Ensuring Diagnosability of the Technological Process with a Minimum Number of Sensors Based on the Entropy Criterion

Abstract

When automating a technological process, it is necessary to determine the number and location of sensors (sensor network) in such a way as to ensure the possibility of effective diagnostics of the process and at the same time to reduce the cost of measuring equipment. To solve this problem, it is proposed to use the entropy criterion. For this, an analysis is carried out and possible faults and failures that may arise in the process are identified. Next, the changes in entropy when a fault or failure occurs for each of the possible sensors, are considered. Those sensors that provide the greatest decrease in entropy (increase in information) are selected.