A hybrid PSO-LSTM-based electricity prediction and optimization technique for home appliances

Impact of battery degradation on energy cost and carbon footprint of smart homes

Inefficient energy use has been a major issue globally. In Malaysia, the statistics reveal that residential energy consumption has been on a steady increase due to the growing population as well as a lack of awareness within households regarding proper energy utilization that causes significant amount of energy wastage. The emergence of Internet-of-Things (IoT) is a consequence and convergence of several key technologies such as real-time analytics, machine learning, sensors and embedded systems. The application of intelligence in IoT, known as Cognitive IoT (CIoT), will enable decision making based on historical data, and automatically train, learn and troubleshoot future issues. This project proposes a smart energy monitoring system for home appliances incorporating CIoT which consists of three parts. Firstly, a Raspberry Pi-based smart plug serving as the gateway, that is able to read current data from individual home appliances, load the trained model from training server and test the verified data using the model. Secondly, Google Colab as the training server will be used to store the training data set and building the Tensorflow-based Long Short-term Memory (LSTM) model. This recurrent neural network model will forecast electricity bill and notify users if abnormal energy consumption of individual home appliances is detected. Thirdly, a dashboard using Matplotlib library where users may monitor the real-time energy consumption. The completed LSTM model demonstrates a high accuracy of more than 80%, where a low mean squared error with train score = 0.1798 and test score = 0.1229 is determined. Furthermore, the R2 test shows a high level of goodness of fit with train score = 0.844 and test score = 0.835.

Electricity peak shaving for commercial buildings using machine learning and vehicle to building (V2B) system

Since the occurrence of oil shocks in the 1970s, a number of countries have introduced fuel poverty programs. However, rebound effects could be problematic even in these programs. In particular, there are two controversies surrounding rebound effects: the magnitude of rebound effects and the influence of income on these effects. This study attempts to resolve these issues by empirically estimating the rebound effects of individual home appliances for low-income households. Thereafter, it compares the rebound effects for low-income families with those for all-income families. Analyses results suggest that the magnitude of rebound effects highly depends on individual home appliances, and that these effects are usually larger for low-income households. Thus, the differences in rebound effects between all-income and low-income households also depend on individual appliances. Therefore, policy-makers should meticulously consider the rebound effects of individual home appliances when planning energy efficiency programs.

Energy management in residential buildings has grabbed the attention of many scientists for the last few years due to the fact that the residential sector consumes the highest amount of total energy produced by different energy resources. To manage the energy in residential buildings effectively, an efficient energy control system is required, capable of decreasing the total energy consumption without compromising the user-preferred environment inside the building. In the literature, many approaches have been proposed to achieve the goals of minimizing the energy consumption and maximizing the user preferred comfort by keeping different parameters under consideration, but all these methods face some problems in resolving the issue properly. The bat algorithm is one of the most recently introduced optimization approaches that has drawn the attention of researchers to apply it for solving different types of optimization problems. In this paper, the bat algorithm is applied for energy optimization in residential buildings, which is one of the most focused optimization problems in recent years. Three environmental parameters, namely temperature, illumination and air quality are bat algorithm inputs and optimized values of these parameters are the outputs. The error difference between the environmental parameters and optimized parameters are inputs of the fuzzy controllers which give energy as output which in turn change the status of the concerned actuators. It is proven from the experimental results that the proposed approach has been effectively successful in managing the whole energy consumption management system.

Electricity is the predominant energy source used in these buildings. Therefore, energy management in residential buildings requires serious attention to ensure the energy consumption in residential building have been managed properly. Currently, the unstable of fuel price give the big impact to electricity price. Due to the issue, consumers require to use electricity more wisely. Using energy efficiently is one of the solution to reduce energy consumption. This paper aims to propose an initiative strategy for energy management system based on an analysis of energy consumption in residential building. The level of energy consumption among the occupants was found by obtaining electricity bills and distributing a questionnaire. A case study was carried out in selected areas in the Southern Zone of Peninsular Malaysia. The results of the study show that energy consumption was significant increased by month and the EEES as one of energy management system was suggested.

This paper presents a battery-aware stochastic control framework for residential energy management systems (EMS) equipped with energy harvesting, that is, photovoltaic panels, and storage capabilities. The model and control rationale takes into account the dynamics of load, the weather, the weather forecast, the utility, and consumer preferences into a unified Markov decision process. The embedded optimization problem is formulated to determine the proportion of energy drawn from the battery and the grid to minimize a cost function capturing a user-defined tradeoff between battery degradation and financial expense by user preferences. Numerical results are based on real-world weather data for Golden, Colorado, and load traces. The results illustrate the ability of the system to limit battery degradation assessed using the Rain flow counting method for lithium ion batteries.

The energy management in residential buildings according to occupant’s requirement and comfort is of vital importance. There are many proposals in the literature addressing the issue of user’s comfort and energy consumption (management) with keeping different parameters in consideration. In this paper, we have utilized artificial bee colony (ABC) optimization algorithm for maximizing user comfort and minimizing energy consumption simultaneously. We propose a complete user friendly and energy efficient model with different components. The user set parameters and the environmental parameters are inputs of the ABC, and the optimized parameters are the output of the ABC. The error differences between the environmental parameters and the ABC optimized parameters are inputs of fuzzy controllers, which give the required energy as the outputs. The purpose of the optimization algorithm is to maximize the comfort index and minimize the error difference between the user set parameters and the environmental parameters, which ultimately decreases the power consumption. The experimental results show that the proposed model is efficient in achieving high comfort index along with minimized energy consumption.

A new technology called smart energy management makes use of IoT concepts to enhance energy efficiency and lower waste in structures. The goal of this study is to comprehend how household energy management knowledge affects energy usage, user behavior, related expenses, and environmental effect. Through a survey of 100 valid replies in Palestine, the research model assessed the knowledge and consumption habits of building occupants. Smart PLS software was used to analyze the research model using partial least squares structural equation modeling (PLS-SEM). Using path coefficients and behavior as a mediating variable, the structural model connected the latent variables. The mediation hypotheses were tested using the Preacher and Hayes method, and the indirect effect and confidence intervals were estimated and calculated using bootstrapping. The findings demonstrated that by lowering energy use and enhancing overall building performance, residential buildings that implement smart energy consumption management systems may move toward a more sustainable future. Furthermore, the study found that education and awareness campaigns are necessary to increase residents’ knowledge of these systems to promote energy savings. The results also indicated statistically significant indirect effects, supporting the existence of mediation of the behavior construct. Path coefficient values and P-values were presented to further support the study’s hypotheses. Such smart energy management systems represent an important innovation in building management and can help create more sustainable and efficient buildings.

Improved energy management technique in pipe-embedded wall heating/cooling system in residential buildings

The study predicted electrical energy management practices among residents, using regression model. Three variables used were (Income levels, Age groups and Educational levels) of electrical energy users. The study adopted a cross sectional survey research design. The population of the study was made up of 191,416 heads of households in residential buildings connected to the distribution network in 25 Local Governments of Niger State. The sample for the study consisted of 1,290 heads of households in residential buildings drawn through multistage sampling techniques. The instrument used for data collection was a structured questionnaire. Correlation and multiple regression analysis were used to analyzed hypothesis which determined the no relationship at (P ˂ .05) level of significance on the practices of residents on electrical energy management in residential buildings in Niger State and to show the variable that has effect and the degree of the effect on the electrical energy management practices. The model developed is this shows that person income level, age group and educational level were significant in predicting ones practices on electrical energy management with significance values of 0.02, 0.00 and 0.00 less than the of P ˂ .05. Furthermore, age groups have a negative impact on electrical energy management practices in contrast to the income levels and educational qualifications. This is indicated by the values of their coefficient of +0.07, -0.062 and 0.097 respectively. The model significantly predicts electrical energy management practices. The conclusion is strengthen by the significant value of 0.00 less than the of P ˂ 0.05. KEY WORDS : Model, Predicting, Electrical Energy, Management, Practices and Residents

The increasing energy demand and rising electricity costs have intensified the need for efficient energy management in residential buildings. Load imbalances in power grids and cost surges during peak demand periods are pushing consumers toward more strategic and conscious consumption models. This study focuses on optimizing household electricity consumption through load shifting strategies. To achieve this, a low-cost, Internet of Things (IoT)-based Smart Energy Management System (SEMS) was developed and implemented in a real household. Its performance was then compared with a commercial reference device, revealing an impressive accuracy rate of 99.98%. Electricity consumption data recorded over a one-year period using SEMS was analyzed to assess voltage fluctuations, frequency variations, and energy usage patterns. Daily, weekly, and seasonal consumption trends were identified. Moreover, classifying shiftable and non-shiftable loads enabled a comprehensive evaluation of the load-shifting potential. Scenario-based analyses demonstrated that shifting 25%, 50%, and 75% of shiftable loads to lower-cost tariff periods could result in 9.8%, 17.6%, and 26.1% cost savings, respectively. These findings indicate that SEMS enhances energy efficiency and reduces electricity costs. With its user-friendly interface and low installation cost, SEMS is well-suited for widespread adoption and presents an effective solution for time-based load shifting. Additionally, by encouraging users to adopt more efficient consumption habits, SEMS is expected to contribute to sustainable energy management.

As an indispensable component of coal-fired power plants, boilers play a crucial role in converting water into high-pressure steam. The oxygen content in the flue gas is a crucial indicator, which indicates the state of combustion within the boiler. The oxygen content not only affects the thermal efficiency of the boiler and the energy utilization of the generator unit, but also has adverse impacts on the environment. Therefore, accurate measurement of the flue gas’s oxygen content is of paramount importance in enhancing the energy utilization efficiency of coal-fired power plants and reducing the emissions of waste gas and pollutants. This study proposes a prediction model for the oxygen content in the flue gas that combines the whale optimization algorithm (WOA) and long short-term memory (LSTM) networks. Among them, the whale optimization algorithm (WOA) was used to optimize the learning rate, the number of hidden layers, and the regularization coefficients of the long short-term memory (LSTM). The data used in this study were obtained from a 350 MW power generation unit in a coal-fired power plant to validate the practicality and effectiveness of the proposed hybrid model. The simulation results demonstrated that the whale optimization algorithm–long short-term memory (WOA-LSTM) model achieved an MAE of 0.16493, an RMSE of 0.12712, an MAPE of 2.2254%, and an value of 0.98664. The whale optimization algorithm–long short-term memory (WOA-LSTM) model demonstrated enhancements in accuracy compared with the least squares support vector machine (LSSVM), long short-term memory (LSTM), particle swarm optimization–least squares support vector machine (PSO-LSSVM), and particle swarm optimization–long short-term memory (PSO-LSTM), with improvements of 4.93%, 4.03%, 1.35%, and 0.49%, respectively. These results indicated that the proposed soft sensor model exhibited more accurate performance, which can meet practical requirements of coal-fired power plants.

Long Short-Term Memory (LSTM) neural networks must consider parameter settings for more accurate prediction results. This study aims to improve the LSTM performance using Particle Swarm Optimization (PSO) for hyperparameters selection. PSO may optimize the weight of the LSTM neural network, which results in minimal error prediction due to its feature selection and data normalization abilities. This study applies the LSTM architecture, consisting of 3 layers (input, hidden, and output) with several 7, 3, and 1. Root mean squared error (RMSE) and processing time evaluate the results. The results show that PSO tuning obtained six hyperparameters values, namely RMSprop, ReLU, MAE, batch size (128), Neuron (78), and Epoch (62). The RMSE of the LSTM with PSO tuning has a minor error value of 21.923 and the fastest processing time of 156 seconds. The results overcome other LSTM models with random parameters. The highest difference of RMSE between the proposed approach and the baseline model is 7.621. In other words, the PSO hyperparameters tuning efficiently reduces the LSTM errors. These findings could be useful as a recommendation for reducing the time-consuming efforts of hyperparameters selection.

Building-owned micro energy hubs (EHs) usually focus on optimal energy consumption cost and emission, whereas, macro energy hubs (MEHs) mainly concentrate on utility’s interest termed as network load deviation. Therefore, a bi-level MEH control capable of simultaneously attaining bilateral stakes is required. However, in real life energy distribution systems, MEH structure operates under uncertainties such as unpredictable solar photovoltaic (PV) irradiance and unplanned electric and natural gas (NG) network outages. Such uncertain conditions may affect the MEH’s performance (energy cost, emission and network load deviation) and resilience (capability of recovering quickly from disturbances) undesirably. Objective of this paper is to attain an optimal compromise between the performance and the resilience of a bi-level residential MEH, under uncertain conditions. In first step, risk neutral bi-level MEH is proposed to optimally reduce the network load deviation, while obeying the customer specified comfort, emission and cost constraints. However, this strategy disregards the risk introduced by the uncertainties. In second step, risk averse strategy is devised by incorporating conditional value at risk (CVaR) in the objective function, for improvement in resilience. Proposed linear bi-level risk averse MEH is mapped in conventional flower pollination algorithm (FPA). Resilience of the MEH is measured in terms of energy stored in the plug-in hybrid electric vehicle (PHEV) and thermal energy storage (TES). Third step concentrates on the development of an efficient solution technique for energy management system to obtain better solution. For this, an improved version of FPA, termed as 2-cored FPA is proposed and employed to solve the model. Proposed technique has a filtration layer (termed as core-1) that consists of random walk, local pollination and global pollination to obtain improved pollinators. Subsequently, these pollinators are injected into conventional FPA layer (termed as core-2) to obtain better solution. Comparison of results demonstrates that under risk averse approach with two cores, the energy retaining capability of the PHEV and the TES increases by 31.66% and 57.66%, respectively.