Energy Optimization Model Research Articles

Time series aggregation – A new methodological approach using the “peak-load-pricing” model

Coping with fluctuating renewables and large-scale integrated systems raises new challenges related to achieving simultaneously manageable computation times and sufficient levels of detail in energy optimization models. The present contribution proposes a new methodological approach for identifying appropriate time series aggregation, which may complement heuristic and numerical approaches. The basic idea is to define an evaluation criterion that assesses the quality of the solution of the model based on the aggregated time series (“reduced problem”) compared to the solution of the original problem. Analytical results may be derived for a simplified version of this problem, known as the “peak-load-pricing” model.

Data-driven ship energy efficiency analysis and optimization model for route planning in ice-covered Arctic waters

As increasing numbers of merchant ships navigate in the Arctic waters, more energy efficient navigation in the Arctic is needed for both economic and environmental purposes. This paper aims to provide a comprehensive analysis of energy efficiency of ice-going ships. Firstly, a data-driven model based on neural network theory is developed to predict the energy efficiency of ships in the Arctic. Then, a route with optimum energy efficiency is presented, intended to cut costs and be as environmentally friendly as possible. Finally, a case study is carried out to analyze the performance of a case study ship sailing at various environmental conditions. The results indicate that when planning the route for such an Arctic ship, it would be shortsighted only to focus on distance rather than energy efficiency. Moreover, the model shows good agreement between the simulation and real-life navigation. It is expected that the construction of this multi-angle, energy efficiency optimization model could help to improve Arctic navigation.

Driving energy management of front-and-rear-motor-drive electric vehicle based on hybrid radial basis function

This paper presents mathematical methods to develop a high-efficiency and real-time driving energy management for a front-and-rear-motor-drive electric vehicle (FRMDEV), which is equipped with an induction motor (IM) and a permanent magnet synchronous motor (PMSM). First of all, in order to develop motor-loss models for energy optimization, database of with three factors, which are speed, torque and temperature, was created to characterize motor operation based on HALTON sequence method. The response surface model of motor loss, as the function of the motor-operation database, was developed with the use of Gauss radial basis function (RBF). The accuracy of the motor-loss model was verified according to statistical analysis. Then, in order to create a two-factor energy management strategy, the modification models of the torque required by driver (Td) and the torque distribution coefficient (β) were constructed based on the state of charge (SOC) of battery and the motor temperature, respectively. According to the motor-loss models, the fitness function for optimization was designed, where the influence of the non-work on system consumption was analyzed and calculated. The optimal β was confirmed with the use of the off-line particle swarm optimization (PSO). Moreover, to achieve both high accuracy and real-time performance under random vehicle operation, the predictive model of the optimal β was developed based on the hybrid RBF. The modeling and predictive accuracies of the predictive model were analyzed and verified. Finally, a hardware-in-loop (HIL) test platform was developed and the predictive model was tested. Test results show that, the developed predictive model of β based on hybrid RBF can achieve both real-time and economic performances, which is applicable to engineering application. More importantly, in comparison with the original torque distribution based on rule algorithm, the torque distribution based on hybrid RBF is able to reduce driving energy consumption by 9.51% under urban cycle.

A realistic energy optimization model for smart‐home appliances

Smart homes have the potential to achieve optimal energy consumption with appropriate scheduling. The control of smart appliances can be based on optimization models, which should be realistic and efficient. However, increased realism also implies an increase in solution time. Many of the optimization models in the literature have limitations on the types of appliances considered and/or their reliability. This paper proposes a home energy management scheduling model that is more realistic and efficient. We develop a mixed integer linear optimization model that minimizes the energy cost while maintaining a given level of user comfort. Our main contribution is the variety of specific appliance models considered and their integration into a single model. We consider the use of energy in appliances and electric vehicles (EVs) and take into account renewable local generation, batteries, and demand response. Our models of a shower, a fridge, and a hybrid EV consider both the electricity consumption and the conventional fuel cost. We present computational results to validate the model and indicate how it overcomes the limitations of other models. Our results, compared with the best competitors, provide cost savings ranging from 8% to 389% over a horizon of 24 hours.

Occupant Comfort Management Based on Energy Optimization Using an Environment Prediction Model in Smart Homes

Occupant comfort management is an important feature of a smart home, which requires achieving a high occupant comfort level as well as minimum energy consumption. Based on a large amount of data, learning models enable us to predict factors of a mathematical model for deriving the optimal result without expensive experiments. Comfort management supports high-level comfort to the occupant in the individual indoor environment, using the optimal power consumption to run home appliances. In this paper, we propose occupant comfort management based on energy optimization, using an environment prediction model. The proposed energy optimization model provides optimal power consumption based on the proposed objective function, which requires temperature and comfort index data as the input parameters. For the input requirement, temperature prediction model and humidity prediction model are presented based on a recurrent neural network with a pre-collected dataset, including indoor and outdoor temperature and humidity sensing data. Using the predicted temperature and humidity data, the comfort index model derives the predicted mean vote value to be used in the energy optimization model with the predicted temperature data. The experimental results present an 8.43% reduction of the optimized power consumption compared to the actual power consumption using mean absolute percentage error to calculate. Moreover, the emulation of an indoor environment using optimal energy consumption presents as approximately similar to the actual data.

Smart Residential Load Simulator for Energy Management in Smart Grids

This paper describes the development of a freeware smart residential load simulator to facilitate the study of residential energy management systems in smart grids. The proposed tool is based on MATLAB-Simulink-GUIDE toolboxes and provides a complete set of user-friendly graphical interfaces to properly model and study smart thermostats, air conditioners, furnaces, water heaters, stoves, dishwashers, cloth washers, dryers, lights, pool pumps, and refrigerators, whose models are validated with actual measurements. Wind and solar power generation as well as battery sources are also modeled, and the impact of different variables, such as ambient temperature and household activity levels, which considerably contribute to energy consumption, are considered. The proposed simulator allows modeling of appliances to obtain their power demand profiles, thus helping to determine their contribution to peak demand, and allowing the calculation of their individual and total energy consumption and costs. In addition, the value and impact of generated power by residential sources can be determined for a 24-h horizon. This freeware platform is a useful tool for researchers and educators to validate and demonstrate models for energy management and optimization, and can also be used by residential customers to model and understand energy consumption profiles in households. Some simulation results are presented to demonstrate the performance and application of the proposed simulator.

A novel DEACM integrating affinity propagation for performance evaluation and energy optimization modeling: Application to complex petrochemical industries

Data Envelopment Analysis (DEA) has been widely used in performance and energy efficiency evaluation. However, in the traditional DEA, the effective of each decision making unit (DMU) is evaluated through its own optimized perspective and regardless of other DMUs influence, which may result in too many effective DMUs. And the DEA cross-model (DEACM) can distinguish the effective DMUs better by constructing a cross-efficiency matrix, but the optimal weight of the DMU may not be unique, so the cross efficiency of the DEACM may be different. Therefore, this paper proposes a novel DEACM based on the affinity propagation (AP) clustering algorithm (AP-DEACM). Through the AP clustering algorithm, the high impact data affecting the performance capacity and energy saving are obtained. Then the better effective DMU is identified through the high discrimination of the AP–DEACM. Finally, the proposed AP-DEACM is used for performance evaluation and energy optimization modeling of the Pure Terephthalic Acid (PTA) production process and the ethylene industrial process in complex petrochemical industries. The experimental results show that the energy saving potential of PTA production plants and ethylene production plants are 2.78% and 1.26%, respectively, and the average value of carbon emission savings potential is 3.62% in ethylene production plants.

Enhancing aquaponics management with IoT-based Predictive Analytics for efficient information utilization

Modern aquaponic systems can be highly successful, but they require intensive monitoring, control and management. Consequently, the Automation Pyramid (AP) with its layers of Supervisory Control and Data Acquisition (SCADA), Enterprise Resource Planning (ERP) and Manufacturing Execution System (MES) is applied for process control. With cloud-based IoT-based Predictive Analytics at the fore marsh, it is worth finding out if IoT will make these technologies obsolete, or they can work together to gain more beneficial results. In this paper, we will discuss the enhancement of SCADA, ERP and MES with IoT in aquaponics and likewise how IoT-based Predictive Analytics can help to get more out of it. An example use case of an aquaponics project with five demonstration sites in different geographical locations will be presented to show the benefits of IoT on example Predictive Analytics services. Innovative is the collection of data from the five demonstration sites over IoT to make the models of fish, tomatoes, technical components such as filters used for remote monitoring, predictive remote maintenance and economical optimization of the individual plants robust. Robustness of the various models, fish and crop growth models, models for econometric optimization were evaluated using Monte Carlo Simulations revealing as expected the superiority of the IoT-based models. Our analysis suggest that the models are generally tolerant to the temperature coefficient variations of up to 15% and the econometric models tolerated a variation of for example feed ration size for fish of up to 4% and by the energy optimization models a tolerance of up to 14% by variations of solar radiation could be noticed. Furthermore, from the analysis made, it can be concluded that MES has several capabilities which cannot be replaced by IoT such as responsiveness to trigger changes on anomalies. It act as proxy when there is no case for sensors and reliably ensure correct execution in the aquaponics plants. IoT systems can produce unprecedented improvements in many areas but need MES to leverage their true potential and benefits.

An energy-efficient process planning system using machine-monitoring data: A data analytics approach

This paper presents a system development of incorporating Computer-Aided Process Planning (CAPP) with energy-efficient machining based on a hybrid approach to take advantage of Generative Process Planning (GPP) and Variant Process Planning (VPP) and compensate for the drawbacks of both GPP and VPP. The GPP decides process plans without human assistance through decision-making algorithms in computers but lacks in ensuring the models’ robustness for different machining conditions. The VPP adopts group technology by reusing existing plans through the identification and classification of part family but does not support predictive and optimum decision-making. The developed Energy-Efficient Process Planning System (EEPPS) builds upon data analytics to efficiently process the machine-monitoring data collected from real machine tool’s operations and to develop energy prediction and optimization models based on historical machine-monitoring data. Particularly, those energy prediction and optimization models allow process planners to anticipate the energy consumed during executing a numerical control program and optimize process parameters at the level of machining features for minimizing energy use. This paper also presents a prototype implementation to show the feasibility of the proposed EEPPS.

Optimization of an Intelligent Community Hybrid Energy System With Robustness Consideration

An intelligent community hybrid energy system (ICHES) includes renewable energy power generation equipment, distributed controllable generation equipment, energy storage systems, mobile agents, and large numbers of fixed loads. In recent years, many studies of hybrid energy systems have been considered from the perspective of cost. However, the pollution and robustness of the system cannot be ignored. To address these issues, this paper introduces the environmental impact ratio of power generation and power supply robustness into the model and proposes a novel intelligent community hybrid energy optimization model. The proposed model is a multi-objective optimization problem. To address this complex multi-objective optimization problem, a direction vector adjustment (DVA) mechanism is introduced into the multi-objective evolutionary algorithm based on decomposition (MOEA/D) using localized penalty-based boundary intersection (LPBI) (MOEA/D-LPBI). Then, an improved MOEA/D-LPBI-DVA is proposed. The experimental results show that our model is more reasonable and outperforms existing models. The solutions obtained allow the problem to have a better effect, thereby effectively optimizing the hybrid energy output and achieving the multi-objective optimization requirements of low cost, low pollution, and high robustness of the system power supply.

A sensor node scheduling algorithm for heterogeneous wireless sensor networks

To improve the regional coverage rate and network lifetime of heterogeneous wireless sensor networks, a sensor node scheduling algorithm for heterogeneous wireless sensor networks is proposed. In sensor node scheduling algorithm, heterogeneous perception radius of sensor node is considered. Incomplete coverage constraint and arc coverage interval are analyzed. Regional coverage increment optimization model, arc coverage increment optimization model, and residual energy optimization model are proposed. Multi-objective scheduling model is established using weight factors and integrated function. Furthermore, the heuristic method is proposed to solve the multi-objective optimization model, and scheduling scheme of heterogeneous sensor nodes is obtained. When the network is in operation for a period of time, some sensor nodes are invalid and relevant regions are uncovered. The repair method is proposed to wake up sleep sensor nodes and repair the coverage blind area. The simulation results show that if keeping the same regional coverage rate, sensor node scheduling algorithm improves network lifetime, increases number of living sensor nodes, and keeps average node energy consumption at a low level. Under certain conditions, sensor node scheduling algorithm outperforms DGREEDY, two-tiered scheduling, and minimum connected cover.

Energy-environment-economy nexus in Pakistan: Lessons from a PAK-TIMES model

Pakistan, alongside other developing countries, is struggling to understand, plan and realize the energy-environment-economic nexus. In this study, we develop a national-scale bottom-up energy optimization model (Pak-TIMES) employing ANSWER-TIMES modeling framework which provides insight into this nexus for Pakistan. The reference energy system of Pak-TIMES model provides energy demand, supply, cost of energy and GHGs emissions. In total, five supply side scenarios enacted in this study are BAU (Business As Usual), RE-30 (30% renewables), RE-40 (40% renewables), Coal-30 (30% Coal) and Coal-40 (40% Coal) for the period 2015–2035. The results suggest that the cumulative investment cost incurred will be 170 billion USD to meet the electricity demand (3091 PJ) by 2035 with 181.5 million tons of GHGs emission in the BAU scenario. The most environmentally ambitious pathway, the RE-40 scenario will require 179.2 billion USD for investment with 96.0 million tons of GHGs emission. The Coal-40 scenario will cause the highest amount of GHGs emissions, i.e., 338.9 million tons with an estimated investment cost of 184.8 billion USD. Based on the results of this study, PAK-TIMES model, it is recommended that government needs to invest in renewables; given that, the environmental cost of non-renewable energy solutions is significantly higher.

Production capacity analysis and energy optimization of complex petrochemical industries using novel extreme learning machine integrating affinity propagation

With the rapid political and economic developments, energy saving and carbon dioxide emission reduction are now recognized as the most important goals worldwide. Therefore, this paper presents a production capacity analysis and energy optimization model that uses novel extreme learning machine integrating affinity propagation clustering. By using the affinity propagation method, clusters of raw data can be obtained automatically to reduce multi-dimensional data and fuse the high-similarity data. Then, the clustering results are taken as the training and testing sets of the extreme learning machine to improve prediction accuracy and analyze production capacity. Through comparisons with the extreme learning machine, back propagation, radical basis function, and extreme learning machine integrated K-means clustering algorithm, the accuracy and validity of the proposed method are verified by using the University of California Irvine data sets. Finally, the proposed method is applied to build the production capacity analysis and energy optimization model of the ethylene and purified terephthalic acid production systems in complex petrochemical industries. The prediction accuracy of the ethylene and purified terephthalic acid production is approximately 99%, thus improving the energy efficiency of these complex petrochemical processes and achieving energy saving and carbon dioxide emission reduction.

Consequences of the National Energy Strategy in the Mexican Energy System: Analyzing Strategic Indicators with an Optimization Energy Model

This paper presents an approach to the assessment of the Mexican energy system’s evolution under the climate and energy objectives set by the National Climate Change Strategy using an energy optimization model. Some strategic indicators have been chosen to analyze the performance of three integration elements: sustainability, efficiency, and energy security. Two scenarios have been defined in the medium and long-term: the business as usual scenario, with no energy or climate targets, and the National Climate Change Strategy scenario, where clean energy technologies and CO2 emissions objectives are considered. The aim of this work is the analysis of some of those strategic indicators’ evolution using the EUROfusion Times Model. Results show that reaching the strategy targets leads to improvements in the integration elements in the medium and long term. Besides, meeting the CO2 emission limits is achievable in terms of technologies and resources availability but at a high cost, while clean technologies targets are met with no extra costs even in the business as usual scenario.

Soft-linking energy demand and optimisation models for local long-term electricity planning: An application to rural India

Rural electricity plans are usually designed by relying on top-down rough and aggregated estimations of the electricity demand, which may fail to capture the real dynamics of local contexts. This study aims at soft-linking a bottom-up approach for short- and long-term forecasts of load profiles with an energy optimisation model in a more comprehensive rural energy planning procedure. The procedure is applied to a small Indian community, and it is based on three blocks: (i) a bottom-up model to project households' electrical appliances, which adopts socio-economic indicators to make long-term projections; (ii) a stochastic load profile generator, which employs correlations and users’ habits for assessing the coincidence and load factors; (ii) an energy optimisation model based on OSeMOSYS to find the economic optimum. The simulations show that demand models based on socio-economic indicators lead to more structured and less arbitrary scenarios. The soft-link with the energy optimisation model confirms that when accounting for short- and long-term variabilities of electricity demand together, the optimal capacities and costs can vary up to 144% and 50% respectively. Integrating optimisation tools to bottom-up models based on socio-economic indicators for forecasting electricity demand is therefore pivotal to set more reliable investments plans in rural electrification.

Energy optimization of parallel programs in a heterogeneous system by combining processor core-shutdown and dynamic voltage scaling

Reducing power consumption and improving efficiency are important aspects of the development of supercomputers into large-scale systems. As a result, heterogeneous systems have become an important development trend in high-performance computing. From the perspective of heterogeneous systems, this study establishes a model for energy optimization of parallel programs (EOPP) and puts forward a method of using it. By considering the energy overheads caused by re-synchronization, voltage switching, and operations in critical sections, the model effectively combines processor core-shutdown and dynamic voltage scaling technologies, which can be applied in a heterogeneous system to guide the optimization process. The results show that the proposed model can effectively reduce the energy consumption of parallel programs. Moreover, increasing the proportion of operations in the critical section enhances the optimal frequency of a processor while decreasing the probability of conflicts in the critical section. It can thus provide optimization space for reducing the frequency of a processor which ultimately reduces the energy overhead of the system.

Dynamic Energy Optimization in Chip Multiprocessors Using Deep Neural Networks

We investigate the use of deep neural network (DNN) models for energy optimization under performance constraints in chip multiprocessor systems. We introduce a dynamic energy management algorithm implemented in three phases. In the first phase, training data is collected by running several selected instrumented benchmarks. A training data point represents a pair of values of cores’ workload characteristics and of optimal voltage/frequency (V/F) pairs. This phase employs Kalman filtering for workload prediction and an efficient heuristic algorithm based on dynamic voltage and frequency scaling. The second phase represents the training process of the DNN model. In the last phase, the DNN model is used to directly identify V/F pairs that can achieve lower energy consumption without performance degradation beyond the acceptable threshold set by the user. Simulation results on 16 and 64 core network-on-chip based architectures demonstrate that the proposed approach can achieve up to 55 percent energy reduction for 10 percent performance degradation constraints. In addition, the proposed DNN approach is compared against existing approaches based on reinforcement learning and Kalman filtering and found that it provides average improvements in energy-delay-product (EDP) of 6.3 and 6 percent for the 16 core architecture and of 7.4 and 5.5 percent for the 64 core architecture.

Efficient control and Fault Detection and Isolation in Building HVAC systems

My thesis work is focused on advancing the state of the art in smart building monitoring and control problems. A major part of my thesis will develop combined model- and data-driven models for the energy systems of buildings and apply them for energy monitoring and optimization, fault diagnosis, and fault-adaptive control. In past work, I have already developed initial machine learning-based models for energy optimization. In this paper, I am focusing on developing a framework using bond graphs to build a model of the HVAC system and understand how different factors affect the measurement variables of this system. This information will help us do subsequent fault diagnosis by studying the characteristics of the system behavior represented by the temporal causal graph. It will also provide the basis for fault-adaptive control of buildings

Air pollutant and CO2 emissions mitigation in urban energy systems through a fuzzy possibilistic programming method under uncertainty

Abstract To address the increasing tension between energy activities and environmental protection, an energy and environment optimization model based on fuzzy possibilistic programming method is proposed. The proposed model can (i) generate optimized solutions for energy and environmental systems; (ii) tackle the uncertainties expressed as fuzzy sets in the goal, left-hand (e.g., economic data), and right-hand (e.g., energy demands) sides of constraints; (iii) analyze the integrated and individual impacts of energy conversion efficiency, air pollutant removal rate, and electric power mix on energy, environmental, and economic systems; and (iv) explore cost-effective approaches for emissions reductions. The proposed model can be applied at city and regional scales, and its effectiveness was verified using Tianjin Municipality, Beijing Municipality, and Hebei Province. Several policy implications for local decision makers are suggested from the proposed model. Firstly, the optimization of the electric power mix and improvement of energy conversion efficiencies are cost-effective ways to reduce pollutant and CO2 emissions. Secondly, the improvement of pollutant removal rates can effectively reduce SO2, NOx, and particulate matter emissions, but with increased system costs. Thirdly, oil refining, coke processing, and coal-fired power are identified as major sectors for potential emissions reduction. The proposed model can effectively support local policymakers in the adjustment of current energy and environmental strategies sustainably and robustly.

Optimization of Cutting Parameters for Minimizing Environmental Impact: Considering Energy Efficiency, Noise Emission and Economic Dimension

Green manufacturing is attracting significant attention from the academic and industrial world under current environmental circumstances. The purpose of the present study is to evaluate trade-offs between the main factors in green manufacturing: energy, noise and cost, through cutting parameter optimization. First, the energy consumption model is developed based on the analysis of the relationship among sub-processes, power and cutting parameters. Then numerical noise emission model, which integrates orthogonal experiment and response surface method, is proposed. Moreover, Analysis of Variance results are employed to analyze the influence of cutting parameters on noise, the results show that the depth of cut is the dominating influencing factor of noise, then the cutting cost model is presented. The multi-objective optimization model for energy saving, noise reducing and cost saving is proposed, which takes cutting speed, feed fate and depth of cut as decision variables. NSGA-II is adopted to obtain the Pareto optimal solutions. The most suitable Pareto-optimal solution is selected by a combination weighting method. A case study involving the cutting process of a CNC lathe is used to validate the proposed methodology, and the results are discussed and analyzed.