Abstract:With the growing proportion of renewable energy sources in power grid, the demands for deep peak shaving and rapid load regulation of ultra-supercritical (USC) unit to provide more flexible and stable power services is further enhanced. Simultaneously, the load tracking speed is no longer the only significant task in the flexible operation of USC unit, the operational economy is more and more important. To this end, a hierarchical reinforcement learning GPC control scheme with a better economy is proposed and apply to USC unit. Firstly, the model mismatch problem in control process is taken into account, and an integral mode is added into the GPC to eliminate the steady state error during the load regulation. Secondly, the hierarchical reinforcement learning GPC scheme integrated the GPC and twin delayed deep deterministic policy gradient algorithms is presented. The complex flexibility demand problem is decomposed into rapid load regulation and economic operation and then respectively solved in the GPC-based upper layer and reinforcement learning agent-based lower layer. Thirdly, under the flexibility demand, a multi-criteria optimization function including the load tracking cost, main steam throttling loss and coal consumption rate of unit is constructed. Then, via the accurate trajectory tracking by GPC and on-line control sequence optimization by reinforcement learning agent, the operational flexibility and economy of USC unit are simultaneously improved within safe boundaries. Finally, to verify the availability of the proposed hierarchical reinforcement learning GPC strategy, large-scale load variations and disturbance rejection ability tests cover the 30%–100 % rated load of a 1000 MW USC unit are carried out, and the load regulation performance is compared to other strategies. Results confirm that the designed hierarchical reinforcement learning GPC control scheme is profit for the flexible and economic operation capability promotion of USC unit.

An effective multi-model based nonlinear control for USC power plant

Nonlinear model predictive control of USC boiler-turbine power units in flexible operations via input convex neural network

A Power Plant Coordinated Control System Using MPC

Development of a Three-Term MPC and Its Application to an Ultra-Supercritical Coal Fired Power Plant

Multi-model Predictive Control of Ultra-supercritical Coal-fired Power Unit

Two-timescale Nonlinear Model Predictive Control for Flexible Operation of Coal-Fired Power Plant with Post Combustion CO2 2 Capture System

Improved Internal-Model Robust Adaptive Control with Its Application to Coordinated Control of USC Boiler-Turbine Power Units in Flexible Operations

A Composite Nonlinear Control Scheme of the Ultra-Supercritical Boiler-Turbine Unit for Large-Scale Load Accommodation

Nonlinear Model Predictive Control of Ultra-Supercritical Once Through Boiler-Turbine Unit

Steam Pressure Control of 1 000MW Ultra-Supercritical Coal-Fired Power Unit Based on Multi-Model Predictive Control

Control of ultra-supercritical once-through boiler-turbine unit using MPC and ESO approaches

A hierarchical reinforcement learning GPC for flexible operation of ultra-supercritical unit considering economy

A Nonlinear Model Predictive Control Strategy For Superheater Process In Thermal Power Plant

Extended State Observer Based Fuzzy Model Predictive Control for Ultra-Supercritical Boiler-Turbine Unit

Nonlinear Predictive Control For A Boiler-Turbine Unit Based On A Local Model Network And Immune Genetic Algorithm

An Adaptive Nonlinear Model Predictive Control Based On Online Svm Regression Algorithm

A Dynamic Nonlinear Model for a Wide-Load Range Operation of Ultra-Supercritical Once-Through Boiler-Turbine Units

Multivariable constrained predictive control of main steam temperature in ultra-supercritical coal-fired power unit

Research and application of multiple model Predictive Control in Ultra-supercritical boiler-turbine system

MLD–MPC for Ultra-Supercritical Circulating Fluidized Bed Boiler Unit Using Subspace Identification

Multivariable Constrained Predictive Control of 1000 MW Ultra Supercritical Once-Through Boiler-Turbine System