Operation Of Power Plants Research Articles

Thermal energy storage integration for increased flexibility of a power plant with post-combustion CO2 capture

Flexible operation of thermal power plants will become increasingly relevant in the coming years. This work evaluates the effect of integrating a steam accumulator into a 598 MW supercritical coal-fired power plant with moving bed temperature-swing adsorption CO2 capture. Charging the accumulator with reheat steam from the turbine train can reduce the net power output by up to 1.4 % (8.2 MW) for around 200 min. Small charging flow rates are recommended to maximize the final pressure of the accumulator. Covering the sorbent regeneration duty and meeting the demand of two feedwater heaters were considered as alternatives for discharging of the accumulator. Thermal storage discharging was found to give relative power plant load increases between 1.7 and 11.2 % (10.2–66.9 MW) for up to 37.5 min, which exceeds the requirement for primary reserve. Increases in the electrical energy output of 10.5–11 MWh were achieved. The flexibility modes studied in this work are compatible with high CO2 recovery rates.

Joint Modeling and Operational Optimization of a Reverse Osmosis-Mechanical Vapor Recompression System for Coal-Fired Power Plant Wastewater.

The operation of coal-fired power plants generates a large amount of wastewater. With the issuance of increasingly strict drainage standards, the cost of wastewater treatment is increasing, and the need to reduce the cost of wastewater treatment is becoming increasingly urgent. Thus, based on the principles of reverse osmosis (RO) and mechanical vapor recompression (MVR) in wastewater treatment, the operational optimization of an RO-MVR joint system was studied in this work with the consideration of reducing the operating costs of wastewater treatment under given operational conditions. Firstly, based on the basic principles of RO and MVR, corresponding mechanism models were established and their accuracy was verified. Then, an economic model of the RO-MVR joint system was established, with the goal of minimizing the water production unit price and daily operating costs of the joint system for optimization analysis. Finally, we analyzed the cost and water production performance of the RO-MVR joint system before and after optimization under different operating conditions. The results show that this optimization based on the RO-MVR joint system will reduce the unit price of water production to 3.16 CNY/m3, with the daily operating costs being decreased by 22% compared to before optimization. This result helps to reduce the cost of zero-discharge wastewater treatment in coal-fired power plants.

ISSUES OF WATER USE IN THERMAL POWER PLANTS. PROBLEMS AND SOLUTIONS

The article considers the most effective methods of water treatment for thermal power plants. Water treatment is one of the most important processes in the operation of thermal power plants (TPP). The quality of water treatment for use in boiler units of thermal power plants is of fundamental importance to ensure the efficiency and durability of process equipment. This article reviews the main aspects of water treatment in TPP plants, including treatment methods, water quality characteristics and standards, and the impact of water treatment on the overall energy efficiency and safety of power generation in TPP plants.

Sorptive Removal of 133Ba from Aqueous Solution Using a Novel Cellulose Hydroxyapatite Composite Derived from Cigarette Waste

133Ba is a hazardous radionuclide generated during the operation of nuclear power plants. 133Ba needs to be removed from waste solutions because its half-life (10.55 years) and gamma energy pose a significant threat to human health. Cigarette butt (CB) is a waste that causes serious environmental problems. Various types of adsorbent materials are prepared based on the cellulose in its structure. The focus of the present study is to synthesize a novel composite material derived from CBs and to investigate its 133Ba removal capability. Microfibrillated cellulose (MFC) obtained from CBs was modified with hydroxyapatite (HAp) via the co‐precipitation method and converted into a composite adsorbent for the removal of 133Ba ions. Response surface methodology (RSM) based on Box‐Behnken Design (BBD) was employed for the examination of process variables such as initial pH, metal concentration, and adsorbent amount on 133Ba sorption. XRD and FTIR data confirmed the successful isolation of cellulose and the modification of the cellulose surface with HAp. The model F-value (100.04) and R2 (0.99) suggested that the proposed model was significant. Optimum conditions were determined as initial pH of 8, contact time of 134 min, and concentration of 0.01 mol/L, and the barium sorption capacity of MFC‐HAp was found to be 0.75 mmol/g under these conditions. The maximum monolayer barium sorption capacity was determined to be 2.92 mmol/g. Combining cellulose and HAp to be a novel composite adsorbent is useful for reusing CB waste and promising for removing 133Ba ions from aqueous solutions.

Use of Data Analytics for Power Plant Operation Optimization

Optimized operation and maintenance of a power plant to improve the KPIs is the key to success. Forced outages not only make the unit unavailable but also increase the number of start-ups and hence high cost and deteriorated plant life. Fault diagnostics and prediction of future behavior of equipment by data analytic techniques will be very handy in improving reliability. On the other hand, the cost of generation depends upon how efficiently the plants are operated. Any deviation from the optimized design point, results in losses and hence as far as possible the deviations should be minimized. However, the deviations are unavoidable due to various reasons. Along with efficient operation, Environmental compliance, and flexibility is also to be ensured to meet the statutory requirements. With the advancement of computing techniques, data analytics, and Artificial Intelligence are being used for the optimization of many industrial processes. Given the background, the objective of this paper is to study and suggest how Data Analytics can be utilized for the optimization of power plant operations to improve reliability, efficiency, and flexibility, and minimize the impact on the environment. Many statistical methods are available to identify the hidden pattern in the data which can be used for optimisation of processes. AI is used for process control, diagnosing faults, and predicting of future behaviors so that advanced action can be taken to avoid surprises. The domain expertise along with data analytic methods can be utilized to find solutions to a variety of problems. In this paper, a comprehensive data analytic tool with four modules i.e., “Efficiency optimization”, “Plant Health monitoring and reporting”, “Optimization of life consumption” and “Environment protection” has been conceived for application. These Modules work in an integrated manner and shall monitor, optimize, control, and report/advise. The tool shall have a digital replica of the individual equipment for simulation individually as well as in combination with other related equipment for whole plant performance prediction and diagnosis. The replica shall use its database for machine learning and for running of diagnostic process.

Reliability Assessment of Power Plant Operations through Boolean Function Expansion and Mean Time to Failure Analysis

This paper presents an investigation into the reliability assessment of power plant operations using Boolean Function expansion and Mean Time to Failure (MTTF) analysis. The study focuses on a complex system comprising two power generators within a power house. The primary objective of this system i. This paper presents an investigation into the reliability assessment of power plant operations using Boolean Function expansion and Mean Time to Failure (MTTF) analysis. The study focuses on a complex system comprising two power generators within a power house. The primary objective of this system is to ensure uninterrupted power supply from the power house to critical consumers via an output main switch. Reliability calculations are conducted considering failure times of various components, including cables, generators, and main switchboards, assuming arbitrary distribution patterns. Additionally, MTTF for the system under an exponential failure time distribution is determined. Graphical representations are employed to illustrate the utility and efficacy of the proposed models to ensure uninterrupted power supply from the power house to critical consumers via an output main switch. Reliability calculations are conducted considering failure times of various components, including cables, generators, and main switchboards, assuming arbitrary distribution patterns. Additionally, MTTF for the system under an exponential failure time distribution is determined. Graphical representations are employed to illustrate the utility and efficacy of the proposed model.

Investigation of Failures during Commissioning and Operation in Photovoltaic Power Systems

Considering global warming and environmental problems, the importance of renewable energy sources is increasing day by day. In particular, the effects of wind and solar power, which are variable renewable power sources, on the power system necessitate their evaluation in terms of the reliability of the power system. Photovoltaic panels, which enable the conversion of solar power into electrical power with semiconductors, have started to take an important place in global energy investments today. Photovoltaic power plants increase the demand for this energy source with continuous energy conversion depending on sunshine duration and radiation intensity. Among the renewable energy sources, the most easily utilized energy source, regardless of geographical conditions, is the sun. To prevent the energy production of PV power plants from being interrupted, it is necessary to address and analyze all kinds of faults that will affect power production in order to increase the reliability of the system. Academic studies in this field are generally grouped under two topics: classification of faults or modeling of electrical faults. Based on this, in this study, the problems that occur during the installation and operation of photovoltaic systems are classified, and the relevant faults are modeled and simulated in MATLAB Simulink version 23.2 (R2023b). Thus, a scientific approach to the problems of photovoltaic power plant operating conditions has been gained, which will be the basis for academic studies.

Redispatch Model for Real-Time Operation with High Solar-Wind Penetration and Its Adaptation to the Ancillary Services Market

Modern electrical power systems integrate renewable generation, with solar generation being one of the pioneers worldwide. In Latin America, the greatest potential and development of solar generation is found in Chile through the National Electric System. However, its energy matrix faces a crisis of drought and reduction of emissions that limits hydroelectric generation and involves the definitive withdrawal of coal generation. The dispatch of these plants is carried out by the system operator, who uses a simplified mechanism, called “economic merit list” and which does not reflect the real costs of the plants to the damage of the operating and marginal cost of the system. This inefficient dispatch scheme fails to optimize the availability of stored gas and its use over time. Therefore, a real-time redispatch model is proposed that minimizes the operation cost function of the power plants, integrating the variable generation cost as a polynomial function of the net specific fuel consumption, adding gas volume stock restrictions and water reservoirs. In addition, the redispatch model uses an innovative “maximum dispatch power” restriction, which depends on the demand associated with the automatic load disconnection scheme due to low frequency. Finally, by testing real simulation cases, the redispatch model manages to optimize the operation and dispatch costs of power plants, allowing the technical barriers of the market to be broken down with the aim of integrating ancillary services in the short term, using the power reserves in primary (PFC), secondary (SCF), and tertiary (TCF) frequency control.

Reducing the effect of engine failure on the aerodynamic performance of the light transport aircraft model

Recently, aircraft engine manufacturers have shown increased interest in developing hybrid powerplants, which are a combination of gas turbine engines (GTE) with electric motor-generators. The use of the hybrid powerplant makes it possible to increase the fuel efficiency of an airplane, as well as to create new configurations with improved aerodynamic and thrust characteristics. The fuel efficiency improvement is achieved as a result of optimizing the powerplant operation mode to meet the cruising flight requirements, compensating insufficient power during the takeoff and go-around procedures by activating battery-powered electric motors. The creation of new configurations with improved performance can be ensured due to the synergetic effect of the propeller-airframe interaction. Successful flight tests of the hybrid powerplant prototypes in light aircraft configurations allow us to rely on their possible application in the future regarding the projects of new propeller-driven aircraft. The potential benefits of using new powerplants on local airlines can lead to both fuel savings and carbon emission reduction. Short-term maintaining a safe flight mode is also practical in case of one engine failure when using multiple power sources. The power, generated by an electric generator connected to the running engine, can be used both for the electric motor drive of the tip propellers and for rotating the thrust producer of the failed engine. The paper presents the study results of the critical engine failure effect on the aerodynamic performance of the light transport aircraft model obtained as under available electrical transmission as under non-available one between a running and a failed engine. Experimental studies were carried out in a low-speed wind tunnel T-102 TsAGI. The simulation of the electric transmission operation was carried out by setting the operation mode of two power-plant simulators corresponding to the half value of the load factor of one engine propeller Bo in the take-off mode.

Assessment of radiation environmental risks for fish fauna of the cooling pools of a number of operating nuclear power plants

Relevance. The need to assess radiation risks for the main types of fish in pools that are under the constant influence of the nuclear power plants. Aim. To assess the radiation environmental risks for the ichthyofauna of cooling pools of the operating nuclear power plants, to analyze the sources and features of formation of the radiation risks at different stages of their operation. Objects: monitoring studies of radiation environmental risks for reference fish species in cooling reservoirs of the Leningrad, Beloyarsk and Novovoronezh NPP. Methods. Calculation of the exposure dose rate and quantitative assessment of the risk of exposure to technogenic radionuclides of the ichthyofauna of the cooling pools of the Beloyarskaya, Leningradskaya and Novovoronezhskaya NPPs were carried out using the international computer complex of the updated version of ERICA Tool 2.0. Results. The paper demonstrates that the radiation ecological risks for the ichthyofauna in the cooling pools of the operating nuclear power plants are caused by the combined effect of radioisotopes of the different origin. The features of the formation of the radiation risks for each pool are specific and depend on the current radiation situation. The cooling pool of the Beloyarskaya NPP is characterized by a pronounced dependence of the radiation load on the types of operating power units. The ichthyofauna experienced the greatest radiation risks during the operation of the first two power units with thermal reactors AMB-100 and AMB-200. The decommissioning of the first power units and the construction of the units with the fast neutron reactors (BN-600 and BN-800) led to a significant reduction in the radiation load on the ichthyofauna of the cooling pool. At present, the radiation risks are caused mainly by the influence of 90Sr, which is almost not discharged from the Beloyarskaya NPP into the cooling pools, but has a predominantly global origin. Therefore, the formation of the risks is more associated with atmospheric fallout of 90Sr, and not with the operation of the nuclear power plant. The radiation risks for the ichthyofauna of the Leningradskaya NPP cooling pool are caused by the combined effect of 137Cs as a result of the radiation accident at the Chernobylskaya NPP and radionuclides with induced activity (14C; 60Co; 3H 54Mn and 65Zn), the origin of which may be associated with the operation of the nuclear reactors of the NPP. The level of radiation risk for the fish fauna of the Novovoronezhskaya NPP cooling pool for 2009–2019 increased 2.2 times due to an increase in the content of 60Co in many components of the pool. As a result, 81.7% of the radiation risks in demersal fish species were formed by the influence of 60Co. It cannot be ruled out that the increase in the 60Co content in the cooling pool of the Novovoronezhskaya NPP is a consequence of radioactive contamination of the groundwater as a result of the 1985 accident. The ecological risks for the fish fauna in all cooling pools are two to three orders of magnitude lower than the maximum allowable hazard coefficient. Thus, the normal operation of the nuclear power plants does not create unacceptable radiation risks that pose a real environmental threat to the fish fauna of the cooling pools.

Use of biofuels in marine diesel engines for sustainable and safe maritime transport

The article presents results of research on the use of fuel mixtures, containing with biodiesel fuel. The research was carried out on specialized sea vessels with a deadweight of 10,820 tons and 29,420 tons, which made passages in the waters of South America and the Caribbean, as well as in the Sulphur Emission Control Areas of Northern Europe. The operation of the ship's power plants (main and auxiliary engines) was provided both on marine fuel of petroleum origin and on a fuel mixture that consisted of petroleum fuel and biodiesel fuel. The experiments were carried out on marine medium-speed diesel engines 5DC-17A Daihatsu Diesel, 16V32 Wärtsilä-Sulzer and Volvo Penta TWD1645GE. Various kinds of marine fuels and as biodiesel fuel Fatty Acid Methyl Esters B99.9 FAME and B99.6 FAME were used. The concentration of biodiesel fuel in the fuel mixture was 5–20 %. The environmental friendliness of the work of sea vessels was assessed by the value of the emission of nitrogen oxides, sulphur, and carbon. As a result of the research, a variant of adjusting the sea crossing was proposed to ensure the continuous operation of the ship's power plant using biodiesel fuel.

Research on the removal of charged colloidal filter materials from coolants in nuclear power plants

In this paper, nano-grade aluminum powder was used as raw material to prepare glass fiber filter material with positive Zeta potential by hydrothermal method. Since colloids of activated corrosion products in primary circuit of nuclear power plant are mainly electronegative, the filter material prepared in this experiment was used to remove colloidal particles in coolant by combining traditional mechanical filtration with electrostatic adsorption, thus improving the removal efficiency of colloidal particles. The leaching problem of soluble silicon in glass fiber filter material was studied. The leaching amount of soluble silicon in glass fiber filter material was decreased while the filtration efficiency was improved. Consequently reduce the risk of safe operation of nuclear power plants.

Optimal dispatch of battery energy storage for multi-service provision in a collocated PV power plant considering battery ageing

This study explores how a battery energy storage system (BESS) can support photovoltaic (PV) power plant operation by simultaneously minimising the PV power plant (PVPP) clipping losses and providing grid ancillary services. For this purpose, a deterministic day-ahead control strategy is developed while considering both calendar and cycling battery ageing on the BESS multi-value stream. The battery degradation cost model has also been reformulated to integrate the proposed formulation and assess the impact of different ageing aspects. The dispatching problem is reformulated to implement a linear programming approach. It allows to optimally dispatch the power flow between the different system components and schedules the ancillary service provision. The Irish regulation for participating in the Renewable Electricity Support Scheme (RESS) and DS3 grid services markets is considered. A market participation framework has also been identified, and numerical analysis is performed while emulating a real case. The obtained outputs emphasise the value of PV-BESS in providing DS3 grid services and the potential of the multi-service provision to create an additional value from energy storage deployment. The proposed approach and control strategy can also be implemented for other BESS applications in different energy markets.

A standardized protocol for assessing the performance of automatic detection systems used in onshore wind power plants to reduce avian mortality

While wind power plants are an important contribution to the production of renewable energy to limit climate change, collision mortality from turbines is a danger for birds, including many protected species. To try to mitigate collision risks, automatic detection systems (ADSs) can be deployed on wind power plants; these work by detecting incoming birds using a detection/classification process and triggering a specific reaction (scaring off the bird or shutting down the turbine). Nonetheless, bird fatalities still occur at ADS-equipped wind power plants, which raises the question of the performance of these tools. To date, the lack of a transparent, peer-reviewed experimental process to compare the performance of types of ADS has meant there is no robust protocol to assess these systems.With the aim of filling this gap, we developed two standardized protocols that provide objective and unbiased assessments of the performance of different types of ADS, based on their probability of detecting/classifying birds at risk of collision. Both protocols rely on precise 3D tracking of wild birds by human observers using a laser rangefinder, and the comparison of these tracks with those detected and recorded by an ADS. The first protocol evaluates a system's general performance, generating comparable data for all types of ADS. In this protocol, detection/classification probability is estimated conditional on several abiotic and biotic environmental factors such as bird size, distance from the target, the flight angle and azimuth of the bird, as well as weather conditions. The second protocol aims to verify that the performance of an ADS installed on a given wind power plant complies with its regulatory requirements. In this protocol, detection/classification probability is specifically estimated for a given target species at a given regulatory detection distance. This protocol also estimates the proportion of time an ADS is functional on site over a year, and the proportion of reaction orders successfully operated by wind turbines. These protocols have been field-tested and made publicly available for use by government agencies and wind power plant operators.

Application of artificial intelligence technologies and big data computing for nuclear power plants control: a review

Nuclear power plants produce a massive amount of clean energy and necessitate safe operation through intelligence technologies. Recently, the rapid advancements in communication infrastructures including artificial intelligence, big data computing, and Internet of Things devices moving the nuclear industries towards digitalization and intelligence to improve safety. The integration of these technologies into the nuclear sector offers effective tactics in addressing several challenges in the control and safe operation of nuclear power plants. This can be achieved through the insights generated from massive amounts of data. This paper comprehensively reviews the literature on artificial intelligence technologies and big data, seeking to provide a holistic perspective on their relations and how they can be integrated with nuclear power plants. The utilization of computing platforms boosts the deployment of artificial intelligence and big data analytics effectively in nuclear power plants. Further, this review also points out the future opportunities as well as challenges for applying artificial intelligence and big data computing in the nuclear industry.

Enhancement of thermal power plant performance through solar-assisted feed water heaters: An innovative repowering approach

Increasing the efficiency of a power plant by optimization or repowering has always been a crucial challenge for energy analysts. At the same time as natural energy resources are limited, the integration of renewable energy sources such as solar energy has been prioritized for the past few decades. This study presents a way of enhancing a thermal power plant with a regenerative reheating Rankine cycle, evaluating the impact of solar-powered heat exchangers on feed water heating at 200 MW single unit of the Shahid Montazeri Power Plant operating at 13 MPa and 540 °C. Increases in energy-exergy efficiencies, power output and fuel save due to this modification have been studied in 14 different cases which have been analyzed in Engineering Equation Solver (EES). The case studies cover replacing single, double, and triple Feed Water Heaters (FWH). Parametric analysis of the four most promising cases has been performed in varying operating conditions such as main steam and reheat steam temperature and pressure and condensation pressure. In the most favorable configuration (case 12), the net thermal power output increases to 225 MW while energy and exergy efficiency increases to 39.73 % and 43.99 % respectively. Exergy and energy loss across different major components have been presented as well which indicates room for further improvement and contributes valuable insights to thermal power plant optimization for sustainable and efficient energy solutions in the future.

Plant Information Management System Using OSIsoft PI System: A Case Study of Cilacap Power Plant Units 1 and 2

In the dynamic industrial sector, the ability to combine and manage data from various sources is essential. This study focuses on the utilization of the PI System to integrate operational data from multiple data sources at power plants, facilitating data driven decision making and enhancing operational efficiency. The application of the PI System in collecting data from Distributed Control Systems (DCS) and asset management systems like Maximo is examined, as well as its potential for integrating data from other power plant units. The study utilizes systems analysis methods to understand the potential of data integration in improving operational visibility and control. Findings from testing reveal that effective data integration can extend monitoring and management capabilities, indicating an overall improvement in operational performance. This study contributes to the literature on industrial data management by demonstrating the effective use of the PI System as a comprehensive tool for integrating data from various systems within power plants. It showcases the system's adaptability in enhancing operational decision-making and providing a cohesive platform for real-time performance monitoring. Additionally, the research offers insights into the practical application of systems analysis methods in the context of power plant operations, contributing to the ongoing discourse on digital transformation in the energy sector.

Design of Multi-Input Multi-Output Non-linear Model Predictive Control for Main Steam Temperature of Super Critical Boiler

Flexible operation of coal-fired power plants is becoming increasingly necessary for successful integration of large-scale renewable power generation into the power grid. The maximum ramp rate and the number of load cycles are generally limited by the thermal stress experienced by the boiler pressure parts, turbine metallurgy and creep and fatigue of critical thick-walled components Main steam temperature is a critical operating parameter that must be controlled within acceptable limits for safe operation. Main steam temperature deviation beyond acceptable limit has impact on boiler pressure parts and turbine material of construction due to creep and fatigue effect. Base load operating units do not require steep ramp rate and hence recommended ramping rates are kept low within the safe operating zone in comparison to the flexible operation of the units with wide range load change width. Thermal stresses are caused by the temperature changes inside the thick-walled components and turbine steam admission parameters. Hence, the quality of main steam temperature control plays a vital role in flexible operation of the coal fired units. Conventional cascaded PID temperature control loop architecture performs well at steady state condition within a limited variation of load change at low ramp rate but it acts slowly and performs poorly at transient operating conditions of flexible operation of the boiler turbine with wide range load variation and load cycle with high ramp rate and remains far from rated conditions. In this paper, a Multi-Input Multi-Output (MIMO) Non-linear Model Predictive Control (MPC) design for regulation of the main steam temperature of a Once-Through supercritical Boiler is proposed. The controller is based on a non-linear dynamic model which incorporates dynamics of the variables of interest. It has the capability to operate effectively across a wide load range while maintaining main steam temperature within acceptable limits. A notable advancement in this design of MPC is the incorporation of coal flow demand and feedwater flow demand as additional control inputs alongside primary and secondary spray flows. In simulation test cases, the MPC controller demonstrates satisfactory performance and computational efficiency.

Hybrid modeling of a circulating fluidized bed boiler for development of a prediction and prescription system for power plant operation

A novel hybrid modeling system of a circulating fluidized bed boiler was developed in this study. The proposed approach combines a computational fluid dynamics model, practice-oriented macro-scale models, and a data-driven reduced order model to capture both the physical and empirical behaviors of the boiler. The application of various modeling strategies allowed for capturing features of the simulated process at various scales and for developing a fast and accurate prediction system. The use of the computational fluid dynamics model improved the predictions of the practice-oriented macro-scale models providing more detailed flow information and the use of the reduced order model allowed for reducing the computation time by a factor of 1350. The hybrid model and its submodels were validated against real measurement data. The validation confirmed good accuracy in predicting key process variables. The developed model is used as a prediction and prescription system to enable operators to optimize the boiler operation and prevent abnormal events. The hybrid system is used for a supercritical, once-through boiler in Łagisza power plant in Poland, however, the presented modeling strategy can be applied in other power plants, which can improve efficiency, reduce emissions, enhance boiler availability, and ultimately contribute to more sustainable energy systems.

Research on Nuclear Power Plant Operational Task Complexity Assessment Method Based on Group Decision Making and Deep Learning of EEG Signals

This study aims to explore the correlation between the operational task complexity of nuclear power plant (NPP) operators and electroencephalogram (EEG) features. Initially, we segmented EEG signals according to operational steps and extracted a total of 120 time domain, frequency domain, and time-frequency domain features. Subsequently, we applied an adaptive principal component analysis (PCA) dimensionality reduction method to process the features. On the other hand, three experts were invited to evaluate the complexity of the operational tasks, and their evaluation data were synthesized using a group decision-making approach. A correlation analysis was performed between these data and the PCA-reduced feature data, identifying the features with the highest correlation coefficient for each participant. Then we built a long short-term memory model with the data of the first group of participants to predict the task complexity value and tested it with the data of the second group of participants. Testing the model with data from the second group yielded favorable results, with a training set mean squared error (MSE) of 0.025 and a testing set MSE of 0.078. The results of this study indicate a significant correlation between specific EEG features and task complexity in the operational tasks of NPP operators. The model established through a combination of group decision making and machine learning methods effectively predicted the task complexity levels for operators in different operational tasks. This research provides a new perspective on NPP operators’ cognitive load and operational tasks, holding practical significance for operator training and workload management.