfrac><mrow><msup><mrow><mtext>σ</mtext></mrow><mrow><mtext>2</mtext></mrow></msup><mtext>ref</mtext></mrow><mrow><msubsup><mrow><mtext>σ</mtext></mrow><mrow><mtext>y</mtext></mrow><mrow><mtext>2</mtext></mrow></msubsup></mrow></mfrac><mtext> ,</mtext></mrow></math><img file="EP1528447A1_D0006.tif" /></maths>

0023The Control Performance Index usually ranges between 0 and 1. If the current spread equals the reference value, the CPI reaches 1. If the current spread is much larger, the CPI returns very small values. Different values ​​can be considered as reference values. For example, the variance in a defined good condition can serve as a reference value. This can be logged after controller commissioning with a PID tuner. As long as the controllers and track parameters remain unchanged, ie the good condition is not left, the CPI is on the order of about 100% corresponding to the value 1. A significant reduction of the CPI value indicates an inappropriate regulator setting, e.g. B. can be caused by changes in the process at constant controller parameters. With a reference variance defined in this way, it is temporarily possible for the CPI to be greater than 100%.

0024Another example of a reference value of variance is the variance that arises with a theoretically ideal PID controller, which requires an explicit process model and an explicit noise filter model for the calculation. However, these conditions are seldom fulfilled in practice.

0025The reference value could also be the variance obtained with a theoretical minimum variance controller. This depends only on the process dead time and the fault model. This form of CPI, called the Harris index, represents a generally unobtainable lower bound for a PID controller, which is why the CPI assumes values ​​between 0% for low-set controllers and close to 100% for well-adjusted controllers. Low CPI values ​​provide a first indication that the controller setting can still be improved. It should be noted, however, that the minimum variance is only a theoretically achievable value and the minimum variance controller has properties that are not desirable in practical applications, such as extremely large manipulated variables. So for a CPI based on the minimum variance, it is not desirable to bring it as close to 100% as possible at all costs.

0026With reference to FIG. 3, a few deterministic features, which are suitable in particular for evaluating the control quality, will be clarified below. Shown is a time diagram with the qualitative progressions of some process variables of the control loop. On the abscissa is the time t, plotted on the ordinate, the setpoint SP, the controlled variable y and the manipulated variable u. It is a schematic diagram in which the indication of units has been omitted since these depend on the respective application. As a substantially predetermined course of the target value SP, a set value jump from an initial value SP0 to a final value SP1 at a time t0 is considered. As a result, the illustrated course of the controlled variable y overshoots over the stationary end value. The absolute height of the overshoot Ovs max is related to the height of the setpoint step SP_height and expressed as a percentage. If the regulator setting is good, the relative overshoot is between 0 and 15%. Overshoots greater than 30% indicate too weak attenuation of the control loop, that is a tendency to oscillating behavior out.

0027Another feature for the evaluation of the control loop is the transient response at a setpoint step. This is calculated as the quotient of a rise time T_rise by a settling time T_settling, the definition of which is readily derivable with reference to FIG. The settling time T_settling indicates which time period passes after the jump until the step response for the last time enters a zone with a maximum distance BW from the setpoint SP1. The transient response ranges from 30% to 100% with well-adjusted controls. Lower values ​​can be caused by an excessive reset time of the controller. By evaluating archive data, overall statements can be made for longer periods and slow trends or Deteriorations are detected.

0028To illustrate the results of the diagnosis to an operator, are in a main view of a faceplate, that on an operating and monitoring station, for example, a SIMATIC PCS7 OS, the current CPI, the standard deviation of the controlled variable y, the mean values ​​of the controlled variable y and the manipulated variable u, the stationary amplification and the relative time shares, while the controller is in manual mode, and the relative time shares, in which the manipulated variable u meets the upper or lower limit, displayed. In another "setpoint" view of the faceplate, the maximum and relative overshoot, the settling time and the transient response are displayed after a transient response has ended as a result of a setpoint jump. These values ​​are retained after the evaluation and will only be deleted at the next jump. In a "Parameter" view, the window length required for averaging can be specified. In addition, an initialization of the reference values ​​for stochastic features can be triggered in this view by means of a button. The reference values ​​of variance and standard deviation are displayed in designated fields. Another "Limits" view is used to set warning and alarm limits for the Control Performance Index CPI and relative overshoot. Corresponding messages are generated only when the CPI falls below a predetermined value or the overshoot is greater than the predetermined upper limit. Thus, only lower limits and, for the overshoot, only upper limits are given for the CPI. In a "Trend" view, calculated time histories of the Control Performance Index CPI, the mean of the controlled variable y, and the standard deviation are output for easier evaluation by an operator. In accordance with the SIMATIC PCS 7 standard, the messages generated by the diagnostics function block are listed in a "Messages" view, with their date, time, class, status, and event description.