Matching Items (26)
Description
The dissertation encompasses the transformer-less single phase PV inverters for both the string and microinverter applications. Two of the major challenge with such inverters include the presence of high-frequency common mode leakage current and double line frequency power decoupling with reliable capacitors without compromising converter power density. Two solutions are presented in this dissertation: half-bridge voltage swing (HBVS) and dynamic dc link (DDCL) inverters both of which completely eliminates the ground current through topological improvement. In addition, through active power decoupling technique, the capacitance requirement is reduced for both, thus achieving an all film-capacitor based solution with higher reliability. Also both the approaches are capable of supporting a wide range of power factor.
Moreover, wide band-gap devices (both SiC and GaN) are used for implementing their hardware prototypes. It enables the switching frequency to be high without compromising on the converter efficiency. Also it allows a reduced magnetic component size, further enabling a high power density solution, with power density far beyond the state-of-the art solutions.
Additionally, for the transformer-less microinverter application, another challenge is to achieve a very high gain DC-DC stage with a simultaneous high conversion efficiency. An extended duty ratio (EDR) boost converter which is a hybrid of switched capacitors and interleaved inductor technique, has been implemented for this purpose. It offers higher converter efficiency as most of the switches encounter lower voltage stress directly impacting switching loss; the input current being shared among all the interleaved converters (inherent sharing only in a limited duty ratio), the inductor conduction loss is reduced by a factor of the number of phases.
Further, the EDR boost converter has been studied for both discontinuous conduction mode (DCM) operations and operations with wide input/output voltage range in continuous conduction mode (CCM). A current sharing between its interleaved input phases is studied in detail to show that inherent sharing is possible for only in a limited duty ratio span, and modification of the duty ratio scheme is proposed to ensure equal current sharing over all the operating range for 3 phase EDR boost. All the analysis are validated with experimental results.
Moreover, wide band-gap devices (both SiC and GaN) are used for implementing their hardware prototypes. It enables the switching frequency to be high without compromising on the converter efficiency. Also it allows a reduced magnetic component size, further enabling a high power density solution, with power density far beyond the state-of-the art solutions.
Additionally, for the transformer-less microinverter application, another challenge is to achieve a very high gain DC-DC stage with a simultaneous high conversion efficiency. An extended duty ratio (EDR) boost converter which is a hybrid of switched capacitors and interleaved inductor technique, has been implemented for this purpose. It offers higher converter efficiency as most of the switches encounter lower voltage stress directly impacting switching loss; the input current being shared among all the interleaved converters (inherent sharing only in a limited duty ratio), the inductor conduction loss is reduced by a factor of the number of phases.
Further, the EDR boost converter has been studied for both discontinuous conduction mode (DCM) operations and operations with wide input/output voltage range in continuous conduction mode (CCM). A current sharing between its interleaved input phases is studied in detail to show that inherent sharing is possible for only in a limited duty ratio span, and modification of the duty ratio scheme is proposed to ensure equal current sharing over all the operating range for 3 phase EDR boost. All the analysis are validated with experimental results.
ContributorsRoy, Jinia (Author) / Ayyanar, Raja (Thesis advisor) / Karady, George G. (Committee member) / Qin, Jiangchao (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2017
Description
The growth of energy demands in recent years has been increasing faster than the expansion of transmission facility construction. This tendency cooperating with the continuous investing on the renewable energy resources drives the research, development, and construction of HVDC projects to create a more reliable, affordable, and environmentally friendly power grid.
Constructing the hybrid AC-HVDC grid is a significant move in the development of the HVDC techniques; the form of dc system is evolving from the point-to-point stand-alone dc links to the embedded HVDC system and the multi-terminal HVDC (MTDC) system. The MTDC is a solution for the renewable energy interconnections, and the MTDC grids can improve the power system reliability, flexibility in economic dispatches, and converter/cable utilizing efficiencies.
The dissertation reviews the HVDC technologies, discusses the stability issues regarding the ac and HVDC connections, proposes a novel power oscillation control strategy to improve system stability, and develops a nonlinear voltage droop control strategy for the MTDC grid.
To verify the effectiveness the proposed power oscillation control strategy, a long distance paralleled AC-HVDC transmission test system is employed. Based on the PSCAD/EMTDC platform simulation results, the proposed power oscillation control strategy can improve the system dynamic performance and attenuate the power oscillations effectively.
To validate the nonlinear voltage droop control strategy, three droop controls schemes are designed according to the proposed nonlinear voltage droop control design procedures. These control schemes are tested in a hybrid AC-MTDC system. The hybrid AC-MTDC system, which is first proposed in this dissertation, consists of two ac grids, two wind farms and a five-terminal HVDC grid connecting them. Simulation studies are performed in the PSCAD/EMTDC platform. According to the simulation results, all the three design schemes have their unique salient features.
Constructing the hybrid AC-HVDC grid is a significant move in the development of the HVDC techniques; the form of dc system is evolving from the point-to-point stand-alone dc links to the embedded HVDC system and the multi-terminal HVDC (MTDC) system. The MTDC is a solution for the renewable energy interconnections, and the MTDC grids can improve the power system reliability, flexibility in economic dispatches, and converter/cable utilizing efficiencies.
The dissertation reviews the HVDC technologies, discusses the stability issues regarding the ac and HVDC connections, proposes a novel power oscillation control strategy to improve system stability, and develops a nonlinear voltage droop control strategy for the MTDC grid.
To verify the effectiveness the proposed power oscillation control strategy, a long distance paralleled AC-HVDC transmission test system is employed. Based on the PSCAD/EMTDC platform simulation results, the proposed power oscillation control strategy can improve the system dynamic performance and attenuate the power oscillations effectively.
To validate the nonlinear voltage droop control strategy, three droop controls schemes are designed according to the proposed nonlinear voltage droop control design procedures. These control schemes are tested in a hybrid AC-MTDC system. The hybrid AC-MTDC system, which is first proposed in this dissertation, consists of two ac grids, two wind farms and a five-terminal HVDC grid connecting them. Simulation studies are performed in the PSCAD/EMTDC platform. According to the simulation results, all the three design schemes have their unique salient features.
ContributorsYu, Jicheng (Author) / Karady, George G. (Thesis advisor, Committee member) / Qin, Jiangchao (Thesis advisor, Committee member) / Ayyanar, Raja (Committee member) / Holbert, Keith E. (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2017
Description
The following report details the motivation, design, analysis, simulation and hardware implementation of a DC/DC converter in EV drivetrain architectures. The primary objective of the project was to improve overall system efficiency in an EV drivetrain. The methodology employed to this end required a variable or flexible DC-Link voltage at the input of the inverter stage. Amongst the several advantages associated with such a system are the independent optimization of the battery stack and the inverter over a wide range of motor operating conditions. The incorporation of a DC/DC converter into the drivetrain helps lower system losses but since it is an additional component, a number of considerations need to be made during its design. These include stringent requirements on power density, converter efficiency and reliability.
These targets for the converter are met through a number of different ways. The switches used are Silicon Carbide FETs. These are wide band gap (WBG) devices that can operate at high frequencies and temperatures. Since they allow for high frequency operation, a switching frequency of 250 khz is proposed and implemented. This helps with power density by reducing the size of passive components. High efficiencies are made possible by using a simple soft switching technique by augmenting the DC/DC converter with an auxiliary branch to enable zero voltage transition.
The efficacy of the approach is tested through simulation and hardware implementation of two different prototypes. The Gen-I prototype was a single soft switched synchronous boost converter rated at 2.5kw. Both the motoring mode and regenerative modes of operation (Boost and Buck) were hardware tested for over 2kw and efficiency results of over 98.15% were achieved. The Gen-II prototype and the main focus of this work is an interleaved soft switched synchronous boost converter. This converter has been implemented in hardware as well and has been tested at 6.7kw and an efficiency of over 98% has been achieved in the boost mode of operation.
These targets for the converter are met through a number of different ways. The switches used are Silicon Carbide FETs. These are wide band gap (WBG) devices that can operate at high frequencies and temperatures. Since they allow for high frequency operation, a switching frequency of 250 khz is proposed and implemented. This helps with power density by reducing the size of passive components. High efficiencies are made possible by using a simple soft switching technique by augmenting the DC/DC converter with an auxiliary branch to enable zero voltage transition.
The efficacy of the approach is tested through simulation and hardware implementation of two different prototypes. The Gen-I prototype was a single soft switched synchronous boost converter rated at 2.5kw. Both the motoring mode and regenerative modes of operation (Boost and Buck) were hardware tested for over 2kw and efficiency results of over 98.15% were achieved. The Gen-II prototype and the main focus of this work is an interleaved soft switched synchronous boost converter. This converter has been implemented in hardware as well and has been tested at 6.7kw and an efficiency of over 98% has been achieved in the boost mode of operation.
ContributorsRaza, Bassam (Author) / Ayyanar, Raja (Thesis advisor) / Qin, Jiangchao (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2019
Description
Prior work in literature has illustrated the benefits of using surge arrester as a way to improve the lighting performance of the substation and transmission line. Installing surge arresters would enhance the system reliability but it comes with an extra capital expenditure. This thesis provides simulation analysis to examine substation-specific applications of surge arrester as a way of determining the optimal, cost-effective placement of surge arresters. Four different surge arrester installation configurations are examined for the 500/230 kV Rudd substation which belongs to the utility, Salt River Project (SRP). The most efficient configuration is identified in this thesis. A new method “voltage-distance curve” is proposed in this work to evaluate different surge arrester installation configurations. Simulation results show that surge arresters only need to be equipped on certain location of the substation and can still ensure sufficient lightning protection.
With lower tower footing resistance, the lightning performance of the transmission line can typically be improved. However, when surge arresters are installed in the system, the footing resistance may have either negative or positive effect on the lightning performance. Different situations for both effects are studied in this thesis.
This thesis proposes a surge arrester installation strategy for the overhead transmission line lightning protection. In order to determine the most efficient surge arrester configuration of transmission line, the entire transmission line is divided into several line sections according to the footing resistance of its towers. A line section consists of the towers which have similar footing resistance. Two different designs are considered for transmission line lightning protection, they include: equip different number of surge arrester on selected phase of every tower, equip surge arresters on all phases of selected towers. By varying the number of the towers or the number of phases needs to be equipped with surge arresters, the threshold voltage for line insulator flashover is used to evaluate different surge arrester installation configurations. The way to determine the optimal surge arresters configuration for each line section is then introduced in this thesis.
With lower tower footing resistance, the lightning performance of the transmission line can typically be improved. However, when surge arresters are installed in the system, the footing resistance may have either negative or positive effect on the lightning performance. Different situations for both effects are studied in this thesis.
This thesis proposes a surge arrester installation strategy for the overhead transmission line lightning protection. In order to determine the most efficient surge arrester configuration of transmission line, the entire transmission line is divided into several line sections according to the footing resistance of its towers. A line section consists of the towers which have similar footing resistance. Two different designs are considered for transmission line lightning protection, they include: equip different number of surge arrester on selected phase of every tower, equip surge arresters on all phases of selected towers. By varying the number of the towers or the number of phases needs to be equipped with surge arresters, the threshold voltage for line insulator flashover is used to evaluate different surge arrester installation configurations. The way to determine the optimal surge arresters configuration for each line section is then introduced in this thesis.
ContributorsXia, Qianxue (Author) / Karady, George G. (Thesis advisor) / Ayyanar, Raja (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2018
Description
Recently, a novel non-iterative power flow (PF) method known as the Holomorphic Embedding Method (HEM) was applied to the power-flow problem. Its superiority over other traditional iterative methods such as Gauss-Seidel (GS), Newton-Raphson (NR), Fast Decoupled Load Flow (FDLF) and their variants is that it is theoretically guaranteed to find the operable solution, if one exists, and will unequivocally signal if no solution exists. However, while theoretical convergence is guaranteed by Stahl’s theorem, numerical convergence is not. Numerically, the HEM may require extended precision to converge, especially for heavily-loaded and ill-conditioned power system models.
In light of the advantages and disadvantages of the HEM, this report focuses on three topics:
1. Exploring the effect of double and extended precision on the performance of HEM,
2. Investigating the performance of different embedding formulations of HEM, and
3. Estimating the saddle-node bifurcation point (SNBP) from HEM-based Thévenin-like networks using pseudo-measurements.
The HEM algorithm consists of three distinct procedures that might accumulate roundoff error and cause precision loss during the calculations: the matrix equation solution calculation, the power series inversion calculation and the Padé approximant calculation. Numerical experiments have been performed to investigate which aspect of the HEM algorithm causes the most precision loss and needs extended precision. It is shown that extended precision must be used for the entire algorithm to improve numerical performance.
A comparison of two common embedding formulations, a scalable formulation and a non-scalable formulation, is conducted and it is shown that these two formulations could have extremely different numerical properties on some power systems.
The application of HEM to the SNBP estimation using local-measurements is explored. The maximum power transfer theorem (MPTT) obtained for nonlinear Thévenin-like networks is validated with high precision. Different numerical methods based on MPTT are investigated. Numerical results show that the MPTT method works reasonably well for weak buses in the system. The roots method, as an alternative, is also studied. It is shown to be less effective than the MPTT method but the roots of the Padé approximant can be used as a research tool for determining the effects of noisy measurements on the accuracy of SNBP prediction.
In light of the advantages and disadvantages of the HEM, this report focuses on three topics:
1. Exploring the effect of double and extended precision on the performance of HEM,
2. Investigating the performance of different embedding formulations of HEM, and
3. Estimating the saddle-node bifurcation point (SNBP) from HEM-based Thévenin-like networks using pseudo-measurements.
The HEM algorithm consists of three distinct procedures that might accumulate roundoff error and cause precision loss during the calculations: the matrix equation solution calculation, the power series inversion calculation and the Padé approximant calculation. Numerical experiments have been performed to investigate which aspect of the HEM algorithm causes the most precision loss and needs extended precision. It is shown that extended precision must be used for the entire algorithm to improve numerical performance.
A comparison of two common embedding formulations, a scalable formulation and a non-scalable formulation, is conducted and it is shown that these two formulations could have extremely different numerical properties on some power systems.
The application of HEM to the SNBP estimation using local-measurements is explored. The maximum power transfer theorem (MPTT) obtained for nonlinear Thévenin-like networks is validated with high precision. Different numerical methods based on MPTT are investigated. Numerical results show that the MPTT method works reasonably well for weak buses in the system. The roots method, as an alternative, is also studied. It is shown to be less effective than the MPTT method but the roots of the Padé approximant can be used as a research tool for determining the effects of noisy measurements on the accuracy of SNBP prediction.
ContributorsLi, Qirui (Author) / Tylavsky, Daniel (Thesis advisor) / Lei, Qin (Committee member) / Weng, Yang (Committee member) / Arizona State University (Publisher)
Created2018
Description
With the growing penetration of plug-in electric vehicles (PEVs), the impact of the PEV charging brought to the utility grid draws more and more attention. This thesis focused on the optimization of a home energy management system (HEMS) with the presence of PEVs. For a household microgrid with photovoltaic (PV) panels and PEVs, a HEMS using model predictive control (MPC) is designed to achieve the optimal PEV charging. Soft electric loads and an energy storage system (ESS) are also considered in the optimization of PEV charging in the MPC framework. The MPC is solved through mixed-integer linear programming (MILP) by considering the relationship of energy flows in the optimization problem. Through the simulation results, the performance of optimization results under various electricity price plans is evaluated. The influences of PV capacities on the optimization results of electricity cost are also discussed. Furthermore, the hardware development of a microgrid prototype is also described in this thesis.
ContributorsZhao, Yue (Author) / Chen, Yan (Thesis advisor) / Johnson, Nathan (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2018
Description
The concept of the microgrid is widely studied and explored in both academic and industrial societies. The microgrid is a power system with distributed generations and loads, which is intentionally planned and can be disconnected from the main utility grid. Nowadays, various distributed power generations (wind resource, photovoltaic resource, etc.) are emerging to be significant power sources of the microgrid.
This thesis focuses on the system structure of Photovoltaics (PV)-dominated microgrid, precisely modeling and stability analysis of the specific system. The grid-connected mode microgrid is considered, and system control objectives are: PV panel is working at the maximum power point (MPP), the DC link voltage is regulated at a desired value, and the grid side current is also controlled in phase with grid voltage. To simulate the real circuits of the whole system with high fidelity instead of doing real experiments, PLECS software is applied to construct the detailed model in chapter 2. Meanwhile, a Simulink mathematical model of the microgrid system is developed in chapter 3 for faster simulation and energy management analysis. Simulation results of both the PLECS model and Simulink model are matched with the expectations. Next chapter talks about state space models of different power stages for stability analysis utilization. Finally, the large signal stability analysis of a grid-connected inverter, which is based on cascaded control of both DC link voltage and grid side current is discussed. The large signal stability analysis presented in this thesis is mainly focused on the impact of the inductor and capacitor capacity and the controller parameters on the DC link stability region. A dynamic model with the cascaded control logic is proposed. One Lyapunov large-signal stability analysis tool is applied to derive the domain of attraction, which is the asymptotic stability region. Results show that both the DC side capacitor and the inductor of grid side filter can significantly influence the stability region of the DC link voltage. PLECS simulation models developed for the microgrid system are applied to verify the stability regions estimated from the Lyapunov large signal analysis method.
This thesis focuses on the system structure of Photovoltaics (PV)-dominated microgrid, precisely modeling and stability analysis of the specific system. The grid-connected mode microgrid is considered, and system control objectives are: PV panel is working at the maximum power point (MPP), the DC link voltage is regulated at a desired value, and the grid side current is also controlled in phase with grid voltage. To simulate the real circuits of the whole system with high fidelity instead of doing real experiments, PLECS software is applied to construct the detailed model in chapter 2. Meanwhile, a Simulink mathematical model of the microgrid system is developed in chapter 3 for faster simulation and energy management analysis. Simulation results of both the PLECS model and Simulink model are matched with the expectations. Next chapter talks about state space models of different power stages for stability analysis utilization. Finally, the large signal stability analysis of a grid-connected inverter, which is based on cascaded control of both DC link voltage and grid side current is discussed. The large signal stability analysis presented in this thesis is mainly focused on the impact of the inductor and capacitor capacity and the controller parameters on the DC link stability region. A dynamic model with the cascaded control logic is proposed. One Lyapunov large-signal stability analysis tool is applied to derive the domain of attraction, which is the asymptotic stability region. Results show that both the DC side capacitor and the inductor of grid side filter can significantly influence the stability region of the DC link voltage. PLECS simulation models developed for the microgrid system are applied to verify the stability regions estimated from the Lyapunov large signal analysis method.
ContributorsXu, Hongru (Author) / Chen, Yan (Thesis advisor) / Johnson, Nathan (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2018
Description
The most important metrics considered for electric vehicles are power density, efficiency, and reliability of the powertrain modules. The powertrain comprises of an Electric Machine (EM), power electronic converters, an Energy Management System (EMS), and an Energy Storage System (ESS). The power electronic converters are used to couple the motor with the battery stack. Including a DC/DC converter in the powertrain module is favored as it adds an additional degree of freedom to achieve flexibility in optimizing the battery module and inverter independently. However, it is essential that the converter is rated for high peak power and can maintain high efficiency while operating over a wide range of load conditions to not compromise on system efficiency. Additionally, the converter must strictly adhere to all automotive standards.
Currently, several hard-switching topologies have been employed such as conventional boost DC/DC, interleaved step-up DC/DC, and full-bridge DC/DC converter. These converters face respective limitations in achieving high step-up conversion ratio, size and weight issues, or high component count. In this work, a bi-directional synchronous boost DC/DC converter with easy interleaving capability is proposed with a novel ZVT mechanism. This converter steps up the EV battery voltage of 200V-300V to a wide range of variable output voltages ranging from 310V-800V. High power density and efficiency are achieved through high switching frequency of 250kHz for each phase with effective frequency doubling through interleaving. Also, use of wide bandgap high voltage SiC switches allows high efficiency operation even at high temperatures.
Comprehensive analysis, design details and extensive simulation results are presented. Incorporating ZVT branch with adaptive time delay results in converter efficiency close to 98%. Experimental results from a 2.5kW hardware prototype validate the performance of the proposed approach. A peak efficiency of 98.17% has been observed in hardware in the boost or motoring mode.
Currently, several hard-switching topologies have been employed such as conventional boost DC/DC, interleaved step-up DC/DC, and full-bridge DC/DC converter. These converters face respective limitations in achieving high step-up conversion ratio, size and weight issues, or high component count. In this work, a bi-directional synchronous boost DC/DC converter with easy interleaving capability is proposed with a novel ZVT mechanism. This converter steps up the EV battery voltage of 200V-300V to a wide range of variable output voltages ranging from 310V-800V. High power density and efficiency are achieved through high switching frequency of 250kHz for each phase with effective frequency doubling through interleaving. Also, use of wide bandgap high voltage SiC switches allows high efficiency operation even at high temperatures.
Comprehensive analysis, design details and extensive simulation results are presented. Incorporating ZVT branch with adaptive time delay results in converter efficiency close to 98%. Experimental results from a 2.5kW hardware prototype validate the performance of the proposed approach. A peak efficiency of 98.17% has been observed in hardware in the boost or motoring mode.
ContributorsMullangi Chenchu, Hemanth (Author) / Ayyanar, Raja (Thesis advisor) / Qin, Jiangchao (Committee member) / Lei, Qin (Committee member) / Arizona State University (Publisher)
Created2018
Description
Nowadays, the widespread introduction of distributed generators (DGs) brings great challenges to the design, planning, and reliable operation of the power system. Therefore, assessing the capability of a distribution network to accommodate renewable power generations is urgent and necessary. In this respect, the concept of hosting capacity (HC) is generally accepted by engineers to evaluate the reliability and sustainability of the system with high penetration of DGs. For HC calculation, existing research provides simulation-based methods which are not able to find global optimal. Others use OPF (optimal power flow) based methods where
too many constraints prevent them from obtaining the solution exactly. They also can not get global optimal solution. Due to this situation, I proposed a new methodology to overcome the shortcomings. First, I start with an optimization problem formulation and provide a flexible objective function to satisfy different requirements. Power flow equations are the basic rule and I transfer them from the commonly used polar coordinate to the rectangular coordinate. Due to the operation criteria, several constraints are
incrementally added. I aim to preserve convexity as much as possible so that I can obtain optimal solution. Second, I provide the geometric view of the convex problem model. The process to find global optimal can be visualized clearly. Then, I implement segmental optimization tool to speed up the computation. A large network is able to be divided into segments and calculated in parallel computing where the results stay the same. Finally, the robustness of my methodology is demonstrated by doing extensive simulations regarding IEEE distribution networks (e.g. 8-bus, 16-bus, 32-bus, 64-bus, 128-bus). Thus, it shows that the proposed method is verified to calculate accurate hosting capacity and ensure to get global optimal solution.
too many constraints prevent them from obtaining the solution exactly. They also can not get global optimal solution. Due to this situation, I proposed a new methodology to overcome the shortcomings. First, I start with an optimization problem formulation and provide a flexible objective function to satisfy different requirements. Power flow equations are the basic rule and I transfer them from the commonly used polar coordinate to the rectangular coordinate. Due to the operation criteria, several constraints are
incrementally added. I aim to preserve convexity as much as possible so that I can obtain optimal solution. Second, I provide the geometric view of the convex problem model. The process to find global optimal can be visualized clearly. Then, I implement segmental optimization tool to speed up the computation. A large network is able to be divided into segments and calculated in parallel computing where the results stay the same. Finally, the robustness of my methodology is demonstrated by doing extensive simulations regarding IEEE distribution networks (e.g. 8-bus, 16-bus, 32-bus, 64-bus, 128-bus). Thus, it shows that the proposed method is verified to calculate accurate hosting capacity and ensure to get global optimal solution.
ContributorsYuan, Jingyi (Author) / Weng, Yang (Thesis advisor) / Lei, Qin (Committee member) / Khorsand, Mojdeh (Committee member) / Arizona State University (Publisher)
Created2018
Description
Over the years, the growing penetration of renewable energy into the electricity market has resulted in a significant change in the electricity market price. This change makes the existing forecasting method prone to error, decreasing the economic benefits. Hence, more precise forecasting methods need to be developed. This paper starts with a survey and benchmark of existing machine learning approaches for forecasting the real-time market (RTM) price. While these methods provide sufficient modeling capability via supervised learning, their accuracy is still limited due to the single data source, e.g., historical price information only. In this paper, a novel two-stage supervised learning approach is proposed by diversifying the data sources such as highly correlated power data. This idea is inspired by the recent load forecasting methods that have shown extremely well performances. Specifically, the proposed two-stage method, namely the rerouted method, learns two types of mapping rules. The first one is the mapping between the historical wind power and the historical price. The second is the forecasting rule for wind generation. Based on the two rules, we forecast the price via the forecasted generation and the first learned mapping between power and price. Additionally, we observed that it is not the more training data the better, leading to our validation steps to quantify the best training intervals for different datasets. We conduct comparisons of numerical results between existing methods and the proposed methods based on datasets from the Electric Reliability Council of Texas (ERCOT). For each machine learning step, we examine different learning methods, such as polynomial regression, support vector regression, neural network, and deep neural network. The results show that the proposed method is significantly better than existing approaches when renewables are involved.
ContributorsLuo, Shuman (Author) / Weng, Yang (Thesis advisor) / Lei, Qin (Committee member) / Qin, Jiangchao (Committee member) / Arizona State University (Publisher)
Created2019