Abstract:The rapid growth of renewable generation and its intermittent nature has modified the role of combined cycle power stations in the energy industry, and the key feature for the operational excellence is now flexibility. Especially, the capability to start an installation quickly and efficiently after a shutdown period leads to lower operational cost and a higher capacity factor. However, most of existing thermal power stations worldwide are designed for continuous operation, with no special focus on an efficient start-up process. In most current start-up procedures, the gas turbine controls ensure maximum heat flow to the heat recovery steam generator, without feedback from the steam cycle. The steam cycle start-up controls work independently with as main control parameter the limitation of the thermal stresses in the steam turbine rotor. In this paper, a novel start-up procedure of an existing combined cycle power station is presented, and it uses a feedback loop between the steam turbine, the boiler and the gas turbine start-up controls. This feedback loop ensures that the steam turbine can be started up with a significant reduction in stresses. To devise and assess this start-up methodology, a flexible and accurate dynamic model was implemented in the SimulinkTM environment. It contains more than 100 component blocks (heat exchangers, valves, meters and sensors, turbines, controls, etc.), and the mathematical component sub-models are based on physical models and experimental correlations. This makes the model generally applicable to other power plant installations. The model was validated against process data related to the three start-up types (cold start, warm start, hot start). On this basis, the optimization model is implemented with feedback loops that control for example the exit temperature of the gas turbine based on the actual steam turbine housing temperature, resulting in a smoother heating up of the steam turbine. The optimization model was used to define the optimal inlet guide vanes position and gas turbine power output curves for the three types of start-up. These curves were used during real power station start-ups, leading to, for cold and warm starts, reductions in the start-up time of respectively 32.5% and 31.8%, and reductions in the fuel consumption of respectively 47.0% and 32.4%. A reduction of the thermal stress in the steam turbines is also achieved, thanks to the new start-up strategy.

Intelligent Steam Turbine Start-Up Control Based on Deep Reinforcement Learning

Optimization on Start-up Process of Steam Turbine in Combined Cycle Generating Units

Optimization of Start-up Process in Combined Cycle Steam Turbine of a Power Plant

Study on Quick Start-up of Steam Turbine in a Gas-Steam Combined Cycle

Research on New Type of Fast-Opening Mechanism in Steam Turbine Regulating System and Optimization of Operation Tactic

Deep-learning modeling and control optimization framework for intelligent thermal power plants: A practice on superheated steam temperature

Towards Predictive Maintenance of a Heavy-Duty Gas Turbine A New Hybrid Intelligent Methodology for Performance Simulation

Start-Up Optimization of a CCGT Power Station Using Model-Based Gas Turbine Control

Start-Up Optimization of a CCGT Power Station Using Model Based Gas Turbine Control

Modelling and regulation of steam temperatures of a 1000-MW double-reheat boiler with long short-term memory

Adaptive Finite‐time Control for Turbo‐generator of Power Systems with Prescribed Performance

Intelligent Control Strategy for Transient Response of a Variable Geometry Turbocharger System Based on Deep Reinforcement Learning

Multi-Objective Optimization of Economy, Safety and Emissions for the Startup Scheduling of a Steam Turbine

An Optimization Method of Steam Turbine Load Resilient Adjustment by Characterizing Dynamic Changes in Superheated Steam Energy

Damage Analysis and Start-Up Schedule Optimization of Steam Turbine Rotors With an Anisotropic Damage Coupled Viscoplastic Constitutive Model

New Monitoring Model of Thermal Stresses on Steam Turbine Rotors

Multi-objective optimization of thermal power and outlet steam temperature for a nuclear steam supply system with deep reinforcement learning

Model-Based Analysis Of The Start-Up Improvement Of A Ccpp Due To Steam Turbine Warm-Keeping With Air

Adaptive control for circulating cooling water system using deep reinforcement learning

Deep reinforcement learning for PMSG wind turbine control via twin delayed deep deterministic policy gradient (TD3)

Superheated Steam Temperature Control Based on Modified Active Disturbance Rejection Control