Data-Driven Method to Detect the Events of Interest in an Electricity Market Payam Zamani-Dehkordi, Logan Rakai, and Hamidreza Zareipour Department of Electrical and Computer Engineering University of Calgary, Calgary, AB, Canada pzamanid@ucalgary.ca; l.rakai@ucalgary.ca; h.zareipour@ucalgary.ca Abstract—Participants in an electricity market expect to have a fair, transparent, and open competition. Independent System Operator (ISO) is responsible to monitor the market outcomes to investigate if they are consistent with fundamentals of electricity market. It can be an intensely time-taking process with high levels of computations in an electricity market with huge number of participants. Besides, a manual review of market operation may be deceitful as human is part of decision making process. If this anomaly detection procedure can be done automatically then it can be a great aid to market surveillance process for having an unbiased and prompt tool to monitor the market. In this paper, an anomaly detection algorithm is proposed to identify the events of interest in an electricity market. This algorithm provides the ISO with a tool to detect the instances in the electricity market which electricity price behavior deviates from normal expected regime. These anomaly hours then can be analyzed further in order to diagnose the reason. Index Terms—Electricity market, Anomaly detection, Market surveillance. I. I NTRODUCTION One of the chief duties of the ISO is to monitor market outcomes to ensure that the conduct of participants supports the fair, efficient, and openly competitive operation of the elec- tricity market. This involves actively monitoring the market in real time to ensure that observed outcomes are broadly consistent with market fundamentals and to identify potential anticompetitive conduct. When ISO relies on manual review of each hour of the market then this process was both labor intensive and reliant upon the judgment of the staff members conducting the review. As a solution, an anomaly detection algorithm is proposed in this paper as a computational method to flag hours for further review. Anomaly detection algorithm provides a reliable method of relating market fundamentals to market outcomes to highlight abnormal results. The algorithm is a fast and accurate tool for detecting events of interest in an electricity market which can be replaced with the manual review of each hour. Hawkins formally defined [1] the concept of an outlier as follows: ”An outlier is an observation which deviates so much from the other observations as to arouse suspicions that it was generated by a different mechanism.” Outliers are also referred to as abnormalities, deviants, or anomalies in the data mining and statistics literature. Basically, data has a normal model, and anomalies are recognized as deviations from this normal model. There are different categories of anomaly detection algorithms which have been used in literature. The most basic form of outlier detection is extreme value analysis of 1- dimensional data. These are very specific kinds of outliers, in which it is assumed that the values which are either too large or too small are outliers [2]. Such methods have often not found much utility in the literature for generic outlier analysis, because of their inability to discover outlier in the sparse interior regions of a data set. In probabilistic and statistical models, the data is modeled in the form of a closed form probability distribution, and the parameters of this model are learned. A downside of probabilistic models is that they try to fit a particular kind of distribution to the data, which may often not be appropriate for the underlying data. Regression model- ing are another category of anomaly detection algorithms. The derived residuals which are basically the error of the modeling will be used in conjunction with extreme value analysis in order to determine the underlying outliers. Finally, proximity- based models model outliers as points which are isolated from the remaining data. This modeling may be performed in one of three ways of cluster analysis, density-based analysis or nearest neighbor analysis [3]. There are many papers published in the field of power systems which utilize anomaly detection algorithms for dif- ferent purposes. In [4], [5] a regression-based anomaly de- tection methodology is proposed for the protection of cyber- physical systems. It is concluded that this technique could make a significant contribution to the security of electricity critical infrastructures. The study in [6] proposes a warning approach to identifying power quality problems and providing early warning prompts based on statistical anomaly detection algorithm. A regression-based anomaly detection algorithm is presented in [7], [8] which helps building facility engineers and property managers to achieve significant energy savings,