Dynamic modeling of a direct reduced iron shaft furnace to enable pathways towards decarbonized steel production
Not Available
Immonen, Jake (ORCID:0000000341231625)↗
Engineering topics
Powell, Kody M.
Publications and source records attributed to Powell, Kody M..
Flexible operation of nuclear hybrid energy systems for load following and water desalination
Nuclear hybrid energy systems (NHES) have the potential to provide dependable and emission-free electricity to the grid while also increasing the flexibility and reliability of the electrical grid. Molten salt reactor (MSR) technology can provide consistent, carbon-free electricity while also increasing efficiency, security, and sustainability and reducing nuclear waste. This study investigates the integration of Molten Salt Reactors (MSR) and conventional Pressurized Water Reactors (PWR) with desalination technologies: Direct Contact Membrane Distillation (DCMD), Multi-Stage Flash Distillation (MSFD), and Reverse Osmosis (RO). Dynamic first-principles models were developed and tested using real grid data from the New York Independent System Operator. The results demonstrate that nuclear power is capable of flexibly responding to changing grid demand while simultaneously producing clean water, particularly during periods of low electricity demand. The MSR-RO system was found to be the most efficient in electricity generation and water production, and all hybrid systems reduced CO2 emissions by 356,000 to 682,000 tons annually. Economic analysis reveals that nuclear desalination technologies are cost-competitive with conventional systems, especially when paired with RO. Finally, these findings confirm the technical feasibility and environmental benefits of nuclear hybrid systems for sustainable electricity and water production.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Transformer Neural Networks with Spatiotemporal Attention for Predictive Control and Optimization of Industrial Processes
In the context of real-time optimization and model predictive control of industrial systems, machine learning, and neural networks represent cutting-edge tools that hold promise for enhancing dynamic modeling. This work presents a novel transformer neural network architecture for real-time optimization and model predictive control. This network design includes a modified attention mechanism inspired by positional embedding attention from vision transformers and task-specific modifications to the input-output structure of the transformer’s decoder stack. Experiments were conducted using data from a 450 MW coal-fired power plant to evaluate this approach's effectiveness. The transformer neural network was compared with conventional recurrent models, including GRU and LSTM. The transformer exhibited a 6% increase in the R-squared (R2) value of predictions and an 83% reduction in mean squared error (MSE). Computation time was also reduced by 84% compared to conventional recurrent models.
Gallup, Ethan R.↗
Effects of input gradient regularization on neural networks time-series forecasting of thermal power systems
This study proposes using neural networks, specifically gated recurrent unit (GRU), long-short-term memory (LSTM), and transformer networks, to improve control strategies in a 450 MW coal-fired power plant. However, neural networks face issues of becoming overly dependent on just a few variables to make predictions, which negatively impacts control decisions that rely on the model to determine the value of all manipulated variables. The paper introduces regularization techniques, including noise injection and input gradient regularization, during the training phase. Here, the work presents novel contributions in adapting neural networks to control industrial systems and applying regularization techniques from computer vision to industrial process control. Results demonstrate the effectiveness of input gradient regularization in reducing model dependence on subsets of variables, emphasizing the balance between fidelity and controllability. Further exploration is recommended, including the development of recurrent transformers, closed-loop control testing, and a sensitivity analysis on computer models to provide further insight.
20 FOSSIL-FUELED POWER PLANTS↗
Electrification of residential and commercial buildings integrated with hybrid renewable energy systems: A techno-economic analysis
Here, this study assesses the techno-economic viability of city-scale building electrification, comparing a base case scenario with conventional HVAC systems to an electrified scenario with heat pump systems. EnergyPlus simulations reveal a 28% increase in annual electricity consumption in the electrified scenario, shifting the peak consumption from summer to winter with a 49% surge. Hybrid renewable energy systems (HRESs) are then evaluated using HOMER Pro, considering renewable penetration from 0% to 100%. The net present cost (NPC) of the base case and electrified scenarios range from $\$10.34$ to $\$24.38$ billion and $\$8.79$ to $\$50.31$ billion, respectively. Sensitivity analysis explores the impact of carbon tax, fuel price, and renewable subsidies on HRESs in the electrified scenario. Results indicate that applying a carbon tax up to $\$120$/tonne of CO 2 increases the optimum renewable fraction from 5% to 26%, while higher fuel prices and renewable subsidies further enhance the feasibility by shifting the optimum renewable fraction to 60% and 54%, respectively, demonstrating the interplay between economic factors and renewable integration in electrified urban contexts.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Sustainability and affordability of building electrification: A state-by-state holistic approach for multifamily buildings
This paper explores the evolving narrative of building electrification, considering its potential to solve climate change. Previous research has predominantly focused on hybrid renewable energy portfolios and increasing home pricing premiums, overlooking the perspectives of renters burdened by housing costs. Moreover, existing studies have primarily examined the effects of electrification on single-family homes and case studies, neglecting multifamily buildings and their relationship with the grid. This comprehensive study leverages calibrated building energy models to address these gaps and evaluates sustainability through carbon dioxide emissions and affordability through economic performance. The findings demonstrate significant progress in electrification since 2017, with nearly all states showing decreased energy usage in the electrified models. Environmentally, twenty-two states perform better or comparably with the electrified models in the most recent study. Economically, challenges persist, but a sensitivity analysis demonstrates how results could improve soon. Furthermore, the study discusses the outlook of electrified multifamily buildings, considering ongoing decarbonization plans for the electric grid. In three case studies, electrification demonstrates improvements in over half of the states by 2026. Overall, the research highlights the importance of expanding analysis to include multifamily buildings and emphasizes the positive impact of electrification as a viable energy, environmental, and economic solution.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Challenges of reaching high renewable fractions in hybrid renewable energy systems
Not Available
Energy & Fuels↗
Intelligent Control of Thermal Energy Storage in the Manufacturing Sector for Plant-Level Grid Response
Industrial facilities are seeking new strategies that help in providing savings mechanisms for demand charges. Demand charges are the charges incurred by industrial facilities as a result of power usage. Thermal energy storage has advanced significantly with lots of new applications, garnering the interest of many industrial facilities. These applications could be used to shave the industrial facilities’ peak electric demand and reduce their demand charges. This paper aims to demonstrate the efficacy of thermal energy storage in reducing demand charges and highlight new developments in the integration of smart control systems with thermal energy storage. The study compares energy consumption and peak demand for a facility equipped with and without thermal energy storage tanks using a fixed schedule for charging and discharging. Additionally, the paper examines the impact of incorporating a smart controller to determine when to charge and discharge the tank based on the facility’s real-time power usage and a given setpoint. The results indicate cost savings from the use of thermal energy storage tanks under two proposed scenarios, reflected in the reduced cost of power consumption for the studied facility. The incorporation of a smart controller with the thermal energy storage tank in the facility studied could provide estimated savings of 3.3% per year of power consumption charges, without considering the contribution of any incentives. The estimated savings provided by the fixed schedule scenario are 2.7% per year.
25 ENERGY STORAGE↗
Conversion of food waste to renewable energy: A techno-economic and environmental assessment
Increasing quantities of food waste have become a concern due to high disposal costs in landfills and high greenhouse gas emissions. With this increase in food waste generation, there is also an increasing demand for renewable natural gas to supplement traditional fossil fuel combustion and offset the impacts of climate change. Collecting food waste from landfills and turning it into renewable natural gas using anaerobic digestion could be a win-win option for both food waste disposal and renewable energy production. While some literature exists on the energy potential, economic feasibility, and environmental benefit of food waste disposal via anaerobic digestion, no existing study simultaneously evaluates the energy, economic and environmental effect of food waste to renewable energy via anaerobic digestion, especially on a plant and city scale. Further, this study is focused on the techno-economic and environmental assessment of food waste to energy via anaerobic digestion in order to fill this gap. Four anaerobic digestion pathways are considered in this study: flare, pipeline natural gas, combined heat and power, and combined cycle for efficient power generation. Using a city of 1M people the results show that renewable natural gas from food waste could supply the natural gas usage for 1.9% of residential use, 2.7% of commercial use, 1.1% of industrial use, 167.5% of the compressed natural gas vehicle fleet, 0.7% of electric power generation, or 2.5% of industrial high-temperature heating processes. All pathways except pipeline natural gas will have a positive net present value in the baseline scenario, and the pipeline natural gas pathway will become economically viable with a net present value of 31 USD/t of food waste with renewable energy credits. Lastly, all of the pathways achieve negative greenhouse gas emissions, which indicates that anaerobic digestion is a more environmentally friendly method for the handling of food waste than landfills.
03 NATURAL GAS↗
Dynamic machine learning-based optimization algorithm to improve boiler efficiency
With decreasing computational costs, improvement in algorithms, and the aggregation of large industrial and commercial datasets, machine learning is becoming a ubiquitous tool for process and business innovations. Machine learning is still lacking applications in the field of dynamic optimization for real-time control. This work presents a novel framework for performing constrained dynamic optimization using a recurrent neural network model combined with a metaheuristic optimizer. The framework is designed to augment an existing control system and is purely data-driven, like most industrial Model Predictive Control applications. Several recurrent neural network models are compared as well as several metaheuristic optimizers. Hyperparameters and optimizer parameters are tuned with parameter sweeps, and the resulting values are reported. Further, the best parameters for each optimizer and model combination are demonstrated in closed-loop control of a dynamic simulation, and several recommendations are made for generalizing this framework to other systems. Up to 0.953% improvement is realized over the non-optimized case for a simulated coal-fired boiler. While this is not a large improvement in percentage, the total economic impact is $991,000 per year, and this study builds a foundation for future machine learning with dynamic optimization.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
System benefits of industrial battery storage: A comparison of grid and facility control and dispatch
Customer-owned, distributed battery installations are being incentivized by utilities to increase installed battery capacity. In many of these incentive agreements, the battery owner relinquishes battery control to the utility in exchange for incentive money. The industrial sector has lagged in storage installation when compared to the residential and commercial sectors. This study compares the economic advantages to utilities and industrial facilities in different dispatch control situations. The study presents a novel framework for the optimization of multiple systems using load profiles from the industrial, residential, and commercial sectors. Case studies are presented to illustrate different dispatch scenarios. Further, the simulations showed more fiscal benefit for the industrial facilities to dispatch the battery for electrical demand reduction than utility dispatch. In the case studies, facility dispatch control resulted in an increase of facility savings by a factor of about 8.7 when compared to utility dispatch. Battery size plays a significant factor on the impact of the grid’s generating costs, showing that larger batteries can provide significant benefit even if dispatched by the facility. Future policies concerning industrial battery installations should consider overall economic benefits to utility and facility in the form of rate structures and incentive participation based on battery size.
25 ENERGY STORAGE↗
Grid‐responsive smart manufacturing: A perspective for an interconnected energy future in the industrial sector
Abstract With the growing amount of renewable energy sources, the grid has become responsible for accounting for intermittency and the flexibility needed to utilize dynamic sources. Expensive peaking plants and energy storage systems have been proposed as ways to mitigate those problems. There is a large group of energy consumers that can respond to grid conditions. Historically, these consumers have been residential and commercial users, but with modern innovations and practices, industrial consumers have the potential to become a major player in this space. Grid‐responsive smart manufacturing can be used to utilize modern tools in manufacturing innovation as enablers for grid response. These modern tools already exist but are not widely used for industrial grid‐side energy management. This article defines grid‐responsive smart manufacturing, identifies five major barriers to its widespread implementation, and portrays the path to getting industrial users to be key players in grid stability and flexibility.
Billings, Blake W.↗
A review on the application of machine learning for combustion in power generation applications
Abstract Although the world is shifting toward using more renewable energy resources, combustion systems will still play an important role in the immediate future of global energy. To follow a sustainable path to the future and reduce global warming impacts, it is important to improve the efficiency and performance of combustion processes and minimize their emissions. Machine learning techniques are a cost-effective solution for improving the sustainability of combustion systems through modeling, prediction, forecasting, optimization, fault detection, and control of processes. The objective of this study is to provide a review and discussion regarding the current state of research on the applications of machine learning techniques in different combustion processes related to power generation. Depending on the type of combustion process, the applications of machine learning techniques are categorized into three main groups: (1) coal and natural gas power plants, (2) biomass combustion, and (3) carbon capture systems. This study discusses the potential benefits and challenges of machine learning in the combustion area and provides some research directions for future studies. Overall, the conducted review demonstrates that machine learning techniques can play a substantial role to shift combustion systems towards lower emission processes with improved operational flexibility and reduced operating cost.
Engineering↗
Dynamic optimization with flexible heat integration of a solar parabolic trough collector plant with thermal energy storage used for industrial process heat
There is an increasing need to reduce fossil fuel consumption used for industrial process heat to slow the effects of climate change. Using solar thermal heat is a viable way to replace fossil fuel use, but solar industrial process heat plants have limited implementation due to large upfront costs and inefficiencies from the inherent variability from solar energy. Having more flexible and optimized control over these plants can enable them to be more efficient. Here in this work, a solar industrial process heat plant with thermal energy storage that can flexibly collect and deliver heat to two industrial processes (flexible heat integration) is dynamically modeled with control setpoints found by a dynamic optimization method. The optimized case can increase the solar efficiency of the plant by 7.5% on average relative to a base case. The results show that it is best to collect heat at lower temperatures for all but ideal solar conditions, only medium to high quality heat should be stored in large quantities, and exchanging heat with a lower temperature heat sink is generally more efficient. The optimized case is able to reduce the levelized cost of heat of the plant to $31.83/MWh th compared to a base case value of $34.10/MWh th . The optimized case can reduce emissions by 22.2% compared to a plant which uses only natural gas. This work shows that dynamic optimization with flexible heat integration can be a cost-effective way to increase efficiency so that more of these types of plants can be implemented.
14 SOLAR ENERGY↗
Industrial battery operation and utilization in the presence of electrical load uncertainty using Bayesian decision theory
Behind the meter battery storage is becoming increasing popular in all sectors, though enthusiasm has recently lagged in the industrial sector. Even though there may be many factors contributing to this including lack of innovation, prohibitive costs, and undesirable rate structures, a difficulty arises in accounting for uncertainty of electrical load in industrial facilities while still attempting to utilize battery storage as much as possible all while trying to achieve fiscal profitability. Here this study utilizes Gaussian process regression and Bayesian decision theory to organize load data and quantify electrical load uncertainty to properly and effectively discharge industrial battery storage. The study employs a simulation model to set battery load setpoints for the span of the utility billing period according to the degree of risk aversion. This combination of economic analysis according to utility billing period and utilization of degree of risk aversion to make decisions on the uncertainty of the data has not before been applied to battery storage. The method resulted in an annual average reduction of peak demand by 3.8 % at the lowest amount of savings and lowest risk aversion. The highest risk aversion resulted in an annual average reduction of peak demand of 7.5 %. The maximum reduction of peak load in any month was 13.8 % in the month of December with a relatively high risk aversion. With a the highest amount risk aversion tested, the model reduced demand ten of the twelve months of the year.
25 ENERGY STORAGE↗
Dynamic energy system modeling using hybrid physics-based and machine learning encoder–decoder models
Three model configurations are presented for multi-step time series predictions of the heat absorbed by the water and steam in a thermal power plant. The models predict over horizons of 2, 4, and 6 steps into the future, where each step is a 5-minute increment. The evaluated models are a pure machine learning model, a novel hybrid machine learning and physics-based model, and the hybrid model with an incomplete dataset. The hybrid model deconstructs the machine learning into individual boiler heat absorption units: economizer, water wall, superheater, and reheater. Each configuration uses a gated recurrent unit (GRU) or a GRU-based encoder–decoder as the deep learning architecture. Mean squared error is used to evaluate the models compared to target values. The encoder–decoder architecture is over 11% more accurate than the GRU only models. The hybrid model with the incomplete dataset highlights the importance of the manipulated variables to the system. The hybrid model, compared to the pure machine learning model, is over 10% more accurate on average over 20 iterations of each model. Automatic differentiation is applied to the hybrid model to perform a local sensitivity analysis to identify the most impactful of the 72 manipulated variables on the heat absorbed in the boiler. The models and sensitivity analyses are used in a discussion about optimizing the thermal power plant.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Development of novel dynamic machine learning-based optimization of a coal-fired power plant
The increasing fraction of intermittent renewable energy in the electrical grid is resulting in coal-fired boilers now routinely ramp up and down. The current state-of-the-art operation for such boilers is to apply steady-state, neural network-based optimization to make control decisions in real-time, and this report demonstrates the feasibility of extending this to dynamic, neural network-based optimization using a long short-term memory neural network. A simplified numerical simulation of a t-fired coal boiler and supporting equipment is used to represent a real plant subjected to both steady-state, neural network-based optimization and dynamic, neural network-based optimization. Using the same intervals and a particle swarm optimization algorithm, the dynamic optimization outperforms the steady-state optimization and realizes up to 4.58% improvement in thermal efficiency. Dynamic optimization with a long short-term memory neural network is shown to both be feasible and beneficial for operation of a coal-fired boiler under changing load.
42 ENGINEERING↗