Terminology
Process+
The following terms apply specifically to Process+:
Loop data: The timestamp, quality and value for each tag in a loop, read from either a historical or real-time source.
Loop configuration information: This data details the configuration of the specific control loop. Variables within the loop include the set point value, process variable value, and manipulated variable value. There is also configuration information for the actual control loop, including: upper and lower limits, the control mode etc. See: loop definition file.
Interim data: Plant data is collected from a data source, and stored as interim data on the CSense Data Server until the time period for a specific collection cycle is over. This collection cycle can be anything from a few seconds for historical sources, to 24 hours for a full day shift using a real-time source. The Process+KPIs are calculated using this interim data.
Process+ KPIs: Process+ KPIs are calculated from collected interim data. Determined by the type of Process+ Action Object deployed, different calculated values are generated, or different aspects of the control loop specified in the loop configuration information is calculated. The Process+ KPIs are stored in the CSense Data Server.
Process+ Action Object: an Action Object is a collection of loop configuration information and calculation instructions, which is deployed online. Using this control loop information, Process+ KPIs are generated from calculated statistics, for easy monitoring and understanding of the status of the control loops.
CSense Data Server: Process+ Action Objects are deployed online using the CSense Runtime Manager installed on a CSense Data Server machine. Multiple machines can be CSense Data Servers, each deploying different Action Objects. This distributed architecture enables effective handling of large numbers of control loops. Collected interim data and calculated Process+ KPI data are stored on the CSense
Control systems
Process Variable, PV: a measurable actual target variable value.
Set Point, SP: a known ideal value for the target variable - the optimal target variable value.
Manipulated Variable, MV: a process input: the manipulated variable will adjust the amount of control exerted on the process variable.
Proportional value (P): The variable control required in proportion to the current needs of the system. The output of this part of the controller is proportional to the amplitude of the input signal, usually the error value. The feedback control action required is therefore determined by the size of the error: enough action is required so as to reduce the error, while avoiding overshooting the set point, and causing system oscillation.
Integral value (I): Integral control applies ever-increasing control actions, until the error is reduced to zero. This means that if the control action being applied does not bring the system to operate at the set point, for whatever reason, integral action increasingly applies more control action, moving the proportional control band relative to the set point value, until the PV error is reduced to zero and the set point is achieved. The output for this part of the controller moves at a speed proportional to the input signal value, usually the error value.
Derivative value (D): The derivative value of the algorithm is determined by how quickly the error changes over time. If the PV approaches the set point rapidly, then manipulated variable is reduced early to allow the PV to gradually reach the desired SP. Conversely, if the PV value begins to move away from the set point rapidly, extra control actions are applied in an attempt to maintain the SP value. The output for this part of the controller is proportional to the input signal speed.
Filter value (F): A filter is used to reduce fast disturbances in the process variable, which are high frequency noise signals. The controller is too slow to remove these fast disturbances, so, without the use of a filter, the noise signals are transmitted through the controller and are shown in the controller output. If the filter constant is properly selected, only the actual process disturbances are applied to the controller, the noise being removed or reduced by the filter.
Error: the error values refer to how far the process variable (PV) is operating from the set point value (SP). The intention of control systems is to minimize this error.
Hysteresis: the amount of error that results when there is a delay between a process action and the reaction of a measuring instrument.
Noise: the unwanted signals from unrelated electrical circuits, magnetic fields and vibrations that may be picked up by the control system sensor. Noise causes fast disturbances in the process variable input that need to be filtered out for control purposes.
Stiction: from "static friction", also influenced by the verb "stick". It describes objects which are in a fixed place, and not sliding, but which still experience a form of friction, called static cohesion. Stiction is a threshold, not a continuous force. A force needs to be applied to the object to overcome the stiction before the object will begin to move.
Related topics: