Actual Energy Research Articles

Current Challenges in Operation, Performance, and Maintenance of Photovoltaic Panels

The installed solar capacity in the European Union has expanded rapidly in recent years. The production of these plants is stochastic and highly dependent on the weather. However, many factors should be considered together to estimate the expected output according to the weather forecast so that these new PV plants can operate at maximum capacity. Plants must be operated in coordination with maintenance operations and considering actual energy market prices. Various methods have recently been developed in the literature, ranging from the most impactful artificial-intelligence-based generation estimation methods to various diagnostic and maintenance methods. Moreover, the optimal operational and maintenance strategy usually depends on market regulation, and there are many concerns related to the distribution system operator. This review article aims to summarize and illustrate the challenges of operating and maintaining solar power plants and the economic and technical importance of these problems.

Closed-loop control of [formula omitted]EDM surface quality with alternate on-machine metrology and in-process roughness prediction

As an established tool-based micro-machining technique, micro electrical discharge machining (μEDM) has received widespread attention from academia and industry for its capability of producing intricate micro-scale structures and features on various difficult-to-cut materials. However, assurance of the as-built surface quality and consistency has been a challenge due to the complex interactions among machining parameters and the unpredictable variability of material properties and the discharge process. In this paper, an efficient closed-loop control methodology based on an innovative application of on-machine metrology (OMM) and in-process roughness prediction (IPRP) is proposed for automatically ensuring the μEDM surface quality. The work takes into consideration process fundamentals related to size and distribution of craters by means of single crater experiments and pulse discrimination studies in order to analyze the influence of actual energy parameter on the evolving surface roughness. The OMM enables accurate in situ surface characterization while the IPRP, which is built on an online Bayesian regression model, allows for the prediction of surface quality according to the monitored process features. This IPRP model presents an increased predictive performance and confidence when more OMM data become available. Depending on the acquired quality outputs, a two-step quality control strategy involving energy adaptation and discharge stability regulation is implemented. This novel control strategy has proven to achieve a consistent areal surface roughness with variations smaller than 0.1 μm for a specific material and dielectric combination in μEDM milling.

High-Loading Lithium-Sulfur Batteries with Solvent-Free Dry-Electrode Processing.

Lithium-sulfur (Li-S) batteries, with their high energy density, nontoxicity, and the natural abundance of sulfur, hold immense potential as the next-generation energy storage technology. To maximize the actual energy density of the Li-S batteries for practical applications, it is crucial to escalate the areal capacity of the sulfur cathode by fabricating an electrode with high sulfur loading. Herein, ultra-high sulfur loading (up to 12mgcm-2) cathodes are fabricated through an industrially viable and sustainable solvent-free dry-processing method that utilizes a polytetrafluoroethylene binder fibrillation. Due to its low porosity cathode architecture formed by the binder fibrillation process, the dry-processed electrodes exhibit a relatively lower initial capacity compared to the slurry-processed electrode. However, its mechanical stability is well maintained throughout the cycling without the formation of electrode cracking, demonstrating significantly superior cycling stability. Additionally, through the optimization of the dry-processing, a single-layer pouch cell with a loading of 9mgcm-2 and a novel multi-layer pouch cell that uses an aluminum mesh as its current collector with a total loading of 14mgcm-2 are introduced. To address the reduced initial capacity of dry-processed electrodes, strategies such as incorporating electrocatalysts or employing prelithiated active materials are suggested.

Lithium magnesium silicate nanoparticles with unique cation acceleration channels as Li-ion rectifiers for stabilizing Li metal batteries

Lithium metal anodes possessing a high theoretical specific capacity and low redox potential are considered the most promising materials for high-energy-density Li metal batteries (LMBs). However, safety concerns due to the growth of lithium dendrites and short-cycle lifetime issues due to more side reactions have hindered the practical implementation of LMBs. To alleviate the above problems and further improve the safeness and cycling stability of LMBs, a Li-Nafion@Laponite-XLG@PP (LNLX@PP) based functional separator is proposed. The as-optimized functional separator (LNLX-30@PP) not only has high ionic conductivity (8.4 × 10−4 S cm−1), Li+ transfer number (0.8), and high mechanical properties, but also induces the formation of a strong LiF-rich SEI and hinders anionic shuttling through electrostatic shielding effects. As a result, the Li|LNLX-30@PP |Li symmetric cells can stably cycle for 650 h under high current density (2 mA cm−2) and high charge-discharge capacity (4 mA h cm−2). Furthermore, this approach enables more than 1000 cycles at 3 C with high coulomb efficiency of 99.9 % in LFP||Li coin cells and achieve an actual initial energy density of 322 Wh kg−1 and more than 100 stable cycles at 0.2 C in an assembled Li-S pouch cell with S loading of ∼9.3 mg cm−2. This work will inspire the design of hybrid functional separators for advanced Li metal batteries.

Dual-stage polyporate framework with redox mediator for high loading lithium sulfur batteries

Realizing a high sulfur loading cathode is a necessary measure for assembling a high actual energy density lithium sulfur battery, which requires consideration of conductivity, reaction interface, confinement of lithium polysulfide, and redox kinetics. Herein, a pomegranate-inspired dual-stage porous graphene foam with large cavity skeletons and submicron small cavity carriers is obtained by combining the bubble-template and the flash-freezing ice-template. Furthermore, the redox mediator of poly (1-aminoanthraquinone) is electropolymerized onto it to obtain a sulfur host with high specific surface area and strong confinement capability. The nanosized sulfur deposition behavior and the rich electronic relay capability of the sulfur host are analyzed by the independent gradient model combined with in-situ Raman spectroscopy. Besides, a few of electro-depolymerization products of poly (1-aminoanthraquinone) are confirmed to be able to migrate to the anode surface to stabilize lithium deposition. With these merits, the assembled lithium sulfur batteries exhibit a discharge capacity of 1214 mAh g−1 and a high capacity retention rate of 75.1 % after 800 cycles at a sulfur loading of 10.0 mg cm−2, and provide an excellent areal capacity of 17.5 mAh cm−2 (867 mAh g−1) even at an ultra-high sulfur loading of 20.1 mg cm−2.

Absorption Chiller Efficiency Increasing at Waste Heat Recovery of Cogen-eration Plants using a Vapor Compression Chiller as a Backup Cold Source

Cooling efficiency in a trigeneration cycles is one of the most frequently problems. A single-stage hot water absorption chiller is a universal solution for the gas engine waste heat utilization. How-ever, trigeneration centers exploitation experience shows the need for a more detailed technical and economic analysis to achieve maximum efficiency of the three-generation cycle. The purpose of this article is to substantiate the use of two-stage absorption cycles increasing the utilization rate of waste gas heat. The study presents the results of assessing the efficiency of using energy resources on the example of a three-generation center with more than 14 years of exploitation experience, including two single-stage absorption chillers (Q0= 0.65 MW each) and a chiller with a screw compressor (Q0= 0.8 MW). A comparison of actual energy consumption with designed indicators was carried out. The influ-ence of external and internal factors on the efficiency of heat recovery from a gas engine in single-stage absorption refrigeration machines with a water-heated generator is considered. The dependence of the change in the average annual cooling coefficient during the calendar year was obtained. Based on the study, proposals were formulated to improve the energy efficiency of refrigeration supply, giving an increase in refrigeration capacity of 6.1% with a commensurate reduction in the load on the screw chiller. The option of using a combined two-stage cycle chiller (type 2) under similar operating conditions showed a COP=1.43, which provides savings of 67.45 MWh of electricity or 5.4%, on an annualized basis (assuming only waste heat is used). Additional advantages of such a cycle include increased accu-racy and stability of maintaining the water temperature at the absorption chiller evaporator due to more flexible adjustment of the temperature of the lithium bromide solution in the high-temperature generator.

Digital twin with augmented state extended Kalman filters for forecasting electric power consumption of industrial production systems

The work aims to develop an effective tool based on Digital Twins (DTs) for forecasting electric power consumption of industrial production systems. DTs integrate dynamic models combined with Augmented State Extended Kalman Filters (ASEKFs) in a learning process. The connection with the real counterpart is realized exclusively through non-intrusive sensors. This architecture enables the model development of industrial systems (components, machinery and processes) on which complete knowledge is not available, by identifying the model's unknown parameters through short online training phases and small amounts of real-time raw data. ASEKFs track the unknowns keeping models updated as physical systems evolve. When a forecast is needed, the current estimates of the uncertain parameters are integrated into the dynamic models. These can then be used without ASEKFs to predict the actual energy use of the system under the desired operating conditions, including scenarios that differ from typical functioning. The approach is validated offline with reference to the electricity consumption of an automatic coffee machine, which represents a real test environment and a blueprint to design DTs for other industrial systems. The appliance is observed by measuring the supply voltage and the absorbed current. The accuracy of the results is analyzed and discussed. This method is developed in the context of energy consumption prediction and optimization in the manufacturing industry through refined energy management and planning.

End-to-end learning of representative PV capacity factors from aggregated PV feed-ins

Energy system models rely on accurate weather information to capture the spatio-temporal characteristics of renewable energy generation. Whereas energy system models are often solved with high abstraction of the actual energy system, meteorological data from reanalysis or satellites provides rich gridded information of the weather. The mapping from meteorological data to renewable energy generation usually relies on major assumptions as for solar photovoltaic energy the photovoltaic module parameters. In this study, we show that these assumptions can lead to large deviations between the reported and estimated energy, as shown for the case of photovoltaic energy in Germany. We propose a novel gradient-based end-to-end framework that can learn local representative photovoltaic capacity factors from aggregated PV feed-ins. As part of the end-to-end framework, we compare physical and neural network model formulations to obtain a functional mapping from meteorological data to photovoltaic capacity factors. We show that all the methods developed have better performance than commonly used reference methods. Both physical and neural network models have much better performance than reference models whereas operational use cases may prefer the neural network due to higher accuracy while interpretable, physical models are more suited to academic settings.

Physics-Based Modeling and Parameter Tracing for Industrial Demand-Side Management Applications: A Novel Approach

The transition to sustainable energy sources presents significant challenges for energy distribution and consumption systems. Specifically, the intermittent availability of renewable energy sources and the decreasing usage of fossil fuels pose challenges to energy flexibility and efficiency. An approach to tackle these challenges is demand-side management, aiming to adapt energy consumption and demand. A key requirement for demand-side management is the traceability of the energy flow among individual energy consumers. In recent years, advancements in industrial information and communication technology have provided additional potential for data acquisition. Complementary to acquired data, a physics-based modeling and analysis approach is proposed, which describes the energy consumption with physical parameters. This results in comprehensive options for monitoring actual energy consumption and planning future energy demand supporting energy efficiency and demand-response goals. To validate the proposed approach, a case study with a 3D printer covering approximately 110 h of active printing time is conducted. The medium-term study results indicate a consistent parameter trend over time, suggesting its conceptual suitability for industrial application. The approach helps to monitor energy efficiency among manufacturing assets by identifying peak loads and consumption hotspots, and provides parameters to estimate energy consumption of manufacturing processes. Results indicate up to 50% energy savings when switching the printing material and indicate further potentials.

Performance improvement of Li-rich Mn-based materials with the submicron-sized single-crystal morphology and cationic doping

Lithium-rich manganese-based cathode materials (LRMCs) can be regarded as one of the most potential cathode materials for the next-generation high-energy Li-ion batteries due to the extra oxygen redox, which can greatly increase the actual output energy density of LRMCs. Nevertheless, the structure degradation and inevitable loss of oxygen will be triggered by the irreversible oxygen redox, which will cause a significant decline in cyclic performance and limited initial Coulombic efficiency, thus restricting their industrial applicability. Herein, the LRMCs with the submicron-sized single-crystal morphology and sodium doping are prepared by molten salt treatment with NaCl. On one hand, the submicron-sized single-crystal cathode effectively reduces the diffusion path for Li+ due to the smaller particle dimensions. On the other hand, Na+ doping expands the Li slab spacing, promoting the transport of Li+. The results show that the as-prepared material can purposefully alleviate oxygen release, improve structure stability, and increase the Li+ diffusion rate for the pristine LRMCs. Moreover, the as-prepared material in this study exhibits a higher reversible capacity of 282.8 mAh g−1 at 0.1 C and displays good cyclic performance, retaining a capacity rate of 80.9% after 200 cycles at 1 C. Therefore, this study offers a potential exploration for designing LRMCs with high energy density and excellent cyclic performance, which are achieved through building submicron-sized single-crystal morphology and introducing cationic doping.

Determination of energy requirements after minor burns using indirect calorimetry: A descriptive cohort study.

Minor burns could be associated with moderate hypermetabolism. In this study, the primary outcome was measured energy expenditure (mEE) determined by indirect calorimetry in patients with minor burns. We also compared mEE with predictive values and actual energy intakes. Adults with minor burns exclusively treated on an outpatient basis were included. During the week following injury, a dietitian performed indirect calorimetry (Q-NRG in canopy mode), calculated the estimated energy expenditure (eEE) based on the Harris-Benedict (HB) and Henry formulas, and evaluated daily energy intakes using a food anamnesis. Forty-nine patients (59.2% male; median age: 35 [interquartile range: 29-46.5] years; body mass index [BMI]: 26.2 [22.3-29.6] kg/m2; burn surface area [BSA]: 1.5% [1%-2%]) were included 4 (2-6) days after injury. The mEE was 1863 (1568-2199) kcal or 25 (22.4-28.5) kcal/kg and 1838 (1686-2026) kcal or 26.1 (23.7-27.7) kcal/kg in patients who were respectively fasting for >10 h or not (P = 0.991 or P = 0.805). The total mEE was 104% (95%-116%) and 108% (99%-122%) of the total eEE using the HB and Henry formulas, respectively, with diet-induced thermogenesis and physical activity level. Hypermetabolism (ie, oxygen consumption at rest ≥3.5 ml/kg/min) was observed in 21/49 (42.9%) patients. Energy intakes corresponded to 71% (60%-86%) of the total mEE. Performing indirect calorimetry in adults with minor burns revealed that ≥40% of the tested adults presented a hypermetabolism and that their mEE was not covered by their energy intakes.

A novel energy reconstruction method for the MAGIC stereoscopic observation

We report the successful development of a novel methodology of energy reconstruction for very high energy gamma rays detected with Imaging Atmospheric Cherenkov Telescopes (IACTs). This methodology, based on the machine learning algorithm Random Forest, and named RF-Erec, has been adjusted for being used with data from the Major Atmospheric Gamma-ray Imaging Cherenkov (MAGIC) stereo telescope system, which is a worldwide leading instrument for gamma-ray astronomy in the energy range from about 20GeV to beyond 100TeV.The RF-Erec has been evaluated using different realistic scenarios with Monte Carlo simulated data and real observations from the Crab Nebula (the standard candle for the VHE gamma-ray community). This new methodology has been validated by the MAGIC software board, and it is implemented and ready-to-use in the MAGIC Analysis and Reconstruction Software (MARS). This new methodology, validated by the MAGIC software board, has been implemented and is ready for use in the MAGIC Analysis and Reconstruction Software (MARS). We demonstrate that, in comparison to the previous energy reconstruction methodology for MAGIC data, which relied on Look-Up-Tables (LUTs- Erec) and has been utilized in over 100 scientific publications over the last decade, RF-Erec significantly enhances the energy reconstruction of gamma rays. This improvement extends the capabilities of the MAGIC telescopes.Specifically, when quantifying the energy resolution with the width of a Gaussian fitted to the error distribution (resolution-σ), the RF-Erec energy resolution-σ is 20% at 100 GeV and 11% above 1 TeV for Zenith distances (Zd) below 35 degrees, while it is 20% at 1TeV and 13% above 10TeV for Zd above 55 degrees. For a wide range of the observable energies, the improvement of energy resolution-σ, compared to LUTs-Erec, reaches roughly a factor of two, and the improvement is even larger for high Zd observations. Differently to many other works in the literature, our evaluation also considers the energy dispersion and the actual energy migration of events, where RF-Erec improves the performance of LUTs-Erec by factors of a few. The manuscript also demonstrates the importance of energy reconstruction methods with a small energy migration in order to prevent the appearance of artificial spectral features. These artifacts are particularly important at the high end of the gamma-ray spectra, where a few extra high-energy photons could have dramatic consequences for studies related to the EBL attenuation, Lorentz invariance violation, or searches for Axion-like-particles.

Kinetic Promoters for Sulfur Cathodes in Lithium-Sulfur Batteries.

ConspectusLithium-sulfur (Li-S) batteries have attracted worldwide attention as promising next-generation rechargeable batteries due to their high theoretical energy density of 2600 Wh kg-1. The actual energy density of Li-S batteries at the pouch cell level has significantly exceeded that of state-of-the-art Li-ion batteries. However, the overall performances of Li-S batteries under practical working conditions are limited by the sluggish conversion kinetics of the sulfur cathodes. To overcome the above challenge, various kinetic promotion strategies have been proposed to accelerate the multiphase and multi-electron cathodic redox reactions between sulfur, lithium polysulfides (LiPSs), and lithium sulfide. Nowadays, kinetic promoters have been massively employed in sulfur cathodes to achieve Li-S batteries with high energy densities, high rates, and long lifespans. A comprehensive and timely summary of cutting-edge kinetic promoters for sulfur cathodes is of great essence to afford an in-depth understanding of the unique Li-S electrochemistry.In this Account, we outline the recent efforts on the design of sulfur cathode kinetic promoters for advanced Li-S batteries. The latest progress is discussed in detail regarding heterogeneous, homogeneous, and semi-immobilized kinetic promoters. Heterogeneous promoters, representatively known as electrocatalysts, function mainly by reducing the energy barriers of the interfacial electrochemical reactions. The working mechanism, activity regulation strategies, and reconstitution/deactivation processes of the heterogeneous promoters are reviewed to provide guiding principles for rational design. In comparison, homogeneous promoters are able to fully contact with the reaction interfaces and regulate the electron/ion-inaccessible reactants in working Li-S batteries. Redox mediators and redox comediators are typical homogeneous promoters. The former establishes extra chemical reaction pathways to circumvent the originally sluggish steps and boost the overall kinetics, while the latter fundamentally modifies the LiPS molecules to enhance their redox kinetics. For semi-immobilized promoters, the active units are generally anchored on the cathode substrate through flexible chains with mobile characteristics. Such a design endows the promoter with both heterogeneous and homogeneous characteristics to comprehensively regulate the multiphase sulfur redox reactions involving both mobile and immobile reactants.Overall, this Account summarizes the fundamental electrochemistry, design principles, and practical promotion effects of the various kinetic promoters used for the sulfur cathodes in Li-S batteries. We believe that this Account will provide an in-depth and cutting-edge understanding of the unique sulfur electrochemistry, thereby providing guidance for further development of high-performance Li-S batteries and analogous rechargeable battery systems.

Energy Measurement and Potential Energy Conservation Measures in Five Healthcare Buildings in Malaysia

Concern is growing around the world as a result of the global economy's overall rise in carbon emissions and energy consumption. The healthcare facility is among the buildings in the world with the highest energy use. Thus, it is necessary to create energy-saving plans that evaluate the energy performance of buildings and put energy-saving measures into practice. Enhancing building energy efficiency is most commonly achieved through a comprehensive energy audit. The Building Energy Index is the current method used to analyses all forms of building energy performance. The operating hours of a healthcare facility are longer than those of a commercial building. The issue emerges when there is no clear criterion for establishing whether or not a healthcare facility is energy efficient, and the present norm may misread healthcare energy consumption. To determine the actual onsite energy use at the healthcare building, a full energy audit is performed. Five places are selected based on the sort of healthcare services offered. According to the on-site complete energy audit, cooling equipment consumes the most energy. The study's findings may be divided into two categories: Healthcare Building Energy Index from detailed energy audit results and derived directly from power bills for the years 2018 and 2019. In terms of Healthcare Building Energy Index, there is a contrast between utilizing detailed energy audit data and using utility bills directly. Nonetheless, a comparison of Healthcare Building Energy Index with existing Malaysian requirements reveals a significant discrepancy. Based on the audit and analysis results, several savings opportunities may be presented.

Calculation of Charge Transport Phenomena and Mobility Analysis in the Insulating Oil Using First Principle and Marcus Theory.

Oil-paper insulation is widely used as a reliable composite insulation system in power transformers. The dielectric property of oil insulation plays an important role in the reliable operation of power equipment. To recognize the charge transfer process in composite insulation, the mobility of the charge in aged insulating oil is studied. However, few studies have been conducted on the microscopic mechanism of charge transport phenomena at the molecular level. In this research, we have studied the molecular electronic structure and the distribution of holes and electrons in the insulating oil by first-principles calculation. By combining with Marcus theory, the corresponding electron coupling energy, reorganization energy, and free energy are obtained. The corresponding charge hopping model is chosen by the parameter relation, and the hopping rate is calculated. At last, the mobility of holes and electrons in insulating oil within the insulation is simulated by the Monte Carlo method. Other possible charge migration methods are also studied and discussed for the comparison. It is observed that the transfer integral of electrons is 2 orders of magnitude larger than that of holes, which is mostly due to the localization of lowest unoccupied molecular orbitals (LUMO). The hole and charge transfers accord with Marcus hopping, the adiabatic charge transfer model, and the charge hopping rate is obtained. The actual free energy action under an external electric field is obtained by calculating polarizability and permittivity. Monte Carlo simulation is used to obtain the charge transfer image and mobility under an actual electric field. Possible types of traps and mobility of ions and clusters in the insulating oil are also studied.

A modular framework for a dynamic residential building stock model with energy retrofit forecasts

Building retrofits are strongly dependent on the individual refurbishment/rehabilitation cycles of buildings. In order to achieve the targeted climate goals for the building sector, it is necessary to make the best possible use of refurbishment opportunities for energy retrofits. Furthermore, it must be considered that the younger the building, the lower the potential for energy savings. The most challenging, yet required, energy retrofits are all yet to come. Given the importance of the refurbishment cycle, the question arises as to what the theoretical refurbishment rate is, what the actual energy retrofit rate is, and what it needs to be in order to meet the climate targets for the residential building stock. The aim of this paper is to model the evolution of the size and retrofit status of national building stocks. This is to be done against the background of the deviation from the actual and theoretical refurbishment cycle. Using current statistics and the past development of the German residential building stock, central parameters like living space, new construction, deconstruction, and the retrofit rate are modeled endogenously. For the latter, influences of economic conditions are implemented through an exogenously set black box variable. The retrofit rate represents a deep energy retrofit and thus allows an easy connection of the model with energy parameters for building classes defined in the web database TABULA for 20 European countries. The results are of relevance for modeling the energy efficiency status of the building stock and deriving suitable conditions for increasing retrofit rates also in other countries. The results show that increasing the energy retrofit rates to 2%/a or more is largely unrealistic. Instead, the focus should be on weak points, especially where short-term savings are concerned. This applies in particular to facades, basement ceilings, and heating systems as well as the efficient use of energy in buildings.

Machine Learning Prediction of Global Photovoltaic Energy in Spain

The growing presence of solar energy in the electrical systems of many countries has made its accurate forecasting an important issue. In this work we will explore the application of Support Vector Regression (SVR), an advanced Machine Learning modelling tool, to forecast the daily photovoltaic generation of Spain. Given the very large geographical spread of photovoltaic installations, we will use as input features NWP forecasts of relevant meteorological variables for the entire Iberian Peninsula. The input dimension is thus very large but, while further work is needed, our results show SVR to be an effective tool to deal with the problem's underlying dimension, yield useful forecasts and further provide some insights on the relationship between NWP and actual solar energy production.

Comprehensive Analysis of Power System

The contemporary power landscape is marked by escalating complexity driven by surging energy demands, the integration of renewable sources, and the imperative for heightened system performance. In response to these challenges, this paper presents a thorough investigation. Firstly, the load factor Lambda (λ) is expounded upon, emphasizing its significant impact on system stability due to load characteristics. Static load models are employed for voltage stability studies, providing insights into resource allocation and optimization by comparing actual energy consumption to the maximum potential consumption. Secondly, the paper delves into the power balance, a critical aspect that scrutinizes the equilibrium among generation, consumption, and distribution, underscoring its pivotal role in ensuring system stability. Within a 6-node network, the concept of power balance, denoted as "load λ," is illustrated by tracking its variations across iterations, representing the disparity between Power Generator (PG) and Load (PL). To maintain balance during continuous power flow (CPF) analysis, power supply can be finely adjusted using the load factor lambda (λ). Thirdly, the integration of renewable sources introduces active load variation, underscoring the necessity to comprehend load fluctuations over time for ensuring grid reliability. Active load variation is demonstrated based on the number of iterations. Additionally, the paper elucidates the increment factor τ, explaining its impact on the number of correction iterations by selecting the size of the step factor. The graphical representation of the increment factor for iterations 8 and 11 in a network with load nodes provides further clarification. This paper explores the intricate interactions of load factors, power balance, active load changes, and increment factors within power systems. The presented findings contribute to the enhancement of the reliability, efficiency, and sustainability of modern power systems.

Improving energy efficiency in ammonia production plants using machine learning

Energy efficiency is becoming increasingly important nowadays due to the need for energy conservation and environmental sustainability. An existing ammonia plant was simulated using Aspen Hysys software, the simulated plant was used to produce large volumes of data to train and test our machine-learning model. In this work a benchmark methodology is proposed through machine-learning (ML) techniques to identify patterns and anomalies in energy consumption. Our ML model was developed in Python programming language using a multiple linear regression algorithm. Microsoft Power BI was used to build interactive visualizations to illustrate insights to users. The ML model was able to predict energy consumption by developing equations that relate the energy consumption and the operating variables for each significant energy user in the ammonia plant. In this study, actual versus optimum energy consumption was analyzed for four ammonia production plants. The ML model identified the ammonia plant operating costs and potential savings by adjusting operating conditions. An annual saving of up to 3.9 million dollars was reached in one of the ammonia production plants operating costs. A reduction in carbon dioxide emissions of up to 4.7 ton per hour was achieved due to energy efficiency adjustment.

Detection of Anomalies in Daily Activities Using Data from Smart Meters.

The massive deployment of smart meters in most Western countries in recent decades has allowed the creation and development of a significant variety of applications, mainly related to efficient energy management. The information provided about energy consumption has also been dedicated to the areas of social work and health. In this context, smart meters are considered single-point non-intrusive sensors that might be used to monitor the behaviour and activity patterns of people living in a household. This work describes the design of a short-term behavioural alarm generator based on the processing of energy consumption data coming from a commercial smart meter. The device captured data from a household for a period of six months, thus providing the consumption disaggregated per appliance at an interval of one hour. These data were used to train different intelligent systems, capable of estimating the predicted consumption for the next one-hour interval. Four different approaches have been considered and compared when designing the prediction system: a recurrent neural network, a convolutional neural network, a random forest, and a decision tree. By statistically analysing these predictions and the actual final energy consumption measurements, anomalies can be detected in the undertaking of three different daily activities: sleeping, breakfast, and lunch. The recurrent neural network achieves an F1-score of 0.8 in the detection of these anomalies for the household under analysis, outperforming other approaches. The proposal might be applied to the generation of a short-term alarm, which can be involved in future deployments and developments in the field of ambient assisted living.