Hybrid agent-based simulation for policy evaluation of solar power generation systems

To encourage the adoption of solar power as well as new technological improvements in solar industry, state and federal governments have employed various kinds of incentives over the past decades, such as rebates, tax return opportunities, and Net Metering credits. At the same time, however, the governments concern regulations to avoid highly steep growth of solar energy without considering necessary supporting structure such as storage components, which will increase the electricity price and threaten the stability of existing transmission systems. The goal of this research is to develop a decision support tool to analyze the effectiveness of various policies (both incentives as well as regulations) on the proper growth rate of distributed photovoltaic (PV) systems avoiding the instability of the transition system or steep rising of the electricity price. To this end, we propose a hybrid two-level simulation modeling framework, which is significantly more detailed than the simplified structures commonly used in most policy evaluations. The lower-level model concerns the calculation of PV system payback period of individual household based on hourly electricity generation (PV) and consumptions, incentive levels, PV module price, and hourly electricity price (grid). The higher-level model, running on a weekly basis for 20 years, concerns the household adoption behaviors of the PV systems influenced by various factors, including payback period, household income, word-of-mouth effect and advertisement effect. Agent-based and system dynamics modeling techniques are leveraged in both levels. The proposed models have been developed for residential areas at two different regions in the US based on real data, which have been used to illustrate the impact of policies in different regions.     

Hybrid agent-based simulation for policy evaluation of solar power generation systems

To encourage the adoption of solar power as well as new technological improvements in solar industry, state and federal governments have employed various kinds of incentives over the past decades, such as rebates, tax return opportunities, and Net Metering credits. At the same time, however, the governments concern regulations to avoid highly steep growth of solar energy without considering necessary supporting structure such as storage components, which will increase the electricity price and threaten the stability of existing transmission systems. The goal of this research is to develop a decision support tool to analyze the effectiveness of various policies (both incentives as well as regulations) on the proper growth rate of distributed photovoltaic (PV) systems avoiding the instability of the transition system or steep rising of the electricity price. To this end, we propose a hybrid two-level simulation modeling framework, which is significantly more detailed than the simplified structures commonly used in most policy evaluations. The lower-level model concerns the calculation of PV system payback period of individual household based on hourly electricity generation (PV) and consumptions, incentive levels, PV module price, and hourly electricity price (grid). The higher-level model, running on a weekly basis for 20 years, concerns the household adoption behaviors of the PV systems influenced by various factors, including payback period, household income, word-of-mouth effect and advertisement effect. Agent-based and system dynamics modeling techniques are leveraged in both levels. The proposed models have been developed for residential areas at two different regions in the US based on real data, which have been used to illustrate the impact of policies in different regions.     

Economic and technical assessment of rooftop solar photovoltaic potential in Brownsville,Texas, U.S.A

Localized assessment of solar energy economic feasibility will benefit the structuring of residential solar energy deployment globally. In the U.S. growing interest in rooftop residential solar among city managers has spurred the development of photovoltaic (PV) feasibility maps of the technical and economic solar potential within cities. The City of Brownsville, Texas was interested in evaluating solar feasibility for their city but lacked information to make informed policy decisions on PV development. This paper presents novel and systems approaches for determining the technical and economic feasibility of solar development for homes in the Brownsville using LiDAR and local information. Residential technical and economic potential was assessed by optimizing the internal rate of return (IRR) and an average residential building demand profile to determine ideal size and placement of solar arrays. Results showed that residential structures in Brownsville have the technical potential to generate approximately 11% of the total energy provided by the local utility; however, average IRR was only 2.9% with a payback period of over 15 years. Five neighborhoods in the City of Brownsville were identified with spatially clustered homes that had relatively higher IRRs compared with other areas in the city. Despite the high technical potential, modeled results indicate that perspective home owners interested in solar development may require additional incentives to improve the economic feasibility of PV in Brownsville. This study provides a demonstration of an interdisciplinary systems approach and methodology that can be adopted internationally to evaluate the feasibility of solar development in other areas.     

基于一种改进的BP神经网络光伏电池建模

由于光伏电池具有高度非线性特性,难以建模,而传统的数学模型难以满足光伏控制系统设计和应用的要求。该文利用神经网络具有逼近任意复杂非线性函数的能力,将神经网络技术应用到光伏阵的建模中,避开了该模块内部的复杂性。模型以太阳能日照、温度以及负载电压作为神经网络辨识模型的输入量,光伏阵输出电流为输出量,采用改进型BP算法,建立了光伏电池的动态响应模型,然后预测了最大功率点。文中给出模型的结构,训练步骤和仿真结果。仿真结果表明,方法可行,建立的模型精度较高,从而为设计光伏实时控制系统奠定了基础。     

Agent-based and analytical modeling to evaluate the effectiveness of greenbelts

We present several models of residential development at the rural–urban fringe to evaluate the effectiveness of a greenbelt located beside a developed area, for delaying development outside the greenbelt. First, we develop a mathematical model, under two assumptions about the distributions of service centers, that represents the trade-off between greenbelt placement and width, their effects on the rate of development beyond the greenbelt, and how these interact with spatial patterns of aesthetic quality and the locations of services. Next, we present three agent-based models (ABMs) that include agents with the potential for heterogeneous preferences and a landscape with the potential for heterogeneous attributes. Results from experiments run with a one-dimensional ABM agree with the starkest of the results from the mathematical model, strengthening the support for both models. Further, we present two different two-dimensional ABMs and conduct a series of experiments to supplement our mathematical analysis. These include examining the effects of heterogeneous agent preferences, multiple landscape patterns, incomplete or imperfect information available to agents, and a positive aesthetic quality impact of the greenbelt on neighboring locations. These results suggest how width and location of the greenbelt could help determine the effectiveness of greenbelts for slowing sprawl, but that these relationships are sensitive to the patterns of landscape aesthetic quality and assumptions about service center locations.     

Agent-based modelling of land use dynamics and residential quality of life for future scenarios

Current LUCC research employs scenario-based analysis to explore possible future trends and impacts by defining a coherent set of plausible future socio-economic development pathways. Typically, computational models are therein used to interpret qualitative future storylines in terms of quantitative future changes. This paper addresses these challenges and illustrates some of the advantages of a scenario-based approach using an Agent-Based Model (ABM). Storylines are shown to be useful in integrate a broad variety of knowledge sources, such as subjective expert judgement and results from other (integrative) models, which rely on a similar set of assumptions about the future. The advantages of ABMs are demonstrated for interpreting future scenarios in the context of spatial and temporal variations in socio-ecological outcomes based on heterogeneous individual behaviour. For example, ABMs are shown to enable potential hotspots of future development and LUCC to be identified. Furthermore, a procedure is presented for downscaling and interpreting storylines from general qualitative trends to local quantitative parameters within an ABM framework. This framework is applied to the Municipality of Koper, Slovenia, where the future impacts of LUCC on the loss of agricultural land and residential quality-of-life are simulated. The results are compared to a “business-as-usual” baseline and it is shown that industrial and commercial development has the greatest impact on the loss of high quality agricultural land across all scenarios. Furthermore, the model indicates an increase in inequality in the perceived quality-of-life of residential households, with new households achieving higher quality-of-life than existing residents.     

Modular ABM development for improved dissemination and training

Agent-based models (ABMs) have become an important tool for advancing scientific understanding in a variety of disciplines and more specifically have contributed gains to natural resource management in recent decades. However, a key challenge to their utility is the lack of convergence upon a common set of assumptions for representing key processes (such as agent decision structure), with the outcome that published ABM tools are rarely (if ever) used beyond their original development team. While a number of ABM frameworks are publicly available for use, the continued development of models from scratch is a signal of the continuing difficulty in capturing sufficient modeling flexibility in a single package. In this study we examine ABM sharing by comparing co-citation networks from several well-known ABM frameworks to those used in the land-use change modeling community. We then outline a different publication paradigm for the ABM community that could improve the sharing of model structure and help move toward convergence on a common set of tools and assumptions.     

Evolutionary algorithms for the design of grid-connected PV-systems

The sale of electric energy generated by photovoltaic (PV) plants has attracted much attention in recent years. The installation of PV plants aims to obtain the maximum benefit of captured solar energy. The current methodologies for planning the design of the different components of a PV plant are not completely efficient. This paper addresses the optimization of the design of PV plants with solar tracking, which consists of the optimization of the variables that make up the PV plant to obtain the minimum electric (Joule) losses possible. These variables are the size and distribution of solar modules in the solar tracker, the distribution of the solar trackers in the field and the choice of inverter. Evolutionary algorithms (EAs) are adaptive methods based on natural evolution that may be used for searching and optimization. Four different EAs have been used for optimizing the design of PV plants: steady-state genetic algorithm, generational genetic algorithm, CHC algorithm and DE algorithm. In order to test the performance of these algorithms we have used different proposed fields to mount PV plants. The results obtained show that EAs, and specifically DE with rand mutation schemes, are promising techniques to optimize design of PV plants. Furthermore, the results are contrasted with nonparametric statistical tests to support our conclusions.     

Modelling the impacts of land system dynamics on human well-being: Using an agent-based approach to cope with data limitations in Koper,Slovenia

To cope with data limitations and to provide insight into the dynamics of LUCC for local stakeholders in the Municipality of Koper, Slovenia, we constructed an ABM (loosely defined) that integrates utility theory, logistic regression, and cellular automaton-like rules to represent the decision-making strategies of different agents. The model is used to evaluate the impact of LUCC on human well-being, as represented by the provision of highly productive agricultural soil, the extent of noise pollution, and quality-of-life measurements. Results for the Municipality of Koper show that, under a range of model assumptions, (1) high quality agricultural soils are disproportionately affected by urban growth, (2) aggregate resident quality of life increases non-linearly with a change in development density, (3) some drivers of residential settlement produce non-linear preference responses, and (4) clustering industrial development had a beneficial impact on human well-being. Additional novel contributions include the incorporation of noise pollution feedbacks and an approach to empirically inform agent preferences using a conjoint analysis of social survey data.     

Solar Radiation Prediction Using Satin Bowerbird Optimization with Modified Deep Learning

Solar energy will be a great alternative to fossil fuels since it is clean and renewable. The photovoltaic (PV) mechanism produces sunbeams’ green energy without noise or pollution. The PV mechanism seems simple, seldom malfunctioning, and easy to install. PV energy productivity significantly contributes to smart grids through many small PV mechanisms. Precise solar radiation (SR) prediction could substantially reduce the impact and cost relating to the advancement of solar energy. In recent times, several SR predictive mechanism was formulated, namely artificial neural network (ANN), autoregressive moving average, and support vector machine (SVM). Therefore, this article develops an optimal Modified Bidirectional Gated Recurrent Unit Driven Solar Radiation Prediction (OMBGRU-SRP) for energy management. The presented OMBGRU-SRP technique mainly aims to accomplish an accurate and time SR prediction process. To accomplish this, the presented OMBGRU-SRP technique performs data preprocessing to normalize the solar data. Next, the MBGRU model is derived using BGRU with an attention mechanism and skip connections. At last, the hyperparameter tuning of the MBGRU model is carried out using the satin bowerbird optimization (SBO) algorithm to attain maximum prediction with minimum error values. The SBO algorithm is an intelligent optimization algorithm that simulates the breeding behavior of an adult male Satin Bowerbird in the wild. Many experiments were conducted to demonstrate the enhanced SR prediction performance. The experimental values highlighted the supremacy of the OMBGRU-SRP algorithm over other existing models.     

ANN based simulation and experimental verification of analytical four- and five-parameters models of PV modules

In this article, artificial neural network (ANN) is adopted to predict photovoltaic (PV) panel behaviors under realistic weather conditions. ANN results are compared with analytical four and five parameter models of PV module. The inputs of the models are the daily total irradiation, air temperature and module voltage, while the outputs are the current and power generated by the panel. Analytical models of PV modules, based on the manufacturer datasheet values, are simulated through Matlab/Simulink environment. Multilayer perceptron is used to predict the operating current and power of the PV module. The best network configuration to predict panel current had a 3–7–4–1 topology. So, this two hidden layer topology was selected as the best model for predicting panel current with similar conditions. Results obtained from the PV module simulation and the optimal ANN model has been validated experimentally. Results showed that ANN model provide a better prediction of the current and power of the PV module than the analytical models. The coefficient of determination (R²), mean square error (MSE) and the mean absolute percentage error (MAPE) values for the optimal ANN model were 0.971, 0.002 and 0.107, respectively. A comparative study among ANN and analytical models was also carried out. Among the analytical models, the five-parameter model, with MAPE = 0.112, MSE = 0.0026 and R² = 0.919, gave better prediction than the four-parameter model (with MAPE = 0.152, MSE = 0.0052 and R² = 0.905). Overall, the 3–7–4–1 ANN model outperformed four-parameter model, and was marginally better than the five-parameter model.     

Data-driven agent-based modeling,with application to rooftop solar adoption

Agent-based modeling is commonly used for studying complex system properties emergent from interactions among agents. However, agent-based models are often not developed explicitly for prediction, and are generally not validated as such. We therefore present a novel data-driven agent-based modeling framework, in which individual behavior model is learned by machine learning techniques, deployed in multi-agent systems and validated using a holdout sequence of collective adoption decisions. We apply the framework to forecasting individual and aggregate residential rooftop solar adoption in San Diego county and demonstrate that the resulting agent-based model successfully forecasts solar adoption trends and provides a meaningful quantification of uncertainty about its predictions. Meanwhile, we construct a second agent-based model, with its parameters calibrated based on mean square error of its fitted aggregate adoption to the ground truth. Our result suggests that our data-driven agent-based approach based on maximum likelihood estimation substantially outperforms the calibrated agent-based model. Seeing advantage over the state-of-the-art modeling methodology, we utilize our agent-based model to aid search for potentially better incentive structures aimed at spurring more solar adoption. Although the impact of solar subsidies is rather limited in our case, our study still reveals that a simple heuristic search algorithm can lead to more effective incentive plans than the current solar subsidies in San Diego County and a previously explored structure. Finally, we examine an exclusive class of policies that gives away free systems to low-income households, which are shown significantly more efficacious than any incentive-based policies we have analyzed to date.     

Modeling and simulation of a grid connected PV system based on the evaluation of main PV module parameters

In this work we present a new method for the modeling and simulation study of a photovoltaic grid connected system and its experimental validation. This method has been applied in the simulation of a grid connected PV system with a rated power of 3.2 Kw_p, composed by a photovoltaic generator and a single phase grid connected inverter. First, a PV module, forming part of the whole PV array is modeled by a single diode lumped circuit and main parameters of the PV module are evaluated. Results obtained for the PV module characteristics have been validated experimentally by carrying out outdoor I-V characteristic measurements. To take into account the power conversion efficiency, the measured AC output power against DC input power is fitted to a second order efficiency model to derive its specific parameters.The simulation results have been performed through Matlab/Simulink environment. Results has shown good agreement with experimental data, whether for the I-V characteristics or for the whole operating system. The significant error indicators are reported in order to show the effectiveness of the simulation model to predict energy generation for such PV system.     

Hybrid simulation and optimization-based design and operation of integrated photovoltaic generation,storage units,and grid

Unlike fossil-fueled generation, solar energy resources are geographically distributed and highly intermittent, which makes their direct control extremely difficult and requires storage units as an additional concern. The goal of this research is to design and develop a flexible tool, which will allow us to obtain (1) an optimal capacity of an integrated photovoltaic (PV) system and storage units and (2) an optimal operational decision policy considering the current and future market prices of the electricity. The proposed tool is based on hybrid (system dynamics model and agent-based model) simulation and meta-heuristic optimization. In particular, this tool has been developed for three different scenarios (involving different geographical scales), where PV-based solar generators, storage units (compressed-air-energy-storage (CAES) and super-capacitors), and grid are used in an integrated manner to supply energy demands. Required data has been gathered from various sources, including NASA and TEP (utility company), US Energy Information Administration, National Renewable Energy Laboratory, commercial PV panel manufacturers, and publicly available reports. The constructed tool has been demonstrated to (1) test impacts of several factors (e.g. demand growth, efficiencies in PV panel and CAES system) on the total cost of the integrated generation and storage system and an optimal mixture of PV generation and storage capacity, and to (2) demonstrate an optimal operational policy.     

Genetic algorithm-trained radial basis function neural networks for modelling photovoltaic panels

Radial basis function neural networks (RBFNs) can be applied to model the I–V characteristics and maximum power points (MPPs) of photovoltaic (PV) panels. The key issue for training an RBFN lies in determining the number of radial basis functions (RBFs) in the hidden layer. This paper presents a genetic algorithms-based RBFN training scheme to search for the optimal number of RBFs using only the input samples of a PV panel. The performance of the trained RBFN is comparable with that of the conventional model and the training algorithm is computationally efficient. The trained RBFNs have been applied to predict MPPs of two different practical PV panels. The results obtained are accurate enough for applying the models to control the PV systems for tracking the optimal power points.     

Qualitative Models in Computational Simulative Sciences: Representation,Confirmation, Experimentation

The Epistemology Of Computer Simulation (EOCS) has developed as an epistemological and methodological analysis of simulative sciences using quantitative computational models to represent and predict empirical phenomena of interest. In this paper, Executable Cell Biology (ECB) and Agent-Based Modelling (ABM) are examined to show how one may take advantage of qualitative computational models to evaluate reachability properties of reactive systems. In contrast to the thesis, advanced by EOCS, that computational models are not adequate representations of the simulated empirical systems, it is shown how the representational adequacy of qualitative models is essential to evaluate reachability properties. Justification theory, if not playing an essential role in EOCS, is exhibited to be involved in the process of advancing and corroborating model-based hypotheses about empirical systems in ECB and ABM. Finally, the practice of evaluating model-based hypothesis by testing the simulated systems is shown to constitute an argument in favour of the thesis that computer simulations in ECB and ABM can be put on a par with scientific experiments.

     

Centrally located yet close to nature: A prescriptive agent-based model for urban design

A common dilemma for planners is how to design urban settlements that give people easy access to a center and nature. Difficulties arise because each household's access to such elements is a function of other households' location and the set of potential arrangements is constrained by the households' degree of acceptance of different density levels. This paper suggests the ideal arrangement of built-up and green areas may be identified by simulating in an agent-based model (ABM) the interactions of virtual households that try to find the best residential location based on their preferences towards distance from the center, proximity to green space and density.Simulations showed that the ABM can, iteration after iteration, develop progressively better configurations and eventually get to an equilibrium if households' locational choice is driven not only by the maximization of individual utility, but also the preservation of the neighbors' well-being. Model's outputs suggest that compact settlements with an even distribution of green spaces offer the greatest benefits to their inhabitants, and that larger green areas are to be preferred when the population is less sensitive to density and/or the travel to the center is faster along some directions. Application of a rent formation model on the configurations generated by the ABM shows that these are relatively equitable, as lower income households could afford at least half of all locations. Future improvements can turn this model into a suitable tool for designing new settlements, guiding the densification of existing settlements or defining zoning regulations.     

Model testing and assessment: Perspectives from a swarm intelligence,agent-based model of forest insect infestations

Model testing procedures represent a major challenge in the development of agent-based models (ABMs). However, they are required stages for a model to be accepted and to serve as a forecasting, management or decision-making tool. This study presents a comprehensive approach for testing ForestSimMPB, an agent-based model (ABM) designed to simulate mountain pine beetle (MPB), Dendroctonus ponderosae Hopkins, outbreaks at the scale of individual trees. ForestSimMPB is a complex system model that is using swarming intelligence, capable to represent individuals’ behaviours and spatial interactions that influence their surrounding environment. Swarm Intelligence (SI) methods are integrated into the ABM in order to reproduce the collective reasoning and indirect communication of autonomous agents representing MPB behaviour within the forest environment. Model testing approach consist of verification, calibration, sensitivity analysis, validation and qualification stages. Model testing is accomplished by simulating MPB infestations using both the ForestSimMPB model and a Random–ABM model that serves as a null model. Outcomes comparison and assessment are performed using raster-based techniques as well as spatial metrics. Aerial photographs of the British Columbia, Canada study sites are used in this model testing approach. Results indicate that ForestSimMPB model representations of MPB outbreaks are more similar than Random model representations to the spatial distribution of MPB-dead trees.     

积灰条件下光伏阵列最优倾角问题研究

在光伏发电系统设计中,安装倾角的选择对光伏发电效率具有重要影响.针对太阳能光伏阵列常见的表层积灰现象,改变传统的只考虑最大辐射量的倾角确定方法,提出了综合考虑表层积灰情况下的最优发电倾角计算方法,使倾角的确定更加合理与完善.建立了积灰辐射量统一发电模型,并基于Matlab对模型进行了仿真验证.以福建某光伏电站为例,搭建实验平台,通过相关数据的检验与预测,得到了光伏电池板的综合最优倾角.实验结果证明该模型所确定的倾角比传统模型可以得到更大的发电量,提高了光伏利用效率.