How Do I Become an Electroneurodiagnostic Technologist?

Currently, the widely used intelligent technology-based fault diagnosis methods for power grids include expert systems, artificial neural networks, Bayesian networks, optimization technologies, support vector machines, fuzzy set theory, Petri nets, information fusion technologies, and multi-agent technologies. This article briefly introduces the basic concepts of these intelligent methods, the current research status in the field of power grid fault diagnosis, and explains their respective characteristics and shortcomings from a practical perspective, as well as their respective future developments. Starting with important issues and future development trends in this area.

Grid fault diagnosis is to identify the faulty component by measuring and analyzing the electrical quantities such as current and voltage in the power grid after the fault, as well as the switching quantity changes of protection and circuit breaker actions. A good diagnosis strategy is of great significance to shorten the failure time and prevent the accident from expanding. When a fault occurs, a large amount of fault information collected by the monitoring system flows into the dispatch center. The diagnosis methods based on traditional mathematical models can no longer guarantee the accuracy and speed of diagnosis. In contrast, based on intelligent technology The diagnostic method has obvious advantages. Intelligent methods can simulate, extend and extend human intelligent behavior, make up for the lack of mathematical model diagnosis methods, and open up new approaches for the field of power grid fault diagnosis. Therefore, the development of fault diagnosis methods from traditional technology to intelligent technology is the focus and hotspot of future research in this field ^[1]

An analytical model that expresses the relationship between electrical equipment, protection, and circuit breakers needs to be constructed in power grid fault diagnosis. Some literatures have studied the timing characteristics of alarm information during faults, and combined with the concept of dynamic correlation paths to build an analytical model for fault diagnosis of power grids, which can more clearly describe the action logic and time sequence relationship of protection and circuit breakers in power grid protection configurations, which can be better To reflect multiple complex faults. There is also literature that studies the fault hypothesis of protection and circuit breaker misoperation and refusal behavior, and establishes a more complete diagnostic model, which can analyze not only protection and circuit breaker misoperation and refusal behavior, but also identify missed or false alarms. Information. Some studies have proposed a complete analytical model for power grid fault diagnosis, which expresses the analytical relationship between the protection configuration and the circuit breaker operation rules by establishing logical constraints, and completely preserves the protection operation status, circuit breaker trip status, and their misoperation and refusal. The coupling relationship formed by rule analysis improves the robustness and fault tolerance of the model. The above model is built for the relay level, and cannot fully reflect the logical relationship at the protection device level during application. The relay protection of the grid operation is configured according to the complete set of devices, and the operation consistency between the devices and between devices is consistent due to operating requirements and circuit connections. ^[5]

At present, the main problems faced in the field of power grid fault diagnosis are: (1) various diagnostic methods have poor fault tolerance when dealing with uncertain and incomplete information, and until now, no clear solution has been given to this problem; (2) As mentioned earlier, these intelligent methods have their own application limitations and deficiencies, and in current practical applications, most power grid fault diagnosis is based on only one intelligent method; (3) the operation mode of the power grid and the network topology Changes in the fault have a significant impact on the results of fault diagnosis; (4) The practical research on intelligent fault diagnosis of the power grid is not enough.

Starting from the current problems in the field of power grid fault diagnosis, future research focuses can be divided into the following aspects:

(1) Research on diagnostic methods based on the fusion of multiple intelligent methods. At present, in the practical application of power grid fault diagnosis, an intelligent method is mostly adopted, and among them, an expert system and an optimization technology are mostly used. Integrating a variety of intelligent technologies to learn from each other's strengths and introduce the latest technology in the field of power grid fault diagnosis in a timely manner is an important trend for the development of future fault diagnosis.

(2) Research on diagnostic methods based on multi-data source information fusion technology. Currently, most of the intelligent technologies used in power grid fault diagnosis systems use switching information. In comparison, in terms of accuracy and fault tolerance, electrical quantity has greater advantages. Fusion of the switching and electrical information of different data sources and making full use of the fault information of multiple data sources can make the diagnosis results more accurate.

(3) Research on fault diagnosis based on distributed intelligent technology. The distributed fault diagnosis method can be used for distributed fault diagnosis after partitioning the large power grid, which effectively solves the problem of fault diagnosis for the large power grid. In 2007, experts from the Chinese Academy of Electric Sciences used the Bayesian network diagnosis method to study MAS collaborative diagnosis problems using a large-scale system containing uncertain fault information as a platform, and achieved good diagnostic results.

(4) Research on the practicality of online power grid fault diagnosis. At present, experts and scholars at home and abroad have made many achievements in the research of power grid fault diagnosis theory, but they are still lacking in practical research. Therefore, in the future research, how to apply theory to practice from the perspective of practicality is still an important research topic. The fault diagnosis pre-processing function is the entrance of the entire practical fault diagnosis system and provides guarantee for subsequent operation. Therefore, the research on the fault diagnosis pre-processing function is also an important step to promote the practical application of power grid fault diagnosis.

Intelligent diagnosis of power grid faults is a new idea in the development of this field, and a lot of substantial results have been achieved. Summarizes the intelligent methods of power grid fault diagnosis that are currently attracting attention, and introduces the characteristics of these methods, the existing problems and recent research results, and points out the problems facing the current field of power grid fault diagnosis, such as insufficient practicality. It handles uncertain information with poor fault tolerance, etc. Finally, it discusses the future development trend of power grid fault diagnosis. Some of these intelligent methods are still in the theoretical stage and have their own disadvantages. Therefore, they need to be continuously improved in practical engineering applications to improve the intelligence level of power grid fault diagnosis.

Judgment of power grid accidents is a prerequisite for accident handling, and it is of great significance to improve the accuracy and speed of diagnosis. With the expansion of the scale of the power grid, it is necessary to equip dispatchers with high-intelligence, high-speed, high-quality real-time fault diagnosis and diagnosis systems to assist dispatchers in real-time adjustment and control of the grid operation.