Artificial intelligence launched at T Plus CHPP

2023: Artificial Intelligence Launch

T Plus has deployed a predictive analytics system in 7 branches located in the Central, Volga and Ural Federal Districts, based on the Russian digital production management platform ZIIoT, developed by Zyfra Group of Companies, which announced this on October 24, 2023. Predictive analytics helps minimize the risks of equipment breakdowns and outages. Thanks to artificial intelligence, it has already been possible to prevent three incidents at the Akademicheskaya CHPP, which could total disable the station for 24 hours. As a result, the company would not have supplied 5.5 million kWh of electricity to the network and did not release 3360 Gcal of heat to consumers.

The basis of the solution was the production data management system developed during the project on the basis of the ZIIoT platform, which collects, processes, stores technological data and presents them in a single information space. On the basis of OMS, a predictive module has been deployed at power plants. The operation of the system has already begun at the Akademicheskaya CHPP. In other branches, mathematical models of equipment are being configured and trained. The developed system is designed to automate monitoring and diagnostics of malfunctions, deviations in the operating modes of power plants, as well as to predict the technical condition of gas and steam turbines, boiler units, auxiliary equipment and equipment of the CHPP electrical facilities. During the project, a diagnostic center was created to service the T Plus PJSC CHPP with the prospect of scaling at the infrastructure level of the entire company.

The prognostic module is based on reference models of equipment operation under different conditions and under different modes. Models are built using neural networks and the results of analyzing data on the operation of an energy facility and are constantly correlated with information received in real time. To do this, about 8500 different sensors are installed on the main (boilers, turbines) and auxiliary (for example, pumps) equipment of the thermal power plant, which analyze the state and operation of each element of the station. The indicators collected by the ZIIoT platform fall directly into the prognostic system. And she, in turn, signals the operational staff about different deviations.

The predictive analysis module monitors the dynamics of changes in the indices of the technical state of the units, the degree of deviation of the parameters of the units from the standard values. A sharp change in the technical condition index allows you to predict the approaching failure of the unit with a high probability, and exceeding the permissible deviation of certain parameters allows you to determine the node and the cause that causes the failure. The system helps to save time on identifying the problem and prevent breakdowns and shutdowns, - said Denis Ulanov, Deputy General Director - Chief Engineer of T Plus.

The created prognostic system is part of a large-scale project to digitalize the T Plus business. At the next stages, it is planned to develop a system for managing the production data of the heat unit, which will combine the management of the generation and transportation of heat energy, prognostics, digital counterparts of stations into a single centralized production management system, - said Boris Makevnin, director of digitalization and information technologies at T Plus, about the project.

During the project, we created an integration solution on the ZIIoT platform, solving the problem of choosing the most optimal CHPP management model and ensuring early detection of potential defects. The created predictive analytics system at T Plus is based on Russian software solutions, easily integrates with related IT systems of the enterprise and quickly scales to new objects. The solution developed jointly with T Plus specialists can be used as a reference industry standard, - commented Sergey Emelchenkov, General Director of Figures.