Performance monitoring, diagnosis and maintenance for industrial controller with automatic operation capability

Abstract

The invention disclose a performance monitoring, diagnosis and maintenance for an industrial controller with the automatic operation capability, and the method comprises the following steps: S1, respectively collecting and storing the data samples of a control loop set value variable SP, a controller output variable OP, a process output variable PV, an actuator state variable AV, a redundancy measurement variable SV and a related process variable RV; S2, recognizing a process model according to the above data samples of the variables, assessing and diagnosing the performance of the controller, so as to obtain the working state of the controller, and adjusting the a variable parameter, corresponding to the working state, of the controller according to a preset processing strategy and the obtained working state. The method can automatically analyze the daily industrial operation data, automatically extracts process information, continuously updates the performance reference automatically according to the changes of dynamic characteristics, and guarantees the real-time monitoring of the performances of the controller.

Description

technical field [0001] The invention relates to the technical field of industrial control, in particular to a method for monitoring, diagnosing and maintaining the performance of an industrial controller with automatic operation capability. Background technique [0002] In common process industries such as oil refineries and chemical plants, the number of control loops is large, and the operating status of related industrial controllers is closely related to product quality and production safety. However, the working state of the actual controller is not satisfactory, and there is still much room for improvement in its control performance. For actual process operators, the workload of manual maintenance of these controllers is huge, so the process industry urgently needs a set of industrial controller performance monitoring, diagnosis and maintenance systems with automatic operation capabilities. [0003] Since the controller performance evaluation technology proposed by Ha...

AI Technical Summary

Problems solved by technology

However, in practical applications, the inventors found that the implementation of this performance evaluation technology still has the following problems: First, the gap between the performance benchmark and the optimal performance that the actual controller can achieve is too large, which makes the performance of the actual controller difficult. been accurately assessed

Second, for different loops, it can be determined that the thresholds for controller performance degradation are quite different, so it is necessary to manually determine the thresholds for different loops, resulting in a rapid rise in work configuration costs

Third, some monitoring methods need to obtain the necessary process model information through specific identification experiments, and the identification experiments will also greatly increase the implementation cost

Images