AI and machine learning in SCADA in SCADA systems - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the machine learning library commonly used for anomaly detection in SCADA data.

SCADA systems

import [1]

Drag options to blanks, or click blank then click option'

Atensorflow

Bsklearn

Cpandas

Dmatplotlib

Attempts:

3 left

💡 Hint

Common Mistakes

Importing visualization or data libraries instead of machine learning ones.

Using TensorFlow which is more complex for beginners.

The sklearn library provides tools for machine learning, including anomaly detection, which is useful in SCADA systems.

2fill in blank

medium

Complete the code to create a simple anomaly detection model using IsolationForest for SCADA sensor data.

SCADA systems

model = IsolationForest([1]=100, contamination=0.1)

Drag options to blanks, or click blank then click option'

An_estimators

Bmax_samples

Cmax_depth

Dlearning_rate

Attempts:

3 left

💡 Hint

Common Mistakes

Using parameters that do not exist in IsolationForest.

Confusing sample size with number of trees.

n_estimators sets the number of trees in the IsolationForest model, which helps detect anomalies in SCADA data.

3fill in blank

hard

Fix the error in the code to correctly fit the anomaly detection model on SCADA data stored in variable 'sensor_data'.

SCADA systems

model.fit([1])

Drag options to blanks, or click blank then click option'

Asensor_data.columns

Bsensor_data.values

Csensor_data.to_list()

Dsensor_data

Attempts:

3 left

💡 Hint

Common Mistakes

Passing only the values or columns instead of the full data.

Using methods that convert data to unsupported formats.

The fit method expects the data as a DataFrame or array-like. Passing sensor_data directly is correct.

4fill in blank

hard

Fill both blanks to create a dictionary of sensor names and their anomaly scores from the model predictions.

SCADA systems

anomaly_scores = {sensor: model.[1](data) for sensor, data in sensor_data.[2]()}

Drag options to blanks, or click blank then click option'

Adecision_function

Bitems

Cpredict

Dcolumns

Attempts:

3 left

💡 Hint

Common Mistakes

Using predict instead of decision_function for scores.

Using columns instead of items() to iterate.

decision_function gives anomaly scores, and items() iterates over sensor data columns and their values.

5fill in blank

hard

Fill all three blanks to filter sensor readings above a threshold and create a dictionary of sensor names and their filtered readings.

SCADA systems

filtered_readings = {sensor: readings[readings [1] [2]] for sensor, readings in sensor_data.[3]()}

Drag options to blanks, or click blank then click option'

A>

Bthreshold

Citems

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' for filtering.

Using 'columns' instead of 'items()' for iteration.

The code filters readings greater than a threshold and iterates over sensor data items.