Real Power Plant Research Articles

Comparison and validation of three process simulation programs during warm start-up procedure of a combined cycle power plant

The energy market is one of the largest and most important markets around the world. Where flexibility and efficient operation in the start-up of combined cycle power plants (CCPPs) play an essential role in the competitiveness of CCPPs in the energy market. Moreover, to study the CCPPs, the Dynamic simulation is suitable tool for predicting the behavior of power plant with high accuracy. In this paper, we study the details of warm start-up procedure of CCPP, with subcritical, natural circulation and three-pressure stages heat recovery steam generator. Examining the capability of Power Plant Simulator & Designer (PPSD) by validating the numerical results of PPSD with operational data of real power plant. In addition to compare numerical data of PPSD by another numerical data acquired using another two dynamic simulation tools APROS and ASPEN Plus Dynamics to ensure the capability of PPSD to perform such simulations. The model built based on a real power plant design data with all geometry including heat exchangers, turbines, pipes valves, pumps, and control circuits as well. PPSD shows accurate results with high quality in both qualitatively and quantitatively, compared to the measured data and other simulations well. Moreover, it shows that it is capable to perform a dynamic simulation of CCPP with relatively shorter computation time and user-friendly tools.

Influence of the active leakage current pathway on the potential induced degradation of CIGS thin film solar modules

Copper Indium Gallium Diselenide (CIGS) shows a degradation due to a voltage between the solar cell and the ground potential, known from other solar cell technologies as well. This degradation is named potential induced degradation (PID). At the cell level, PID has been studied in laboratory experiments and can be traced back to sodium diffusion, which is accelerated by the mentioned voltage against ground potential. The applied potential will lead to a leakage current, which can serve as an indicator for PID.The question whether PID of CIGS thin film solar modules also occurs in real PV power plants inevitably arises. To solve this question the degradation behaviour will be analysed by accelerated aging tests. Furthermore, the influence of different leakage current pathways will be shown. The results of this study serve as a basis for modelling the lifetime of CIGS thin-film solar modules taking degradation due to PID into account.

Linearity Decomposition-Based Cointegration Analysis for Nonlinear and Nonstationary Process Performance Assessment

The assessment of operating performance plays an important role in guaranteeing high efficiency, low cost, large economic benefit, etc., for modern industry under routine operating conditions. However, the nonlinear and nonstationary properties that widely exist in real industrial processes, make it a great challenge to obtain accurate assessment of operating performance. To solve the problem of performance assessment for nonstationary processes with nonlinearly cointegrated relationships, the linearity decomposition based cointegration analysis (CA) is proposed through multi-objective optimization in the present work. First, the nonstationary critical-to-performance variables are selected to remove the redundant information and improve the accuracy for assessment. Second, the selected nonstationary critical-to-performance variables are then decomposed into a set of local sub-blocks. Each sub-block contains a group of linearly cointegrated variables, based on which a two-layer multiblock assessing model can be established for nonlinear and nonstationary process using CA. Bayesian inference based criterion is adopted to indicate the operating performance for online assessment. Finally, the feasibility and efficacy are illustrated via a numerical example and a pulverizing process of a real thermal power plant.

An approach to discovering event correlations among edge sensor services

In an IoT environment, a surge in sensor data volume has exposed the shortcomings of cloud computing, particularly the limitation of network transmission capability and centralised computing resources. To handle these issues, this paper proposes a service-oriented framework, called as INFOG, to support the dynamic cooperation among sensors with the fog computing paradigm. Proactive data services and service hyperlinks, which are our previous work, are two key abstractions for the INFOG framework. The services are software-defined abstraction of physical sensors. They are deployed in edge nodes in INFOG. And service hyperlinks, encapsulation of service correlations, enable the cooperation of sensors at the software layer. We also propose a frequent sequential pattern-based approach to effectively discover service hyperlinks. Based on the dataset from a real power plant as well as several synthetic datasets, we do lots of experiments to verify the effectiveness and efficiency of our algorithm.

An approach to discovering event correlations among edge sensor services

In an IoT environment, a surge in sensor data volume has exposed the shortcomings of cloud computing, particularly the limitation of network transmission capability and centralised computing resources. To handle these issues, this paper proposes a service-oriented framework, called as INFOG, to support the dynamic cooperation among sensors with the fog computing paradigm. Proactive data services and service hyperlinks, which are our previous work, are two key abstractions for the INFOG framework. The services are software-defined abstraction of physical sensors. They are deployed in edge nodes in INFOG. And service hyperlinks, encapsulation of service correlations, enable the cooperation of sensors at the software layer. We also propose a frequent sequential pattern-based approach to effectively discover service hyperlinks. Based on the dataset from a real power plant as well as several synthetic datasets, we do lots of experiments to verify the effectiveness and efficiency of our algorithm.

A numerical study for the assessment of air pollutant dispersion with chemical reactions from a thermal power plant

In this work, a computational simulation of the pollutants spread generated during fuel combustion at Ekibastuz SDPP-1 and their chemical reaction in the atmosphere have been presented. Using the example of a real thermal power plant (Ekibastuz SDPP -1), the dispersion NO, NO2, CO and products NO2, HNO3, CO2 during a chemical reaction with oxygen was modeled. The numerical method was verified by solving three test problems and the received computational solutions were compared with data from the measurement and the computational data of other authors. The purpose of the work was to investigate the pollution level at various ranges from the source. As a result, the mass fraction of the concentration and product during a chemical reaction were determined. According to the obtained data, with increasing distance from the source, the concentration of pollution spreads more widely under the influence of diffusion. The farther the distance from the chimney, the lower the concentration of the substance and products. Thus, the obtained computational values may allow in the future predicting the optimal distance from residential areas for the construction of thermal power plants, at which the concentration of emissions will remain at a safe level.

Enhancing peak shaving capability by optimizing reheat-steam temperature control of a double-reheat boiler

Double-reheat coal-fired power plants elicit much research attention due to their high efficiency and low emissions. However, steam temperatures fluctuate heavily, especially in transient processes, thereby compromising the load cycling capacity of these power plants and even threatening their stable operation. To overcome these challenges and optimize the operations of double-reheat boilers, a dynamic boiler model was established and validated with values measured from a real power plant. A revised reheat steam control logic was proposed that considers the heat storage variation in boiler metals and the deviation of steam temperatures during peak shaving transient processes. The performance of the original and revised reheat steam control logics in terms of temperature control and energy delivery characteristics was compared and discussed. Calculation results show that the revised control logic performs better than the original logic in large-range load cycling processes. The steam temperature fluctuations are alleviated, and the overshoots are reduced substantially. The revised reheat steam control logic ensures that the accessible maximum load variation rate is 4.0% BMCR min−1, which cannot be safely guaranteed by the original logic. These improvements are beneficial to increasing the threshold of the load variation rate and improve the flexibility of the power plant. The average thermal (ηI) and exergetic (ηII) efficiencies of the original and revised control logics in load cycling are compared. For loading up processes, ηI and ηII decrease by 1.04% and 0.28% at the most, respectively. For the loading down processes, the maximum increments in ηI and ηII are 1.44% and 0.77%, respectively.

Microstructure evolution and oxidation behaviour of shot-peened S30432 stainless steel during exposure in double reheat USC power plant steam environment

ABSTRACTThe microstructural evolution and oxidation behaviour of shot-peened S30432 stainless steel during exposure in steam environment in a real double reheat USC power plant has been investigated through microstructural characterisation using electron microscopes. The results indicated that shot-peening treatment introduced a thick deformation layer to the material, and this deformation layer consisted of an outer nanostructured layer and an inner layer with deformation bands. During thermal exposure, the nanostructured layer experienced extensive recrystallisation and coarsening, and numerous σ particles were found to precipitate in the deformation layer as well. Furthermore, a double layer structured oxide formed on the material, and the oxide layer grew slowly after exposure for 6219 h, which was related to the formation of a thin and dense SiO2 layer at the metal/oxide interface. The mechanism of the microstructural evolution and oxidation has been discussed.

Thermo‐Mechanical Analysis of a Steam Turbine rotor

AbstractIn power plants, high temperatures prevail during long holding times. Furthermore, power plants are often started and shut‐down in order to account for gaps or oversupplies in energy production. These loading conditions induce both creep and fatigue loads. Due to their excellent thermo‐mechanical properties, such as high tensile strength and elevated corrosion resistance, heat‐resistant steels are established materials for power plant components. Nevertheless, these steels tend to soften under deformation, which should be accounted for by a constitutive model.The contribution at hand analyses the thermo‐mechanical behavior of a steam turbine rotor. For this purpose, a unified phase mixture model is introduced. This constitutive model accounts for rate‐dependent inelasticity, hardening, as well as softening by employing an iso‐strain approach with a soft and a hard constituent. While the soft constituent represents areas with a low dislocation density, such as the interior of subgrains, the hard constituent refers to regions with a high dislocation density, i.e. the subgrain boundaries. Furthermore, two internal variables are introduced: a backstress tensor of Armstrong‐Frederick type and a scalar softening variable. The model results in a coupled system of three evolution equations with respect to the inelastic strain, the backstress, and the softening variable.To allow for the analysis of real power plant components, the model is implemented into the finite element method such that the evolution equations are integrated based on the backward EULER method. The applicability of the model is demonstrated by conducting a thermo‐mechanical analysis of a steam turbine rotor with complex geometry under realistic boundary conditions. In a first step, the instationary temperature field in the rotor is computed in a heat transfer analysis. Thereby, typical steam temperatures in power plants and the corresponding heat transfer coefficients are prescribed. As a next step, the structural analysis is conducted based on the phase mixture model and the obtained temperature field as input. In addition, the time‐dependent rotational frequency and steam pressure are taken into account. Note that the influence of different start‐up procedures such as a cold or a hot start is examined in detail. As a result, the structural analysis provides the stress‐strain hystereses, which constitute the basis for further fatigue and damage assessment.

Heat losses and thermal stresses of an external cylindrical water/steam solar tower receiver

A receiver serves as a pivotal component in the solar power system as it is responsible for the light-heat conversion. Extensive research has been carried out on cavity receivers while external receivers have been neglected hitherto. Considering this imperative research gap, this work endeavours to narrow the gap by numerically analyzing the thermal performance of an external cylindrical receiver. A methodology is proposed to determine the efficiency of a cylindrical shaped receiver. A heliostat field is simulated using Monte-Carlo Ray Tracing tool to obtain heat flux distribution on the receiver. The peak heat flux obtained, i.e., 425 kW/m2 lies at the centre of the receiver’s front. By designing a tube layout and using boiling heat transfer correlations, temperature at the receiver’s surface and water are obtained. Numerical analysis and simulations are then carried out to evaluate receiver’s thermal efficiency in six different wind directions and four different wind velocities between 3 m/s and 12 m/s. Natural convection and radiation losses were also considered. Combined heat transfer coefficients obtained through numerical simulations are compared with the previous experimental data. The effect of wind in a single direction is precisely evaluated by dividing the cylinder into panels and evaluating heat losses on each panel individually. The thermal efficiency evaluated oscillates between 71% and 77% based on wind velocity, and the results are validated against the real power plants and experimental data for cylindrical solar receivers. A tube at receiver’s centre having the highest temperature gradient is then selected to evaluate thermal stresses. The equivalent stress obtained is less than the yield strength with safety factor > 2.5.

Impact of Inverter-Interfaced Renewable Energy Generators on Distance Protection and an Improved Scheme

With different structures and control strategies, inverter-interfaced renewable energy generators (IIREGs) have different fault characteristics from synchronous generators. This makes the conventional distance protection used in networks with synchronous generators not applicable for transmission lines emanating from IIREGs. Therefore, supported by the fault current analysis, operating performances of distance relays on both sides of the transmission line are unveiled. It reveals the problem that the conventional distance relay on the IIREG side has a high risk of malfunction or refusing to operate. To cope with this adaptability problem, an improved scheme based on time delay and zero-sequence impedance is proposed in this paper. To validate the operating performances of the scheme, a detailed IIREG model is built in a real-time digital simulator, and simulation tests are carried out. Apart from these, a field short-circuit test is performed in a real wind power plant to examine the practical feasibility of the proposed scheme. Both the simulation results and the field test confirm the problem of the conventional distance protection and verify the reliability of the improved scheme.

Sea trial results of a predictive algorithm at the Mutriku Wave power plant and controllers assessment based on a detailed plant model

Improving the power production in wave energy plants is essential to lower the cost of energy production from this type of installations. Oscillating Water Column is among the most studied technologies to convert the wave energy into a useful electrical one. In this paper, three control algorithms are developed to control the biradial turbine installed in the Mutriku Wave Power Plant. The work presents a comparison of their main advantages and drawbacks first from numerical simulation results and then with practical implementation in the real plant, analysing both performance and power integration into the grid. The wave-to-wire model used to develop and assess the controllers is based on linear wave theory and adjusted with operational data measured at the plant. Three different controllers which use the generator torque as manipulated variable are considered. Two of them are adaptive controllers and the other one is a nonlinear Model Predictive Control (MPC) algorithm which uses information about the future waves to compute the control actions. The best adaptive controller and the predictive one are then tested experimentally in the real power plant of Mutriku, and the performance analysis is completed with operational results. A real time sensor installed in front of the plant gives information on the incoming waves used by the predictive algorithm. Operational data are collected during a two-week testing period, enabling a thorough comparison. An overall increase over 30% in the electrical power production is obtained with the predictive control law in comparison with the reference adaptive controller.

Application of the response surface methodology (RSM) in heavy metal removal from real power plant wastewater using electrocoagulation

In this study, application and efficiency of electrocoagulation (EC) in the removal of nickel and iron from real powerplant wastewater was evaluated. Fe electrode (St 12) was used as anode and cathode, connected with parallel monopolar configuration. Tests were conducted in two phases. Phase I mainly focussed on changing the range of the parameters to attain the possible range in which the tests can be conducted and also reached the optimum efficiency. A stirred batch reactor was used to define the test parameters range, including initial pH, electrical current (A), electrode distance (cm) and electrolysis time (min). Then, phase II of this study was conducted using the response surface methodology to design and optimise operational parameters. Outcomes indicated that considering pH = 8.1, d = 1 cm, I = 1.5 A and t = 18 min as operational conditions can result in 99.9% removal efficiency. Remarkably, the high correlation between experimental and predicted values (R 2=99.5% and 99.6% for iron and nickel, respectively) demonstrated that the EC process is a promising method to remove heavy metals from power plant wastewater.

A Deep Neural Network for Unsupervised Anomaly Detection and Diagnosis in Multivariate Time Series Data

Nowadays, multivariate time series data are increasingly collected in various real world systems, e.g., power plants, wearable devices, etc. Anomaly detection and diagnosis in multivariate time series refer to identifying abnormal status in certain time steps and pinpointing the root causes. Building such a system, however, is challenging since it not only requires to capture the temporal dependency in each time series, but also need encode the inter-correlations between different pairs of time series. In addition, the system should be robust to noise and provide operators with different levels of anomaly scores based upon the severity of different incidents. Despite the fact that a number of unsupervised anomaly detection algorithms have been developed, few of them can jointly address these challenges. In this paper, we propose a Multi-Scale Convolutional Recurrent Encoder-Decoder (MSCRED), to perform anomaly detection and diagnosis in multivariate time series data. Specifically, MSCRED first constructs multi-scale (resolution) signature matrices to characterize multiple levels of the system statuses in different time steps. Subsequently, given the signature matrices, a convolutional encoder is employed to encode the inter-sensor (time series) correlations and an attention based Convolutional Long-Short Term Memory (ConvLSTM) network is developed to capture the temporal patterns. Finally, based upon the feature maps which encode the inter-sensor correlations and temporal information, a convolutional decoder is used to reconstruct the input signature matrices and the residual signature matrices are further utilized to detect and diagnose anomalies. Extensive empirical studies based on a synthetic dataset and a real power plant dataset demonstrate that MSCRED can outperform state-ofthe-art baseline methods.

Comparative numerical analysis between two designs of Francis runner blades in sediment affected conditions

Sediment erosion of hydraulic turbine is the major problem from the perspective of operation and maintenance in power plants of Himalayan and Andes region. The effects of sediment erosion on turbine components need to be predicted in advance during the design phase so that the best design with better sediment handling can be installed in real power plants. In this paper, comparison of performance and erosion of two different designs of Francis turbine i.e. design I and design II are carried out. Full turbine steady state numerical simulations are carried out for 8 different guide vane openings using shear stress transport (SST) k-ω turbulence model. Sediment erosion analysis is carried out using Tabakoff erosion model for both the designs, numerical hill charts are constructed using Ned and Qed values for all the operating conditions. Comparison of pressure distribution along pressure and suction side of GV surface between design I and a similar measurement in a previous experiment is carried out, which shows a good agreement between numerical and experimental results. Hydraulic efficiency and the sediment erosion rate density on the runner blades of design I for all operating conditions is higher than that of design II. The difference in efficiency is less than 1.85%for all operating conditions while sediment erosion rate density is much less in design II, which shows that design II is a better option for this turbine.

Electrosorptive removal of salt ions from water by membrane capacitive deionization (MCDI): characterization, adsorption equilibrium, and kinetics.

Capacitive deionization (CDI) was demonstrated to be an affordable technology for reduction of salt concentrations in brackish water. In this study, a novel membrane capacitive deionization (MCDI) cell was assembled by incorporating ion exchange membranes into the CDI cell which was built with high-adsorption electrodes based on ordered mesoporous carbon. The synthesized mesoporous carbon electrode was fully characterized. The simultaneous analysis of the electrosorption capacity and adsorption/desorption kinetics was evaluated by using real power plant desulfurization wastewater. The ordered mesoporous carbon was favorable for salt ion electrosorption, and the best performance was obtained by using MCDI which improved the removal efficiency of total dissolved solids (TDSs) from 65 to 82%. The total hardness and alkalinity of the effluent after treatment could meet the requirement of water quality standard for industries. Langmuir isotherm and pseudo-first-order kinetic models were found to be in best agreement with experimental results of salt ion electrosorption. The selective transport of ions between the electrode surface and bulk solution due to the ion exchange membranes resulted in a better desalination performance of MCDI. The results presented in this paper could be used for developing new electrode materials of MCDI for desalination from water.

Dynamic modeling of NOX emission in a 660 MW coal-fired boiler with long short-term memory

With the rapid development of renewables, increasing demands for the participation of coal-fired power plants in peak load regulation is expected. Frequent transients result in continuous, wide variations in NOX emission at the furnace exit, which represents a substantial challenge to the operation of SCR systems. A precise NOX emission prediction model under both steady and transient states is critical for solving this issue. In this study, a deep learning algorithm referred to as long short-term memory (LSTM) was introduced to predict the dynamics of NOX emission in a 660 MW tangentially coal-fired boiler. A total of 10000 samples from the real power plant, covering 7 days of operation, were employed to train and test the model. The learning rate, look-back time steps, and number of hidden layer nodes were meticulously optimized. The results indicate that the LSTM model has excellent accuracy and generalizability. The root mean square errors of the training data and test data are only 7.6 mg/Nm3 and 12.2 mg/Nm3, respectively. The mean absolute percentage errors are within 3%. Additionally, a comparative study between the LSTM and the widely used support vector machine (SVM) was conducted, and the result indicates that the LSTM outperforms the SVM.

Dynamic simulation of the circulating fluidized bed loop performance under the various operating conditions

In a circulating fluidized bed boiler, the large thermal mass and flow characteristics of the solids strongly affect the transient response of the circulating fluidized bed loop temperature, which determines the heat transfer rate to steam flow. Therefore, it is essential to interpret the dynamic response of the solid behavior in the circulating fluidized bed loop for the stable and efficient operation of the circulating fluidized bed boiler. In this study, the dynamic simulation of the solid behavior along with the flue gas flow in a circulating fluidized bed loop was performed by applying the core-annulus approach for the solid-gas flow inside the furnace and selected models for other physical phenomena of the fluidized bed. The circulating fluidized bed loop of a commercial boiler was selected as the target system. Especially, the model simulates the characteristics of the solid behavior, such as the local solid mass distribution, and the solid flow inside the furnace and the circulating solid according to the various operating conditions. These aspects are difficult to measure and quantify in a real power plant. In this paper, the simulated furnace temperature behavior as the representative performance parameter of the circulating fluidized bed loop was discussed along with the qualitative operation experiences reported in the literature. The operating conditions include the feed rate of fuel and air, the particle size, the solid inventory and the solid circulation rate. Furnace temperature behavior was reproduced through the simulation for each operating case in the literature and was analyzed with the solid behavior along with the combustion rate and heat transfer rate of the circulating fluidized bed loop. The simulation enables quantitative evaluation of the effect of the solid behavior on the temperatures of the furnace and return part in the various operating conditions.

Transient Simulation of Geothermal Combined Heat and Power Generation for a Resilient Energetic and Economic Evaluation

Geothermal power plants based on the organic Rankine cycle (ORC) are used to convert the thermal power of brine into electricity. The efficiency and profitability of these power plants can be increased by an additional heat supply. The purpose of this study is to evaluate different combined heat and power (CHP) concepts for geothermal applications by thermodynamic and economic considerations. Therefore, a dynamic simulation model of a double-stage ORC is developed to perform annual return simulations. The transient ORC model is validated in a wide range by operational data of an existing power plant in the German Molasse Basin. A district heating system is considered and the corresponding heat load profiles are derived from a real geothermal driven heating network. For CHP, parallel and combined configurations are considered. The validation of the transient model is satisfying with a correlation coefficient of 0.99 between the simulation and real power plant data. The results show that additional heat extraction leads to a higher exergetic efficiency and a higher profitability. The exergetic efficiency and the profitability are increased by up to 7.9% and 16.1%, respectively. The combined concept shows a slightly better performance than the parallel configuration. The efficiency can be increased by up to 1.3%. In economic terms, for CHP the annual return can be increased by at least 2,500,000 €. In principle, the dynamic model shows reliable results for high power gradients. This enables an investigation of geothermal ORC models for the reserve market in future works.

Separation of Carbon Dioxide from Real Power Plant Flue Gases by Gas Permeation Using a Supported Ionic Liquid Membrane: An Investigation of Membrane Stability.

The separation of carbon dioxide from coal-fired power plant flue gases using a CO2/N2-selective supported ionic liquid membrane (SILM) was investigated and the performance and stability of the membrane during operation are reported. The membrane is composed of a polyacrylonitrile (PAN) ultrafiltration membrane as a support and a selective layer of an ionic liquid (IL), 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (EMIM Tf2N). The feasibility of large-scale SILM production was demonstrated by the formation of a square-meter-scale membrane and preparation of a membrane module. A flat-sheet envelope-type SILM module containing 0.67 m2 of the membrane was assembled. Prior to real flue gas operation, the separation behaviour of the membrane was investigated with single gases. The stability of the SILM during the test stand and pilot plant operation using real power plant flue gases is reported. The volume fraction of carbon dioxide in the flue gas was raised from approx. 14 vol. % (feed) to 40 vol. % (permeate). However, issues concerning the membrane stability were found when SO3 aerosols in large quantities were present in the flue gas.