基于改进序列模式挖掘算法的告警关联模型

在电力故障发生时，会产生大量的电力故障告警信息数据，如何从电力故障告警信息中挖掘出可靠的关联规则，对后续电力的调度运维有着重要的影响。广义序列模式（Generalized Sequential Pattern,GSP）算法通过增加时间上的约束条件提高算法的效率，适合应用于电力故障告警信息挖掘的场景。针对GSP算法中的关键参数多和不同的参数组合影响算法的准确性和可靠性的问题，将遗传算法与GSP算法相结合，自适应地得到一组较好的参数，将参数代入GSP算法，从而得到更加可靠的关联规则，以此来解决在电力故障告警信息应用中很难为不同的数据集找到合适的参数组合的问题。通过实例验证，电力故障告警信息数据应用遗传算法结合GSP算法能够有效地得到更加准确和可靠的计算结果。

Alarm association model based on improved sequence pattern mining algorithm

When a power fault occurs,a large amount of power fault alarm information data will be generated.How to mine reliable association rules from power fault alarm information has an important impact on the subsequent power dispatching operation and maintenance.The Generalized Sequential Pattern(GSP) algorithm can improve the efficiency of the algorithm by increasing the time constraints,which is suitable for the scenario of power fault alarm information mining.To address the problem that many key parameters and different parameter combinations in GSP algorithm affect the accuracy and reliability of the algorithm,the genetic algorithm is combined with GSP algorithm to obtain a better set of parameters adaptively.And the parameters are substituted into GSP algorithm to obtain more reliable association rules to solve the problem that it is difficult to find suitable parameter combinations for different data sets in power fault alarm information applications.It is verified by example that the application of genetic algorithm combined with GSP algorithm can effectively obtain more accurate and reliable calculation results in power fault warning information data.