Parameter optimization method for superheated steam temperature control system

Abstract

The invention discloses a parameter optimization method for a superheated steam temperature control system, and the method comprises the steps: a data collection unit which obtains the historical operation data of a steam temperature at set intervals under a stable load; a LSTM (Long Short-Term Memory Neural Network) prediction unit which is used for establishing a time prediction model and predicting steam temperature output data at the future moment through the steam temperature data at the historical moment; a performance index calculation unit which is used for constructing indexes of system performance evaluation through the Hurst index and the Mahalanobis distance and calculating performance indexes of multiple groups of steam temperature prediction data acquired in a period of timeat the future moment; and a parameter optimization guidance unit which gives suggestions for adjustment of the controller parameters. The data acquisition unit, the LSTM prediction unit and the parameter optimization guidance unit are connected in sequence. The parameter optimization method for the superheated steam temperature control system is simple in calculation, and the control performance of the superheated steam temperature control system is improved.

Description

technical field [0001] The invention relates to the technical field of superheated steam temperature control systems in thermal power plants, in particular to a parameter optimization method for superheated steam temperature control systems. Background technique [0002] The performance improvement of the control loop is a key factor determining the production efficiency of the industrial process. However, in actual situations, the characteristics of the controlled object change or the aging of the actuator often occurs, resulting in a decline in the performance of the controller, affecting economic benefits and even bringing safety hazards. , data show that up to 60% of the controllers can not play its expected performance. Therefore, monitoring the performance of these control loops and modifying the controller parameters before performance degradation is of great significance to improve the efficiency, quality and safety of the entire industrial process. The steam temper...

AI Technical Summary

Problems solved by technology

The steam temperature system in a thermal power unit is a complex system. Due to its unique complexity, it is difficult to obtain the model in engineering practice. PID adjustment is the most important control strategy of the thermal system, which directly affects the unit. Although the PID controller has many advantages, it is difficult to set the parameters of the PID controller applied to thermal objects. Especially for thermal objects with large inertia and large delay, conventional tuning methods such as Z-N, Engineering setting methods such as C-C, ITAE step response curve method and decay curve method make it more difficult to set PID parameters, and it is not easy to obtain good control effect

Images