Pradeep K. Behera

Solar radiation prediction based on recurrent neural networks trained by Levenberg-Marquardt backpropagation learning algorithm

In response to the growing concern over the use of fossil fuels, renewable energy industries have been significant economic drivers in many parts of the United States. In the recent years there is a strong growth in solar power generation industries that requires prediction of solar energy to develop highly efficient stand-alone photovoltaic systems as well as hybrid power systems. In order to accomplish the goal, we propose a predictive model that is based on recurrent neural networks trained with the Levenberg-Marquardt backpropagation learning algorithm to forecast the solar radiation using the past solar radiation and solar energy. This computational intelligence modeling tool explored the impact of solar radiation and solar energy in forecasting reliable long-run solar energy. Based on the excellent experimental results including the mean squared error analysis, error autocorrelation function analysis, regression analysis, and time series response, it demonstrated that the proposed neural network structure and the learning algorithm could be very useful in training the recurrent neural network for the solar radiation prediction.

Characterization of Urban Runoff Quality: A Toronto Case Study

This chapter presents an overview of the characterization of urban runoff quality constituents. Characterization includes descriptive statistics, correlation a…

Optimization of a Stormwater Quality Management Pond System

Stormwater system management in urban watersheds is often achieved by effective design and use of detention ponds which help mitigate impacts of urban drainage from water quantity and quality perspectives. The costs associated with these tasks for management can be minimized considering the fulfillment of objectives of environmental and regulatory compliance. Nonlinear and mixed integer nonlinear programming (MINLP) formulations with discrete and binary variables is developed in this study to obtain an optimal design for a multiple stormwater detention pond system. The main objective considered is the minimization of cost constrained on system performance related to pollution control. Analytical probabilistic expressions in mathematically closed form for system performance depicting watershed hydrology, control system hydraulics and pollution removal processes are integrated into the optimization formulations. Gradient-based NLP and genetic algorithm-based solvers are used to obtain optimal solutions. Application of the methodology is demonstrated with a hypothetical case study system with realistic hydrologic and water quality parameter values and the benefits of solutions for effective pollution control are reported. Results from the solutions of the formulations provide optimal design parameters considering the runoff control and pollutant reduction considering environmental and regulatory constraints. A comparison of results from these formulations to those from a dynamic programming (DP) formulation developed in an earlier study indicates that limitations associated with discretization within DP can be overcome with the proposed optimal formulations.

Solar radiation prediction based on particle swarm optimization and evolutionary algorithm using recurrent neural networks

Over the last decade, there has been emphasis on the reduction of the dependency of fossil fuels that resulting in the growth of renewable energy industries. These industries have been significant economic drivers in many parts of the United States supported by both government and private sectors. As a part of renewable energy industries, there is a strong growth in solar power generation industries that often requires prediction of solar energy to develop highly efficient stand-alone photovoltaic systems as well as hybrid power systems. Specifically solar radiation prediction is a important component in the solar energy production. However, some computational intelligence methods that have most successful applications on time series prediction have not yet been investigated on solar radiation prediction. Only a limited number of neural networks models were applied to the solar radiation monitoring. Therefore, we propose an Elman style based recurrent neural network to predict solar radiation from the past solar radiation and solar energy in this research. A hybrid learning algorithm incorporating particle swarm optimization and evolutional algorithm was presented, which takes the complementary advantages of the two global optimization algorithms. The neural networks model was trained by particle swarm optimization and evolutional algorithm to forecast the solar radiation. The excellent experimental results demonstrated that the proposed hybrid learning algorithm can be successfully used for the recurrent neural networks based prediction model for the solar radiation monitoring.

Precipitation Extremes and Climate Change: Evaluation Using Descriptive WMO Indices

Assessment of spatial and temporal extreme precipitation events due to climate variability and change is critical for future hydrologic design. Evaluation of these extremes in the past has been limited to evaluation of annual and partial duration series. However, climate-change sensitive hydrologic design requires evaluation of precipitation extremes at different temporal levels using a variety of indices. This study evaluates the variability of precipitation extremes in two climatic regions in the U. S. using WMO (World Meteorological Organization) proposed and adopted eleven indices. These indices relate to precipitation extremes at a daily temporal scale and encompass a variety of conditions including user-defined precipitation thresholds. Quantitative evaluation, statistical analyses and spatial variability of indices in a region as well across different climate zones indicate that infilling of precipitation data and existence of in homogeneities influences the assessment of trends in extreme events using indices. This paper presents preliminary results of an ongoing study.

Design of adaptive feature extraction algorithm based on fuzzy classifier in hyperspectral imagery classification for big data analysis

We proposed a new adaptive feature extraction (FEA) approach that integrates concepts of per-pixel/field classification and spectral ummixing. It combines their advantages in adaptive feature selection while minimizing the disadvantages associated with the high-complexity of each technique. The approach consists of local gradients calculation, reference clusters determination, prototype classification using fuzzy classifier, and feature vectors selection. Multiple experiments were performed using a simulated hyperspectral cube composed by 123 samples and 1254 features and classification was done only for verification purposes. Cross-validation demonstrated that FEA generated an average improvement of 7% on the misclassification error when compared to full feature analysis.

Advanced sensor-computer technology for urban runoff monitoring

The paper presents the project teams advanced sensor-computer sphere technology for real-time and continuous monitoring of wastewater runoff at the sewer discharge outfalls along the receiving water. This research significantly enhances and extends the previously proposed novel sensor-computer technology. This advanced technology offers new computation models for an innovative use of the sensor-computer sphere comprising accelerometer, programmable in-situ computer, solar power, and wireless communication for real-time and online monitoring of runoff quantity. This innovation can enable more effective planning and decision-making in civil infrastructure, natural environment protection, and water pollution related emergencies. The paper presents the following: (i) the sensor-computer sphere technology; (ii) a significant enhancement to the previously proposed discrete runoff quantity model of this technology; (iii) a new continuous runoff quantity model. Our comparative study on the two distinct models is presented. Based on this study, the paper further investigates the following: (1) energy-, memory-, and communication-efficient use of the technology for runoff monitoring; (2) possible sensor extensions for runoff quality monitoring.

Environmental urban runoff monitoring

Urban stormwater runoff has been a critical and chronic problem in the quantity and quality of receiving waters, resulting in a major environmental concern. To address this problem engineers and professionals have developed a number of solutions which include various monitoring and modeling techniques. The most fundamental issue in these solutions is accurate monitoring of the quantity and quality of the runoff from both combined and separated sewer systems. This study proposes a new water quantity monitoring system, based on recent developments in sensor technology. Rather than using a single independent sensor, we harness an intelligent sensor platform that integrates various sensors, a wireless communication module, data storage, a battery, and processing power such that more comprehensive, efficient, and scalable data acquisition becomes possible. Our experimental results show the feasibility and applicability of such a sensor platform in the laboratory test setting.

Storm Event Characteristics based on IETD for different Climatic Regions within United States

Given the climate variability and change throughout the world, local level continuous assessment of storm event characteristics is critical for analyzing adequacy of existing urban drainage infrastructures and for updating the critical rainfall information for future hydrologic designs. Given such new information, it is instructive to analyze this key input to the most of the stormwater management planning, analysis, design, and operation. Based on the Inter-event Time Definition (i.e., minimum dry period between storm events, or IETD), a long-term point rainfall record can be analyzed for different storm event characteristics such as event volume, event duration, event average intensity, and inter-event time. Such analysis will provide valuable information to our engineers, water resources professionals, and regulatory authorities about the change in storm event characteristics (i.e., storm event volume for different durations) over time at a specific location. This study evaluates storm event characteristics based on varying IETDS for different climatological regions within the United States. The available long-term hourly rainfall records at the representative stations within each of the climatic regions have been employed to conduct the analysis. The analysis will provide the storm event characteristics, particularly event volume and duration over time and their statistical analysis and spatial variability in a region and across different climatic zones. Such information is very useful for volume-based hydrology for managing urban stormwater management systems. The paper presents preliminary results of an ongoing study.

Optimization of Regional Stormwater Quality Control Systems Using Genetic Algorithms

Stormwater system management with a set of objectives in urban watersheds is often achieved by effective design and use of detention ponds. These ponds help mitigate impacts of urban drainage from water quantity and quality perspectives. The design, site selection for these ponds, operation and management over time are important aspects of effective stormwater management. This involves costs to land developers and there is always a need for minimizing these costs at that same time fulfilling the objectives of the management with environmental and regulatory compliance. A nonlinear programming formulation and genetic algorithm (GA) based optimization approaches are proposed to design a single stormwater detention pond system in the current study. The optimization methodology is also extended to consider multiple detention systems. The main objective is to minimize the cost constrained on system performance related to pollution control. The optimization techniques employs analytical probabilistic models for urban stormwater management analysis which mathematically closed form are enabling them easily integrated into optimization frame work. The results provide optimal design parameters considering the runoff control and pollutant reduction considering environmental and regulatory issues. Application of the methodology for a case study system is demonstrated with results and the benefits of using GA based optimization approach are illustrated.