Central Plant Cost Optimization Across Subplants and Thermal Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Central plants face challenges in optimizing energy distribution across multiple subplants to minimize energy costs, particularly when integrating thermal energy storage with real-time pricing and demand charges, due to complex control technology requirements.
Innovation Solution
An optimization system that uses a processing circuit to receive load prediction data and utility rate data, generating an objective function to optimize the monetary cost of operating the central plant through high-level optimization modules employing linear programming and dynamic programming, determining optimal distribution of energy loads across subplants and their operating statuses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If thermal energy storage is integrated with multiple subplants and real-time pricing, then energy cost reduction is improved, but control technology complexity increases
Solution Approach 1:
The control technology is segmented into a modular optimization system with distinct functional components: a processing circuit for receiving data, an objective function generation module for cost modeling, and an optimization module for load distribution. This segmentation transforms the complex integrated control problem into manageable modular units that can be developed, implemented, and maintained independently while working together to reduce energy costs.
Solution Approach 2:
An optimization module acts as an intermediary between the thermal energy storage system, multiple subplants, and real-time pricing signals. This intermediary receives pricing data and storage status, processes this information through the objective function, and generates optimized load distribution commands. The intermediary abstracts the complexity of coordinating multiple subsystems with real-time pricing, presenting a simplified control interface while managing the intricate optimization calculations internally.
2Loss of energy
If load distribution across multiple subplants is optimized in real-time, then energy cost minimization is improved, but computational complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-defining the objective function that models energy costs as a mathematical relationship between load distribution and pricing signals. This objective function is prepared in advance with all necessary cost parameters and constraints established, so that when real-time optimization is needed, the system only needs to solve the optimization problem using pre-established cost models rather than building cost models from scratch each time.
Solution Approach 2:
The optimization system manages computational complexity by changing parameters dynamically - using fixed parameters for cost coefficients and constraints that don't change frequently, while only varying the load distribution variables that need optimization. This parameter separation allows the system to reuse established cost models and constraints across multiple optimization cycles, significantly reducing the computational burden of real-time optimization while still achieving cost minimization.
Data Source
AI summary
A controller for equipment obtains utility rate data indicating a price of one or more resources consumed by the equipment to serve energy loads. The controller generates an objective function that expresses a total monetary cost of operating the equipment over an optimization period as a function of the utility rate data and an amount of the one or more resources consumed by the equipment at each of a plurality of time steps. The controller optimizes the objective function to determine a distribution of predicted energy loads across the equipment at each of the plurality of time steps. Load equality constraints on the objective function ensure that the distribution satisfies the predicted energy loads at each of the plurality of time steps. The controller operates the equipment to achieve the distribution of the predicted energy loads at each of the plurality of time steps.