Predictive Drum Level Control for Power Plant Transients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional drum level control in power generation plants is ineffective during transient operations due to complex two-phase flow dynamics, wave presence, unknown heat and pressure disturbances, and load demands, leading to equipment damage and plant disruptions.
Innovation Solution
A predictive control system using sensors to measure liquid level, vapor flow rate, pressure, temperature, and feed-water flow rate, which predicts the liquid volume and generates a set point to adjust control elements, anticipating and mitigating disturbances before they affect the water level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional reactive control is used to maintain fixed water level set point, then the control system is simple to implement, but it cannot anticipate disturbances and provides insufficient response time during transient operations
Solution Approach 1:
The control system performs preliminary actions by predicting future drum water level and required feed water flow rates based on current system state and anticipated disturbances. The predictive controller calculates the necessary control actions in advance, allowing the system to prepare and respond proactively rather than reactively to level changes, thereby improving reliability without excessive complexity.
Solution Approach 2:
The system implements feedback by continuously monitoring drum water level, feed water flow rate, steam flow rate, and other parameters. This feedback is fed into the predictive controller which uses it to update predictions and adjust control actions, creating a closed-loop system that adapts to changing conditions while maintaining manageable complexity through automated decision-making.
2Ease of operation
If conventional control reacts to observed changes in level and steam flowrate, then the control logic is straightforward, but it exhibits inverse response and cannot handle complex two-phase flow dynamics effectively
Solution Approach 1:
The patent replaces conventional mechanical/reactive control logic with a predictive control system that uses computational algorithms to anticipate system behavior. Instead of reacting to observed changes, the system uses mathematical models and sensor data to predict future states and calculate optimal control actions, handling complex two-phase flow dynamics more effectively while maintaining ease of operation through automated control.
3Adaptability or versatility
If the controller maintains a fixed water level set point, then the control target is simple, but it does not account for variation in drum states and predicted disturbances
Solution Approach 1:
The control system transitions from a static fixed set-point approach to a dynamic adaptive approach. The predictive controller continuously updates the water level set point based on varying drum states (pressure, temperature, flow rates) and predicted disturbances (load changes, feed water variations). This dynamic adaptation improves versatility while the automated computational process manages the complexity of handling multiple varying parameters.
Data Source
AI summary
A level control system for controlling a liquid level in a vessel containing a two-phase fluid includes a plurality of sensors configured to measure parameters related to the vessel. The parameters include liquid level in the vessel, vapor flow rate leaving the vessel, pressure in the vessel, temperature of the vessel, and feed-liquid flow rate entering the vessel indicative of a state of the vessel. A predictive controller is configured to receive output signals from the plurality of sensors and predict a volume of liquid over a predetermined time period in the vessel based on output signals from the plurality of sensors and a variation in pressure, thermal load, or combinations thereof in the vessel. The controller is configured to generate a liquid level set point of the vessel based on the predicted volume of liquid in the vessel; and further control a liquid level in the vessel based on the generated liquid level set point by manipulating one or more control elements coupled to the vessel.