Voltage Sensor Placement in Electrical Networks Under Sensor Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing the placement of voltage sensors in electrical distribution networks are either time-consuming, non-optimal, or fail to account for the limited number of sensors available, leading to suboptimal state estimation and voltage control.
Innovation Solution
A genetic algorithm-based method for optimizing sensor placement, which involves generating a population of solutions, evaluating their performance, selecting the best solutions through tournaments, crossing over genes to create new solutions, and mutating genes to explore the solution space, ultimately placing sensors at positions determined by the predominant gene with the maximum evaluated performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of sensors in the network is increased, then the accuracy of voltage estimation is improved, but the cost of instrumenting the network increases
Solution Approach 1:
The patent changes the parameter of sensor placement configuration from random or expert-based to optimized positions determined by the genetic algorithm. This optimization of the spatial parameter arrangement maximizes the information gained from a limited number of sensors, thereby improving voltage estimation accuracy without increasing sensor quantity.
Solution Approach 2:
The patent replaces the mechanical approach of simply adding more sensors with an intelligent computational system (genetic algorithm) that optimizes the placement of existing sensors. This substitution transforms the problem from a quantitative solution (more sensors) to a qualitative solution (smarter placement).
2Reliability
If expert opinion is used for sensor placement, then electrotechnical knowledge is utilized, but the method is time-consuming and not optimal
Solution Approach 1:
The patent implements a self-service system where the genetic algorithm automatically determines optimal sensor placements without requiring expert intervention. The system uses computational evolution to self-optimize the placement configuration, eliminating the time-consuming manual process while maintaining or improving placement quality.
Solution Approach 2:
The patent substitutes the manual expert-based method with an automated genetic algorithm system. This replacement eliminates human time investment while using computational power to explore and evaluate numerous placement configurations, achieving both speed and optimality.
3Reliability
If iterative methods are used to determine sensor positions, then reliability of state estimation is guaranteed, but the search process is time-consuming
Solution Approach 1:
The patent employs periodic iterative generations of the genetic algorithm, where each generation systematically evaluates and improves sensor placements. This structured periodic iteration maintains reliability through systematic exploration while improving productivity by using parallel evaluation and efficient convergence mechanisms inherent in genetic algorithms.
Solution Approach 2:
The patent introduces dynamics into the optimization process by allowing the sensor placement configuration to evolve and adapt across generations. The genetic algorithm dynamically adjusts placements based on performance feedback, enabling the system to converge to reliable solutions more efficiently than static iterative methods.
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
A method for optimizing the positioning of sensors configured to provide measurements of electrical variables in an electrical network comprising a set of nodes comprises: a. a random generation (E1) of a first generation population of solutions corresponding to a set of possible positions of the sensors within the set of nodes of the network, a solution being associated with a vector, called gene, and b. at least one iteration of reproduction of the population, comprising a selection (E3) by tournaments of solutions within the population, by random drawings of pairs of solutions in the population and selection of a solution having the best performance within each pair drawn randomly.At the end (E6) of the reproduction iteration(s), the method comprises a placement (E7) of the sensors within the network at the positions determined by a predominant gene associated with a maximum evaluated performance within the solutions of the last generation population.