Generator Set Operating Sequence Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current optimization methods for determining the operating sequence of generating sets in electrical networks are inefficient for networks with more than four generator sets, as they fail to optimize fuel consumption, wear, and greenhouse gas emissions due to computational limitations.
Innovation Solution
A computer-implemented method that assigns ranks to generating sets, determines a minimum number based on forecast load curves, and optimizes a cost function to assign on/off states and operating powers, reducing computational complexity and symmetry in the problem.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If optimization methods by mixed linear programming or dynamic programming are used, then optimal planning in terms of fuel consumption, wear and tear, and greenhouse gas emissions can be achieved, but the method cannot be applied to electrical networks having more than 4 generator sets due to computational limitations
Solution Approach 1:
The patent segments the problem by introducing a hierarchical structure with a coordination level and execution levels. The coordination level determines the operating sequence by assigning ranks to generating sets, while execution levels handle real-time power distribution. This segmentation divides the computationally intensive optimization problem into smaller, more manageable sub-problems that can be solved efficiently even for networks with more than 4 generators.
Solution Approach 2:
The patent changes the problem parameters by introducing rank assignments and a simplified cost function that focuses on key objectives (fuel consumption, wear, emissions) without requiring exhaustive optimization. This parameter transformation allows the system to achieve near-optimal solutions with significantly reduced computational complexity, enabling application to larger networks.
2Quantity of substance
If the number of generating sets increases beyond 4, then the electrical network can meet higher energy demand, but the computational complexity makes optimal planning infeasible with traditional methods
Solution Approach 1:
The patent divides the control architecture into a coordination level that handles sequence determination and execution levels that handle real-time operation. This segmentation allows the system to manage larger numbers of generating sets by distributing computational tasks, with the coordination level using a simplified ranking approach that scales better than traditional optimization methods.
Solution Approach 2:
The patent applies local quality by allowing different execution levels to operate with different degrees of autonomy and different cost function priorities. Each execution level can be optimized for specific local conditions while the coordination level ensures global optimality, enabling the system to handle larger networks efficiently.
3Manufacturing precision
If traditional optimization methods are applied to large networks, then complete optimization can be achieved, but the computational time and resources become prohibitively large
Solution Approach 1:
The patent performs preliminary action by pre-determining the operating sequence and rank assignments before real-time operation. This advance planning allows the system to prepare optimized sequences that can be executed efficiently during real-time operation, significantly reducing the computational time required during actual network operation while maintaining high optimization quality.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
One aspect of the invention relates to a method for determining an operating sequence of at least one set of generator sets in an electrical network comprising the following steps: - Assigning a rank to each generator set; - From a forecast load curve, determining a minimum number of generator sets; - For each generator set having a rank less than or equal to the minimum number of generator sets, assigning the "on" state to the generator set in the operating sequence for each time interval;- For each generator set with a rank higher than the minimum number of generator sets, assign, in ascending order, for each time interval of the operating sequence, the state "on" or "off" to the generator set, and for each generator set to which the state "on" is assigned at a given time interval of the operating sequence, assign an operating power, by optimizing a cost function over the period with a constraint that the number of time intervals over the entire period for which the state "on" is assigned to the generator is greater than or equal to the number of time intervals over the entire period for which the state "on" is assigned to each generator set with a rank higher than the generator.