Data Driven Modelling of Nuclear Power Plant Performance Data as Finite State Machines

Abstract

Accurate modelling and simulation of a nuclear power plant are important factors in the strategic planning and maintenance of the plant. Several nonlinearities and multivariable couplings are associated with real-world plants. Therefore, it is quite challenging to model such cyberphysical systems using conventional mathematical equations. A visual analytics approach which addresses these limitations and models both short term as well as long term behaviour of the system is introduced. Principal Component Analysis (PCA) followed by Linear Discriminant Analysis (LDA) is used to extract features from the data, k-means clustering is applied to label the data instances. Finite state machine representation formulated from the clustered data is then used to model the behaviour of cyberphysical systems using system states and state transitions. In this paper, the indicated methodology is deployed over time-series data collected from a nuclear power plant for nine years. It is observed that this approach of combining the machine learning principles with the finite state machine capabilities facilitates feature exploration, visual analysis, pattern discovery, and effective modelling of nuclear power plant data. In addition, finite state machine representation supports identification of normal and abnormal operation of the plant, thereby suggesting that the given approach captures the anomalous behaviour of the plant.

Simulation of human risk scenarios for nuclear power plant operation based on finite state machine

“…Considering the human factors domain in the system safety analysis, the function-based analysis method should be proposed. Therefore, the states of the physical part of the system can be analyzed by utilizing an FSM, and the transitions between states can be determined to obtain the physical process of NPP power generation based on state transitions [7][8]. Meanwhile, the operator's behaviors can be described by employing FSMs, thus making it possible to overcome the mismatch between the continuous nature of the physical process and the discrete nature of the human cognitive process.…”

An integrated method for the leakage fault mode diagnosis and life prediction of the reactor coolant pump

“…Data-driven model-based methods utilize the historical or current data of an object under certain functional constraints to establish a type that can approximate the implicit mapping mechanism between object data and lifespan for prediction [ 14 ]. The main steps of data-driven methods are data preprocessing, feature extraction, feature selection, model selection, and model evaluation [ 15 – 17 ]. Data-driven model-based methods mainly includes methods based on statistical regression [ 18 ], similarity [ 19 ], and the stochastic process [ 5 , 20 ].…”