Search

The rising demand for energy and an still increasing world population are prompting the need to further improve power supply services upon a “zero interruption” target. This, combined with the process of shifting fossil fuel usage toward electrification, may escalate the economic and social impacts caused from interruptions. In this context, scheduled inspection plays a major role to manage risks. The challenge, however, is the trade-off of low O&M costs and high reliability. To solve that, this paper proposes a reliability-centered maintenance approach that combines heuristic and genetic algorithm (GA). The heuristic, based on benchmark practices, determines the cumulative total line length to be inspected to reach a targeted reliability level. The GA, on the other hand, yields a priority order list to portions of the grid to be inspected. The methodology relies on corporate systems historic data from outage management system, supervisory control and data acquisition, geographic information system and enterprise resource planning, typically available on power utilities. Tests are performed to a real 13.8 kV network in the South West of Brazil. Results demonstrate that the proposed methodology can significantly improve the decision-making by prioritizing feeder segments with higher load density, number of interruptions and voltage/current problems, up to a preset target, thus, improving reliability while limiting O&M costs.

During power outages in power distribution systems, a framework to detect and locate faults and broken cables is required to promptly restore the power supply and, therefore, to mitigate continuity indexes degradation. In developing countries, such as Brazil, customers’ phone calls are usually the only way for the power distribution utility to become aware of a power outage. Researches on this matter have proposed meta-heuristics, artificial intelligence, and traveling waves-based methodologies. Currently, they are mostly based on a final specific technology, restricted to some monitoring devices. Such condition makes the proposed solutions unfeasible for immediate application to Brazilian distribution systems. This work presents the development of a robust Fault and Broken Cable Location framework (FBCL), successfully deployed at a Brazilian power distribution network. Two major guidelines are considered: (1) use of existing devices’ records through corporate systems integration and (2) aggregation of innovative smart devices devised under the R&D scope and destined to expand monitoring capability. Aside from assisting the distribution utility in locating conventional low-impedance faults, the FBCL framework can also locate high-impedance faults with broken cables, which is a major contribution to mitigate safety issues in distribution networks. Based on the case studies conducted, the FBCL accurately located faults and broken cable events, considering multiple combinations of monitoring conditions.

The recent technological innovations, related to advanced measurement and automation infrastructures, and even sophisticated computational intelligence mechanisms, create opportunities to improve the operational efficiency of electrical systems and power quality indicators within the smart grids context. Aiming to this purpose, distribution companies use supervision, monitoring, control, management and georeferencing systems and have invested in automation and modernization projects of their electrical grids. Thus, this paper presents an architecture deployment for advanced distribution automation (ADA) applications, which includes state estimator, Volt/VAr control and FLISR (fault location, isolation and service restoration) tools and an interoperability bus that integrates the corporate systems with the operational management. The developed applications were tested and applied for real networks, simulated in a computational environment and integrated with the Smart Grids NAPREI/USP Laboratory environment. The results also demonstrate the processes of integrating ADA applications into a commercial service bus, which communicates the company’s legacy systems, and a new contribution with information for advances in the distribution automation area.

This paper presents a methodology for distribution systems planning considering operational performance and power quality indices. This methodology uses real data from the power distribution company’s legacy systems, such as GIS, SCADA and OMS, and reunites a range of tools used in the planning process, including planning studies’ automatic cost calculation, based on real modular costs used by the company. At the last stage of the planning process, investment optimization is addressed by a genetic algorithm considering multiple regions of study. The proposed methodology is applied to a case study, where all stages of the planning process are covered for five real distribution systems planning projects divided in three different regions of study, composed by eight substations and their respective 74 power distribution feeders, with 299,982 consumers and 3,689.35 kilometers of primary distribution networks. The proposed approach provides significant improvements to the current planning process adopted by the company and assists planning engineers in the decision-making of investment in distribution systems planning studies.

In a stochastic-flow network, a (d, T, b, P1, P2)-MP is a system state vector for which d units of flow can be transmitted through two separate minimal paths (SMPs) P1 and P2 simultaneously from a source node to a sink node satisfying time and budget limitations Tand b, respectively. Problem of determining all the (d, T, b, P1, P2)-MPs, termed as the (d, T, b, P1, P2)-MP problem, has been attractive in reliability theory. Here, some new results are established for the problem. Using these results, a new algorithm is developed to find all the (d, T, b, P1, P2)-MPs, and its correctness is established. The algorithm is compared with a recently proposed one to show the practical efficiency of the algorithm.

Advances in communications technologies, data processing and storage benefit power distribution utilities, allowing them to enhance the use of data provided by field monitors, remote-controlled switches and smart meters. Today, utilities can gather a variety of data regarding the power grid operation, such as customers demands, alarms and measurements, taking steps toward the Smart Grids. An interoperability bus (IB) can provide those data to any other corporate system, allowing one to develop power grid operational tools executed by distribution management systems (DMS). In this context, the present paper proposes a new fault location methodology for real power distribution networks, that resorts to data provided by an IB, such as alarms from protection relays and fault indicator sensors, and measurements from power quality monitors and smart meters. The methodology can be implemented in the DMS level and is based on evolutionary strategies, which is responsible for estimating the exact location of faults in the MV level of power distribution networks. The effect of the availability of data as alarms and measurements is assessed, considering a real 319-km-long power distribution feeder. Test results obtained from 113 short-circuit cases have indicated that the locating error is inferior to 2.9%.

The present paper presents an alternative methodology to determine frequency-dependent network equivalents from the frequency response of electric power networks. According to this methodology, one can substitute part of the power network through a simplified electric circuit, reducing the number of elements to be modeled without introducing any inaccuracy to the precise evaluation of large power networks. The main challenge relays on determining a rational function capable of producing the same frequency response characteristic of the original circuit that would be replaced by the equivalent. The rational function determines the impedance to be used in the network equivalent. The methodology is based on Evolutionary Algorithms (EAs). Basically, different set of values for the rational function parameters are evaluated. During the convergence process, the EAs modify the values for the function’s parameters suitably, reducing the RMS error between the original frequency response and the one produced by the equivalent circuit. The results obtained through this methodology were compared with those obtained through the “Vector Fitting” methodology, which is the methodology commonly used to solve similar problems.