The operations and maintenance (O&M) of thermal power plants (TPPs) is essential to ensure efficient and uninterrupted operations. In view of the challenges in the coal-based power generation segment such as low fleet utilisation, ageing assets, rising fuel costs, tighter environment and safety regulations and competition from cheaper renewables, adopting efficient O&M strategies has become crucial. Under flexible operations of power plants, characterised by high ramping frequency and shutdowns, the components’ life could get adversely impacted, resulting in early life failure as compared to baseload operations. One of the traditional O&M practices adopted by developers is routine and capital maintenance of turbines to assess their actual condition and restore them to their desired condition. However, digital O&M solutions at the boiler, turbine and generator (BTG) level have gained traction in the past couple of years. The digitalisation of O&M generates plant-specific operational and contextual data, including operational history, unexpected degradation of plant equipment, etc.
O&M use cases
Generation forecast and optimisation
For a competitively bid asset, the spread between revenue and cost per unit varies based on the coal price and the overall surrender per cent (the gap between the availability as declared by a power plant and the actual demand of the power procurer). Value creation is maximised through optimum utilisation of the available plant capacity, be it for long-term power generation or for maximising returns on surplus capacity from liquid short-term markets. As a result, significant value can be unlocked simply by timing the planned maintenance correctly. This requires accurate forecasting of the cost drivers (coal prices and freight rates), demand (expected surrender per cent and market demand for short-term power) and technological factors (station heat rate impact of delayed maintenance). Digital interventions can help improve both forecast accuracy and the running of optimisation programmes.
Coal blending
Operationally, assets need to develop the capability of blending coal (either in the yard, or on a conveyor belt or through choice bunkering) to work within the constraints imposed by the boiler’s make (especially on sulphur, emissions and coal feed rates). Digital intervention can help drive coal blending on the conveyor in a cost-effective manner, by calibrating reclaimers through a feedback loop, allowing the blend ratio control in real time (with minimal capital expenditure). Once blending has been operationalised, an optimiser can help determine the ideal sourcing mix by taking into account different coal indices, freight rates from different sources, coal specifications and operating limits (gross calorific value, ash per cent, sulphur per cent) of the asset.
Optimising unit heat rate
This is a well-established use case adopted by most generating units in India. However, most assets limit the use of digital tools (such as PADO) to monitor performance, identify major deviations and prepare monthly MIS on loss reasons within BTG islands, that is, a form of post-mortem analysis on historical deviations, rather than proactive and prospective analysis of future deviations. Analytics on metrics impacted by controllable parameters (for example, main steam temperature impacted by superheater spray) can help identify positive spray biases in units, air ingress or steam/ water leakage areas, which can easily be addressed through operator intervention or minor maintenance activities.
Predictive maintenance
Predictive maintenance to minimise asset downtime and maintenance costs is another key area where digital can drive significant efficiencies. The potential benefits of predictive maintenance are only limited by the number of critical parameters within a power plant that are monitored on a real-time basis. Thus, the key requirement to get this started is the installation of a sufficient number of sensors across the cycle, setting up centralised monitoring and analysis centres and hiring analytics resources that can perform time series data analysis and modelling. Once set up, this provides a threefold benefit to the asset owner, such as reducing the downtime by predicting errors and faults through the use of big data and pattern recognition software; the ability to shift from an interval-based maintenance to a condition/risk-based maintenance strategy; and driving manpower efficiency, and overcoming talent availability issues in remote locations by centralising monitoring of the entire fleet.
Summing up
A number of new use cases are currently in the pilot phase globally and may soon become commonplace in generation assets. Some of these are the use of drones for surveillance and coal inventory reconciliation, use of augmented reality to reduce critical breakdown maintenance lead time by leveraging remote expertise, 3D printing of spare parts and use of robots in stores management. Almost half of the downtime of TPPs is due to poor O&M. O&M practices that involve real-time asset monitoring and data and predictive analytics can help ensure greater efficiency and higher cost savings.