Power Network Topology Estimation Using Current Sensitivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying the topology of electric power networks are inefficient due to the need for accurate network parameters, computationally intensive calculations, and sensitivity to noise, especially in low-voltage networks where distribution system operators lack up-to-date knowledge, and require extensive measurement data and deployment of devices across all nodes.
Innovation Solution
A method that uses timestamped current measurements from a limited number of metering units to estimate current sensitivity coefficients, allowing for the inference of the network's incidence matrix without requiring knowledge of network parameters or extensive device deployment, and is robust to noise, using a linear system of equations for efficient computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If measurement data from metering infrastructures is used for topology identification, then the method is applicable to distribution networks, but the results can be imprecise depending on the length of lines connecting the metering units
Solution Approach 1:
The patent replaces traditional voltage-based measurement approaches with current-based measurements from metering units. By using current measurements and calculating current sensitivity coefficients through statistical estimation, the method identifies topology without being affected by line length variations that plague voltage-based methods. This substitution of the measurement fundamental (from voltage to current) resolves the contradiction between broad applicability and precision.
Solution Approach 2:
The patent changes the measurement parameter from voltage to current. By measuring current at metering units and computing current sensitivity coefficients, the system achieves topology identification that is independent of line length. This parameter change allows the method to work accurately across distribution networks with varying line lengths, simultaneously improving both adaptability and precision.
2Adaptability or versatility
If optimization based approaches are used for topology identification, then the method can handle complex networks, but it is computationally intensive to solve
Solution Approach 1:
The patent substitutes optimization-based computational approaches with a direct statistical estimation method. By calculating current sensitivity coefficients through statistical relationships rather than solving optimization problems, the system achieves topology identification for complex networks without the computational burden. This replaces the computational mechanism from iterative optimization to direct statistical computation.
Solution Approach 2:
The patent uses a computationally efficient method that does not require extensive computational resources. By using statistical estimation of current sensitivity coefficients rather than optimization algorithms, the approach provides a 'cheap' computational solution in terms of processing power and time, while still handling complex network topologies effectively.
3Measurement precision
If all loads in the network are metered, then complete topology information can be obtained, but the deployment cost and complexity increase
Solution Approach 1:
The patent applies partial action by using current measurements from a subset of metering units rather than requiring all loads to be metered. The method calculates current sensitivity coefficients based on available measurements and infers topology information without needing complete network-wide metering. This partial measurement approach achieves sufficient topology identification accuracy while reducing deployment complexity and cost.
Solution Approach 2:
The patent uses current sensitivity coefficients as an intermediary to infer topology from limited current measurements. Instead of requiring direct measurements from all network points, the method uses statistical relationships in current measurements as intermediaries to deduce the complete topology, thereby reducing the need for extensive metering deployment.
4Loss of information
If correlation analysis between synchronized energy measurements is used, then the interrelationship between monitoring devices can be identified, but energy losses are a source of noise and error in the estimation
Solution Approach 1:
The patent substitutes energy loss-based correlation analysis with current-based statistical estimation. By using current measurements and calculating current sensitivity coefficients, the method identifies relationships between monitoring devices without being affected by energy losses that introduce noise and errors in power-based methods. This substitution eliminates the harmful effect of energy losses.
Data Source
AI summary
Disclosed is a method for identifying the topology of an electric power network allows for the automatic and efficient identification of network topology without the knowledge of network parameters. The method is based on the estimation of mutual current sensitivity coefficients and on an algorithm to obtain the network incidence matrix from the estimated sensitivity coefficients. This algorithm is based on the general assumption that the metering unit that is the most sensitive to the variations of the measured current in a branch connected to a particular node is the metering unit that is arranged at the physical node immediately upstream form the particular node. The method effectively considers the presence of noise in the measurements and its time correlation.