Search

This paper proposed a low-cost method to build the machine learning training dataset for assessing service transformer health by using fuzzy logic method. The training dataset is tested on a stimulated 50kVA server transformer. The monitoring data is collected from the real-time energy monitoring device which is installed near the transformer to measure ambient temperature, current, and voltage. The condition of transformer is evaluated by using Support Vector Machine algorithm. The data generation proposed in this paper has high feature continuity and good scalability that can be used as a training data for machine learning, deep learning models.

Due to the increasing penetration of distributed behind-the-meter photovoltaic (PV) systems and the installed utility revenue metering limited to monitoring only the net power import/export of the household, it is increasingly challenging for utilities to effectively plan and operate the grid. This paper proposes a methodology that estimates behind-the-meter PV generation using a selected subset of monitored PV systems. It is a data-driven approach, and the PV output is estimated utilizing a statistic regression model. A Minimum Redundancy Maximum Relevance (MRMR) algorithm is applied to preselect the optimal subset of the monitored PV systems. The performance of this approach is compared with a spatial interpolation method and a model-based approach. The proposed method is validated using high-resolution meter data recorded from 18 residential rooftop PV systems located on the island of Maui, Hawaii.

As the electricity generation is shifting to renewable energy sources (RES), the grid infrastructure faces multiple challenges, such as intermittency and volatility of a wide range RES. A high penetration of renewables requires profound changes to the current energy distribution system. The conventional grid is increasingly becoming a bottleneck for expanding the share of RES because of its rigid architecture, which is built around centralized energy source. We propose a new energy exchange model for a routed energy distribution system, which can perform electricity routing based on smart routing algorithms and presented protocols. We utilize a concept of an energy router device that controls energy flows and utilizes protocols stack to smartly route the energy between houses in the grid. This paper describes current results with experimental network of Maui village with multiple houses interconnected through energy routers.

Conservation voltage reduction (CVR) is widely adopted by utilities for energy savings through reducing the voltage level of the distribution system while maintaining the service voltage within standard voltage limits. However, an unexpected voltage violation can occur due to the intermittent nature of photovoltaic (PV) output. In this paper, a fast and adaptive voltage control algorithm is proposed based on the realtime measurement data from customers' smart meters to improve the CVR effect and prevent voltage violations. In particular, a Power Hardware-in-the-loop (PHIL) test platform is set up to validate the proposed control algorithm. The test circuit and measurement data are from an actual LV residential feeder located on the island of Maui, Hawaii. Simulation results demonstrate that the proposed algorithm can achieve additional energy savings compared to a traditional CVR approach while maintaining the feeder voltages within acceptable limits.

This paper presents an online method of monitoring the health of distribution service transformers using data from a real-time data acquisition and analysis system located at the transformer. The results can be provided to utility operators in near real-time, enabling them to assess the status of the distribution transformer and take actions to avoid or mitigate costly transformer failures and service outages. After thorough review of potential non-invasive monitoring techniques, top oil temperature, vibration, transformer loading, and power factor were chosen as monitored parameters, and the overall health index is determined using fuzzy logic. A case study is implemented on a specific distribution transformer to show the advantages of the proposed method.

Computer networks passed many stages before they became so much developed and effective as we see them nowadays. It is time to make a big step to the future in the energy distribution too. It might be hard to believe but energy networks and computer networks have much in common. Both have customers that are connected to network cables, and in both cases the main task of the network is to provide the requested flow. On the one hand it is information, on the other hand it is energy. The same as computer network customers, houses in energy network can either request energy for their needs or generate and send energy to the network when there are photovoltaic panels or other generators installed. In this paper we show how achievements from computer world can contribute to energy network and quickly bring it to the next generation of energy management.

The impact of global warming on personal health is a growing concern, as high temperatures can cause heat stress disorders. Personal cooling systems are a promising solution in environments where air conditioning is not available or infeasible such as the outdoor working environment of utility workers, and treatment of hyperthermia for athletes. However, current untethered personal cooling systems are expensive, heavy, or have a limited cooling duration due to high power consumption. Here, we have designed and prototyped a lightweight and low-power active personal cooling system that uses the non-toxic gallium-based liquid-metal, Galinstan, as the coolant. The Galinstan absorbs body heat as it is pumped through tubing attached to a cooling vest by a low-power DC pump. After absorbing the body heat, the warm Galinstan is cooled by passing it through a network of tubing embedded in cold phase-change material in a cold pack. The cold pack is insulated from the external environment, allowing heat absorption from only the warm liquid metal. To the best of the author’s knowledge, the liquid metal active personal cooling system is 1/3rd of the weight, 1/8th of the price, and cools for 4 times longer compared to the current state-of-art commercial active personal cooling systems.

Due to the large number of distribution transformers in the distribution grid, the status of distribution transformers plays an important role in ensuring the safe and reliable operation of the these grids. To evaluate the distribution transformer health, many assessment techniques have been studied and developed. These tools will support the transformer operators in predicting the status of the distribution transformer and responding effectively. This paper will review the literature in the area, analyze the latest techniques as well as highlight the advantages and disadvantages of current methodologies.

Distributed photovoltaic (PV) generator systems (especially rooftop PV) have been increasing significantly in electrical power systems. However, due to the variability of their output power, it is challenging to integrate large number of these PV generator systems into the existing electrical grid. These distributed PV generator systems can cause large voltage fluctuations due to the oscillation in their power output and reverse power flow in highly penetrated areas. While real power curtailment is an option to manage voltage issues, it reduces the production of this clean energy and is governed by policies and contracts. Traditionally, the use of reactive power to control voltage on a power grid is done through maintaining the voltage within a tolerable range by using transformer tap-changers (TPC), capacitor banks (CB), voltage regulators (VR), static synchronous compensator (STATCOM), and static Var compensators (SVC). Installation and maintenance costs of these devices can be quite high and some have relatively slow response times on the order of many seconds. A smart PV inverter can help regulate voltage by absorbing and injecting reactive power (Var) to/from the grid by using the Volt-Var control function. This paper presents an experimental analysis of the inverter Volt-Var control method for voltage regulation. The capacitive (i.e., Var injection) and inductive (i.e., Var absorption) effects of using a smart inverter and its ability to influence the voltage at the distribution level is investigated in this paper. When the smart PV inverter injects reactive power, it increases the distribution voltage. Conversely, voltage is reduced when the smart inverter absorbs reactive power. As a result, electric power utilities can control the distribution voltage without installing additional devices on the power network. The analyses from a field deployment under the Maui Advanced Solar Initiative Project in Hawaii is discussed here.

As the electricity generation is shifting to renewable energy sources (RES), the grid infrastructure faces multiple challenges, such as intermittency and volatility of a wide range RES. A high penetration of renewables requires profound changes to the current energy distribution system. The conventional grid is increasingly becoming a bottleneck for expanding the share of RES because of its rigid architecture, which is built around centralized energy source. We propose a new energy exchange model for a routed energy distribution system, which can perform electricity routing based on smart routing algorithms and presented protocols. We utilize a concept of an energy router device that controls energy flows and utilizes protocols stack to smartly route the energy between houses in the grid. This paper describes current results with several experimental networks of multiple houses interconnected through energy routers.

A battery energy storage system (BESS) is an available solution for utilities to deal with intermittency issues resulting from renewable energy resources. A BESS needs to have a control algorithm to provide a very good estimation of the load on the grid at each time step. A short-term load forecast (STLF) is necessary for efficient and optimized control of BESSs that are connected to the grid. In this work, two parallel-series techniques for load forecasting are proposed to optimize the performance of a grid-scale BESS (1 MW, 1.1 kWh) in 15-min steps within a moving 24-h window. In both techniques, a complex-valued neural network (CVNN) is used for parallel forecasting. The parallel component is based on the search for similar days of historical data that have a weekly index comparable to the forecast day. For series forecasting, historical data of each day is used within a moving forecast window by CVNN along with the spline method. For both techniques, parallel forecasting is mixed with series forecasting by an adjustment coefficient. Both techniques are tested on a set of real data for a grid with high PV penetration, and the obtained results are compared.

In 2008, the State of Hawaii initiated a clean energy initiative that set an ultimate goal of 70% clean energy by 2030 (40% from renewable energy and 30% from energy efficiency). A controllable Battery Energy Storage Systems (BESSs) can be used to manage intermittent renewable resources on a power system to address both circuit and system level issues. Simulation and experimental results of applying a novel algorithm for the charging and discharging of a BESS are presented, using actual grid data for controlling a BESS for the purpose of peak load shaving, power curve smoothing, and voltage regulation of a distribution transformer. Two optimization objectives for peak shaving are presented in which proposed load forecasting methods are used. The application of a BESS for voltage regulation is examined and analyzed with different tests, and the observed results are discussed.

Large-scale deployments of variable renewable energy sources, such as photovoltaic (PV) systems, are increasing around the world. PV systems are a potential source of clean energy but they experience power intermittency and lack controllability. However, smart PV inverters provide grid-friendly functionalities that control the power output of PV systems. Power intermittency of PV systems causes major problems such as voltage fluctuations and frequency deviations in an electric power grid. Together with varying loads and other renewable distributed generations, the grid frequency and voltage become difficult to manage. A smart PV inverter allows for active power curtailment, volt-watt, and frequency-watt control systems to mitigate the voltage and frequency deviations for a smart power grid. Therefore, smart PV inverters can reduce the installation of voltage and frequency control devices in an electric power grid, which in turn can reduce installation and maintenance costs. Real-time analyses of active power curtailment, volt-watt control, frequency-watt control using smart PV inverters and their effects on voltage and frequency are presented in this paper. Power curtailment operations were executed in low and high PV penetration areas. From the experimental analyses, distributed voltage deviations can be mitigated using active power curtailment and volt-watt control systems, and the over-frequency of an electric power grid can be reduced using frequency-watt control method. Controller performances of smart PV inverters are also evaluated in this paper. The experimental analyses were executed in the Maui Advanced Solar Initiative Project.

With the increase of load demand and distributed photovoltaic (PV) systems on the electric grid, maintaining the required voltage tolerance at the point delivery (customer homes/businesses) is becoming more challenging for electric power utilities. In a residential neighborhood, the peak load typically occurs in the early evening hours while maximum PV generation occurs during mid day. As a result, the lowest voltage operating points occur in the evening hours; whereas the highest voltage operating point occur during the day, when the PV systems are injecting more power than what is locally consumed. Static VAR Compensators (SVCs) can be used to mitigate voltage violations and smooth out the overall profile of a distribution feeder by managing the voltage at the secondary side of distribution service transformers. In this paper, a new method to optimize the placement of SVC devices in a residential neighborhood on the island of Maui, Hawaii, USA with highly distributed PV penetration is presented as a case study using real data to show the efficiency of the proposed optimal placement methodology.

The integration of distributed generations (DGs) in the electric power grid introduces distribution voltage (DV) fluctuations. Load demand variations are another source of DV fluctuations. The DV needs to be maintained within five percent of its nominal value. Several approaches exist for controlling the DV e.g. transformer tap-changers (TPCs), capacitor banks (CBs), voltage regulators (VRs), and static VAR compensators (SVCs). However, these DV control methods require installation and maintenance of new equipment for the electric power utilities. In addition, these devices have a relatively long response time. Due to the increase of DG integration, the DV needs to react promptly. Distributed photovoltaic (PV) generation have been increasing rapidly. The smart PV inverter enables several strategies to control the DV. For example, the smart PV inverter can absorb and generate reactive power (Var) using the Volt-Var control strategy to regulate the DV. This paper shows an experimental study of the Volt-Var control method using a smart PV inverter. The Volt-Var control strategy was tested using a smart SMA PV inverter. When the DV drops, the smart PV inverter dispatches Var to increase the DV. On the other hand, when the DV is higher than its nominal value, the smart PV inverter absorbs Var to decrease the DV. The real-time analysis of the Volt-Var control method was investigated in the Maui Advanced Solar Initiative project.

This paper examines the active power control of a photovoltaic (PV) power generation system. A traditional PV power generation system lacks controllability. However, smart PV inverters provide a mechanism to control the power output of PV generators. Distributed PV generator systems have been increasing for the past few decades as a clean renewable energy resource. As more distributed PV is interconnected to the power grid, the integration of these highly variable resources presents a challenging task for the utility. The power output of the PV generator system is proportional to the solar irradiance. Solar irradiance may vary due to water vapor, clouds, pollution, the season of the year and the time of day. As a result, the power output of a PV generator is a highly varying resource. Coupled with varying loads, the grid frequency and voltage become difficult to manage. One method to control voltage is by active power curtailment of the PV inverter. This paper presents analyses of active power curtailment and its effect on voltage in the Maui Advanced Solar Initiative Project where smart inverters were field deployed in a residential neighborhood.

As the world moves toward greenhouse gas reduction, there is increasingly active work around Li-ion chemistry-based batteries as an energy source for electric vehicles (EVs), hybrid electric vehicles (HEVs) and smart grids. In these applications, the battery management system (BMS) requires an accurate online estimation of the state of charge (SOC) in a battery pack. This estimation is difficult, especially after substantial battery aging. In order to address this problem, this paper utilizes SOC estimation of Li-ion battery packs using a fuzzy-improved extended Kalman filter (fuzzy-IEKF) for Li-ion cells, regardless of their age. The proposed approach introduces a fuzzy method with a new class and associated membership function that determines an approximate initial value applied to SOC estimation. Subsequently, the EKF method is used by considering the single unit model for the battery pack to estimate the SOC for following periods of battery use. This approach uses an adaptive model algorithm to update the model for each single cell in the battery pack. To verify the accuracy of the estimation method, tests are done on a LiFePO 4 aged battery pack consisting of 120 cells connected in series with a nominal voltage of 432 V.

Power demand of isolated communities or islands primarily relies on heavy oil based diesel generator systems where these fossil fuel based power generators are not only costly but also create adverse effect on the environment. Photovoltaic generator system (PV system), wind energy conversion system (WECS) and battery energy storage system (BESS) are clean sources of alternative energy. These energy sources can be the alternative of diesel generator based power system for an isolated communities such as islands. However, both PV systems and WECSs are uncertain and fluctuating sources of energy causing power fluctuations and frequency deviations. This paper proposes a coordinated approach of the frequency control for a zero emission based small power system with renewable energy penetration. PV system, WECS and BESS have been utilized to control frequency of the proposed zero emission based power system. Efficacy of the proposed frequency control system is shown with simulations results.

Starting in 2009, the Hawaii Natural Energy Institute, the Hawaiian Electric Company, and the General Electric Company, in collaboration with DOE, jointly developed and validated state-of-the-art power systems models of the Oahu and Maui electrical systems to study the operational impacts of increasing levels of wind and solar generation. These models were central to a series of coordinated, detailed technical studies which included the 2010 Hawaiian Islands Transmission Interconnection Studies (Stage I: Cable), the 2011 Oahu Wind Integration and Transmission Study (OWITS), the 2012 Hawaii Solar Integration Study (HSIS), and the 2013 Hawaiian Islands Interconnection Studies (Stage II: Grid Tie). This article focuses on key OWITS and HSIS findings for Oahu.

Energy systems integration in the Hawaiian island of Maui is unique because of the island's high renewable energy (RE) penetration, advanced adoption of plug-in vehicles, distributed power generation, and isolated island electrical system. Currently, a large amount of wind power produced on Maui is curtailed because the power cannot be accommodated by the existing power grid. Analysis of changes in electrical grid operations to reduce wind and anticipated solar curtailment and look at energy pathways across the electrical and transportation sectors has resulted in a more comprehensive understanding of energy use on the island.

Variability of power generation from intermittent resources such as wind and solar plants presents an operational challenge for grid operators. The economic incentives and technical challenges that accompany large amounts of variable generation in island power systems are often much greater. The Hawaiian Electric Company and its subsidiaries, the Maui Electric Company and the Hawaii Electric Light Company have considerable experience in planning and operating power systems with relatively high levels of wind power. The islands of Hawaii and Maui operate power systems with high levels of wind power (more than 10% by energy) and have experienced and addressed challenges associated with the variability and uncertainty of wind power. The island of Maui is anticipating further wind plant deployments in the near future. Recent analyses of possible near-term deployment of large amounts of wind power on the Oahu power system (500MW of wind power; approximately 1200MW peak and 520MW minimum annual load) has shown the potential for this system to accept almost 25% of its energy from wind and solar power. This paper will identify some of the wind integration challenges and highlight the benefits of a variety of strategies that are expected to improve system economics and operational reliability, including proposed modifications to the baseload thermal fleet (deeper turndown, higher ramp rates, and tuned droop characteristics), advanced wind turbine grid support features, new operating strategies, wind forecasting and refinements to the up and down reserve requirements. This paper will present the key findings in the context of useful insights and lessons learned that are relevant to other island power systems considering very high levels of wind power.