Industrial Boiler Cascade MPC for Pressure and Level Stability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Industrial boilers face challenges in controlling multiple variables such as steam production, water level, and drum pressure effectively, as existing PID controllers are limited in multivariable systems and prior MPC methods lack practical implementation strategies for real-world boilers.

Innovation Solution

A supervisory cascade MPC control architecture is implemented, using an MPC controller with custom estimation algorithms for steam pressure and fluid level control, integrated with PI controllers to optimize boiler operation, and a cost function focused on fuel mass flow, with a Kalman Filter for state estimation and backwards linear interpolation to address quantization phenomena.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If PID controllers are used for boiler control, then the control structure is simple and easy to implement, but the control performance is limited and cannot effectively handle multivariable systems

Engineering Contradiction:
Improvecontrol implementation simplicityVSAvoidcontrol performance
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The control system is segmented into two hierarchical levels: an MPC controller for optimal multivariable control and PI controllers for basic regulation. This segmentation allows the system to leverage the strengths of both control approaches while managing complexity through functional division.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The PI controllers are nested within the overall control architecture as subordinate controllers to the MPC controller. The MPC controller operates at the higher level for strategic decision-making, while PI controllers handle tactical adjustments, creating a nested control structure where simpler controllers are embedded within a more sophisticated framework.

Inventive Principle:
Principle #7Nested doll (Nesting)

2Reliability

If MPC control is applied to improve boiler performance, then optimal control is achieved, but the system complexity increases significantly

Engineering Contradiction:
Improvecontrol performanceVSAvoidcontrol system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The control system is segmented into two hierarchical levels: an MPC controller for optimal multivariable control and PI controllers for basic regulation. This segmentation allows the system to leverage the strengths of both control approaches while managing complexity through functional division.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The MPC controller is applied partially to only the most critical multivariable aspects of boiler control (steam production, water level, drum pressure), while PI controllers handle routine regulation tasks. This partial application of advanced control minimizes complexity while achieving performance improvements where they are most needed.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If prior MPC methods are used, then theoretical optimization is achieved, but practical implementation strategy is lacking for real industrial boilers

Engineering Contradiction:
Improveoptimization performanceVSAvoidimplementation feasibility
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The MPC controller incorporates real-time feedback from boiler sensors (steam production, water level, drum pressure) to continuously adjust control inputs. This feedback mechanism bridges the gap between theoretical optimization and practical implementation by adapting the control strategy to actual operating conditions.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The MPC controller uses predictive modeling to anticipate future boiler states and pre-adjust control inputs before deviations occur. This preliminary action enables the system to proactively maintain optimal performance rather than merely reacting to changes, improving both optimization and implementability.

Inventive Principle:
Principle #10Preliminary action

4Measurement precision

If advanced control strategies are implemented, then control precision is improved, but cost effectiveness decreases

Engineering Contradiction:
Improvecontrol precisionVSAvoidcost effectiveness
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The MPC controller is applied partially to only the most critical multivariable aspects of boiler control (steam production, water level, drum pressure), while PI controllers handle routine regulation tasks. This partial application of advanced control minimizes complexity while achieving performance improvements where they are most needed.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The control system is segmented into two hierarchical levels: an MPC controller for optimal multivariable control and PI controllers for basic regulation. This segmentation allows the system to leverage the strengths of both control approaches while managing complexity through functional division.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3889696B1A model predictive control method for an industrial boiler and relating boiler structure
Publication Date: 2023.08.16 BONO ENERGIA SPA
  • EP3889696B1 patent drawingFigure 1
  • EP3889696B1 patent drawingFigure 2
  • EP3889696B1 patent drawingFigure 3

AI summary

The present invention relates to a method and system for optimizing the control of an industrial boiler. An MPC control method is applied to an industrial boiler (680) and an industrial boiler system (450) wherein at least a couple of variable parameters (P, L) are controlled through a supervisory cascade MPC control architecture (400) for optimizing the boiler operation, for instance the level (L) sensed by level sensor of the fluid to be heated in the boiler and the value of the steam pressure (P) produced by the boiler (450) and sensed by pressure sensors. The MPC control is applied on the pressure value (P) through a custom estimation algorithm to tackle a quantization phenomenon affecting the pressure sensors.