Search

This paper presents a data-driven method for estimating annual continuous dynamic rating of power transformers to serve the long-term planning purpose. Historically, research works on dynamic rating have been focused on real-time/near-future system operations. There has been a lack of research for long-term planning oriented applications. Currently, most utility companies still rely on static rating numbers when planning power transformers for the next few years. In response, this paper proposes a novel and comprehensive method to analyze the past 5-year temperature, loading and load composition data of existing power transformers in a planning region. Based on such data and the forecasted area load composition, a future power transformer's load shape profile can be constructed by using Gaussian Mixture Model. Then according to IEEE std. C57.91-2011, a power transformer thermal aging model can be established to incorporate future loading and temperature profiles. As a result, annual continuous dynamic rating profiles under different temperature scenarios can be determined. The profiles can reflect the long-term thermal overloading risk in a much more realistic and granular way, which can significantly improve the accuracy of power transformer planning. A real utility application example in Canada has been presented to validate and demonstrate the practicality and usefulness of this method.

This paper points to some forecasting problems, especially the one which occur after some sudden, intensive and durable reduction of peak load. In such circumstances, decreasing peak load trend can last for several years, causing the problems by appliance of classical, trend-based extrapolation forecasting methods. The paper presents three alternative methods for the peak load forecasting, developed in order to eliminate, avoid or reduce these problems. The methods are illustrated here by some examples, in which the peak load value has been reduced by changed tariff system for electricity sale. However, the methods can be applied also in the case of any other-sudden and intensive-influence which could reduce the peak load value in an local area or wider region.

Operational performance in the low-voltage distribution network (LVDN) can be undermined by its inherent unbalances, which may become worse as the penetration of rooftop solar continuously increases. To address this issue, load balancing via phase-reconfiguration devices (PRDs), which can change phase positions of residential customers as required, provides a cost-efficient option. However, most reported approaches to control PRDs require that demands of all residential customers are available, which are not viable for many LVDNs without smart meters or advanced metering infrastructure (AMI) installed. To bridging the gap in this field, this paper proposes a novel method to control PRDs purely based on measurable data from PRDs, and its controller. Based on limited information, sensitivity analysis in the network with PRDs is studied, followed by the optimization model that comprehensively considers operational requirements in the network. Moreover, slack variables are introduced to the model, and penalized in the objective function to assure either a strategy that is secure or with minimized violations can always be provided. The model is a challenging mixed-integer non-convex programming (MINCP) problem, which is reformulated as an efficient solvable mixed-integer second-order cone programming (MISOCP) based on exact reformulations or accurate linear approximations. Simulations based on two modified IEEE systems, and a real system in Australia demonstrate that an efficient strategy can be provided to mitigate unbalances in the network.

Building automation technology is rapidly developing towards more reliable communication systems, devices that control electronic equipment. This equipment, if controlled, leads to efficient energy management, and savings on the monthly electricity bill. Power line communication (PLC) has been one of the dreams of the electronics industry for decades, especially for building automation. It is the purpose of this paper to demonstrate communication methods among electronic control devices through an AC power line carrier within the buildings for more efficient energy control. The paper outlines methods of communication over a power line, namely the X-10 and CE bus. It also introduces the spread spectrum technology as to increase speed to 100-150 times faster than the X-10 system. The power line carrier has tremendous applications in the field of building automation. The paper presents an attempt to realize a smart house concept, so called, in which all home electronic devices from a coffee maker to a water heater microwave to chaos robots will be utilized by an intelligent network whenever one wishes to do so. The designed system may be applied very profitably to help in energy management for both customer and utility.

The paper deals with the power properties of three-phase circuits with unbalanced loads supplied from a three-wire line. It discusses various definitions of the apparent power in such circuits and shows that they cannot be interpreted in terms of any physical phenomena. The paper shows that the unbalanced power does not occur as an effect of the power oscillations and explains why this power does occur. The paper presents a decomposition of the supply current into the active, reactive, and unbalanced currents and shows that this decomposition can be considered as a fundamental for the comprehension of the power phenomena in such circuits and for the power factor evaluation. >

A communication system for distribution automation and demand side management typically supports a multi-tasking environment. In such a system, "time" is a precious commodity, and the loss of time due to a dysfunction of the system can be used as a measure of added costs. This paper explores how unforeseen "added costs of utilization" are incurred and what the root causes are. A simple "two-way transaction equation" is derived showing how added costs can be assessed and studied. The paper also points out what remedial steps should be taken to reduce these added costs, and points out that no communication system is immune to outages in the distribution network. >

This paper presents autonomous control algorithms for electrical loads like HVAC (Heating, Ventilation, and Air Conditioning) systems. These algorithms permit a load to become “aware” of the operation of neighboring loads using nonintrusive measurements of the utility voltage. Analogous to the way a good driver is aware of neighboring cars while driving, loads can use this information to become “self-driving.” For HVAC systems, this permits maintaining occupant comfort while simultaneously reducing bulk peak electrical demand. Electric energy consumption for each load occurs on a schedule, interleaved with the operation of neighboring units. The algorithms are demonstrated here using simulation models developed from nonintrusive load monitoring (NILM) data collected from the microgrid of a US Army forward operating base (FOB). The general approach, however, can be applied in many other venues, e.g., to minimize peak load on distribution transformers in a section of the conventional utility.

An improved switching scheme for static VAr compensators (SVCs) that have fixed capacitor-thyristor-controlled reactors (FC-TCRs) for load balancing and power factor improvement has been addressed in this paper. The proposed control scheme can eliminate negative-sequence current caused by unbalanced load while improving the source power factor simultaneously. Thus total current supplied by the source can be minimized along with improvement of the load voltage profile. As the switching of TCR can introduce additional harmonics in the line, an optimized switching function is adopted in the proposed scheme for minimum injected harmonics without using an external filter. The gravitational search algorithm has been used for obtaining the optimized switching angle. The proposed scheme has been verified through proper simulations backed by suitable experimental results on a practical system.

This paper presents the experimental determination of the ZIP coefficients model to represent (static) modern loads under varying voltage conditions. ZIP are the coefficients of a load model comprised of constant impedance $Z$ , constant current $I$ , and constant power $P$ loads. A ZIP coefficient load model is used to represent power consumed by a load as a function of voltage. A series of surveys was performed on typical residential, commercial, and industrial customers in New York City. Household appliances and industrial equipment found in the different locations were tested in the laboratory by varying the voltage from 1.1-p.u. voltage to 0 and back to 1.1 pu in steps of 3 V to obtain the individual $P$ – $V$ , $Q$ – $V$ , and $I $ – $V$ characteristics. Customer load tables were built using seasonal factors and duty cycles to form weighted contributions for each device in every customer class. The loads found in several residential classes were assembled and tested in the lab. It was found that modern appliances behave quite differently than older appliances even from only 10 years back. Models of the different customer classes were validated against actual recordings of load variations under voltage reduction.

Distribution network designers use a transformer load-management system to estimate and examine the historical and current loads occurring on transformers and test proposed load situations. Due to a lack of measurements at the feeders and distribution transformers, the confidence on the transformer load-management reports was low in the past. Intelligent equipment and advanced information and communication technology are now being integrated into distribution networks at an unprecedented speed. The technology advancement provides an opportunity to enhance transformer load modeling and management. In this paper, a framework using synchronized measurements obtained from different information systems via a utility's enterprise service bus is proposed to calculate the actual load profiles of distribution transformers and reflect changing system conditions. Network analyses resulting from using improved system models based on distribution state estimation allow the system operators and designers to make informed decisions concerning where new loads can be added and when new transformer capacity must be included.

In this paper, an agent-based energy-management system is proposed to facilitate power trading among microgrids with demand response and distributed storage (DRaDS). The key focus of this approach is to utilize diversity in load consumption patterns of the customers and energy availability from the distributed energy resources, distributed storage, and demand response (DR) in reducing the peak demand and minimizing the cost of electricity. An index-based incentive mechanism is also proposed to encourage customers participating in DR based on frequency and size of the participation. The proposed strategies are validated using two case studies with interconnected microgrids. Based on extensive simulation results of the system being developed using the Java Agent DEvelopment framework (JADE), it has been found that multiagent-based energy management with DRaDS is successful in reducing the system peak and cost benefit for customers with a high priority index.

This paper proposes an inference methodology that can minimize the spreading effect caused by the occurrence of a feeder fault, a feeder overload, a main transformer fault, or a main transformer overload problem in a distribution system under a heavy load condition. First, the load transfer procedure is decomposed into multisteps and multistages. Then, a new inference methodology for searching the goal nodes as well as the fail nodes based on the local minimum tree search is developed.

Control and operation of a dc microgrid, which can be operated at grid connected or island modes, are investigated in this paper. The dc microgrid consists of a wind turbine, a battery energy storage system, dc loads, and a grid-connected converter system. When the system is grid connected, active power is balanced through the grid supply during normal operation to ensure a constant dc voltage. Automatic power balancing during a grid ac fault is achieved by coordinating the battery energy storage system and the grid converter. To ensure that the system can operate under island conditions, a coordinated strategy for the battery system, wind turbine, and load management, including load shedding, are proposed. PSCAD/EMTDC simulations are presented to demonstrate the robust operation performance and to validate the proposed control system during various operating conditions, such as variations of wind power generation and load, grid ac faults, and islanding.

A load balancing transformer (LBT) has already been suggested for improving the unbalance of three-phase primary currents. Each phase includes one extra pair of coupling windings in addition to the usual primary and secondary windings. Two coupling windings, located on two different phases, are made in series, where the resulting circuit is then reversely paralleled with the secondary winding placed on the third phase. Under unbalanced conditions, the load currents are distributed between the coupling and secondary windings that are supplied through different primary phases. This paper proposes a method in order to define all possible LBTs, which paves the way to cross-examine and select the best connections of windings for the LBT. Starting from the original LBT, it is shown that the best LBT is practically connected, such as a zig-zag winding. Then, a novel idea is suggested in which a controller and some four-quadrant semiconductor switches contribute to the improvement of the performance of the LBT. These switches should control the amount of current transfer from one phase to another. A combinatorial selection problem is arranged to find the best way of switching modulation. Both simulations and experimental works (using a designed 12-kVA laboratory prototype) verify the studied examination and proposals, showing the switch-mode zig-zag LBT as having the best performance.

This paper proposes a reconfiguration methodology based on an Ant Colony Algorithm (ACA) that aims at achieving the minimum power loss and increment load balance factor of radial distribution networks with distributed generators. A 33-bus distribution system and a Tai-Power 11.4-kV distribution system were selected for optimizing the configuration and to demonstrate the effectiveness of the proposed methodology for solving the optimal switching operation of distribution systems. The simulation results have shown that lower system loss and better load balancing will be attained at a distribution system with distributed generation (DG) compared to a system without DG. Furthermore, the simulation results also satisfy and suitability reference merits of the proposal method.

This paper describes the active power and frequency-control principles of multiple distributed generators (DGs) in a microgrid. Microgrids have two operating modes: 1) a grid-connected mode and 2) an islanded mode. During islanded operation, one DG unit should share output generation power with other units in exact accordance with the load. Two different options for controlling the active power of DGs are introduced and analyzed: 1) unit output-power control (UPC) and 2) feeder flow control (FFC). Taking into account the control mode and the configuration of the DGs, we investigate power-sharing principles among multiple DGs under various system conditions: 1) load variation during grid-connected operation, 2) load variation during islanded operation, and 3) loss of mains (disconnected from the main grid). Based on the analysis, the FFC mode is advantageous to the main grid and the microgrid itself under load variation conditions. However, when the microgrid is islanded, the FFC control mode is limited by the existing droop controller. Therefore, we propose an algorithm to modify the droop constant of the FFC-mode DGs to ensure proper power sharing among DGs. The principles and the proposed algorithm are verified by PSCAD simulation.

In restructured power systems, the independent system operator (ISO) has to maintain system reliability at an acceptable level. In addition to the expected energy not supplied (EENS) as a reliability index, the total grid losses are another system operators' anxieties. Load management (LM) is an effective mean to achieve a suitable network operational condition. The optimum allocation of LM program is of great practical interest. This paper presents a multiattribute decision-making for allocation of network-driven load-management (NDLM) measures. The objectives of the optimization problem are defined as minimization of the EENS, minimization of total grid losses, and maximization of load-management capacity. It has been shown that these objectives are conflicting aspects of an NDLM program which is solved by the STEM interactive method. The IEEE reliability test system is used to demonstrate the effectiveness of the proposed methodology.

The electricity supply industry, unlike other industries, is hard pressed by the very nature of electricity. The inability to be stored economically and the requirement for instant response to consumption demands are the two major business challenges facing the electricity supply industry's service providers. The industry has, therefore, to invest in assets with capacities that may be called upon, in total or in part, only for limited periods of time during any given year. For these reasons and for other potential cost saving measures the need for consumers' continuous, ?on-line?, interaction arises. This paper presents two mathematical equations to continuously compute the capacity and energy charges with built in incentives and penalties to the consumers. The proposed equations lend themselves, to a large extent, to replace currently applied yearly averaged incentives and penalties. The application of the proposed equations would, in essence, avoid the current potential adverse impact on the economics of the electricity industry service providers, that are typically associated with the unaccounted for fluctuations of the consumption. The primary objective of the proposed equations is to entice consumers to interactively manage and/or customize, on their own, their consumption patterns to their benefits and that of the service providers.

In this paper, an extended impedance-based fault-location formulation for generalized distribution systems is presented. The majority of distribution feeders are characterized by having several laterals, nonsymmetrical lines, highly unbalanced operation, and time-varying loads. These characteristics compromise traditional fault-location methods performance. The proposed method uses only local voltages and currents as input data. The current load profile is obtained through these measurements. The formulation considers load variation effects and different fault types. Results are obtained from numerical simulations by using a real distribution system from the Electrical Energy Distribution State Company of Rio Grande do Sul (CEEE-D), Southern Brazil. Comparative results show the technique robustness with respect to fault type and traditional fault-location problems, such as fault distance, resistance, inception angle, and load variation. The formulation was implemented as embedded software and is currently used at CEEE-D's distribution operation center.

The benefit of shifting load from the peak to the off-peak period is well known from the viewpoints of demand response and deferral of capacity upgrade. However, the impact of load leveling on reducing transmission and distribution (T&D) losses is often ignored or lightly mentioned in the literature. A detailed analysis of T&D loss reduction is very specific to the considered power system and cannot be readily generalized. Therefore, the objective of this paper is to identify the key parameters and quantify the saved T&D losses as a function of the size or power of the energy storage in a set of normalized charts to help assess the benefits of energy storage as a means to level a utility load. This is a benefit of utility-scale energy storage that is not fully recognized.

Frequency recording units (FRUs), designed to measure power systems' frequency at low voltage, are affected by the connection of large loads causing sudden phase shifts and magnitude changes in the voltage. These phase shifts introduce spikes in the frequency measured by FRUs. In this paper, we propose an algorithm to detect the phase shifts and to produce an accurate frequency estimation regardless of load changes

Due to dispersed locations and various capacities, emerging Distributed Resources (DR's) have diverse and highly nonlinear impact (Contribution Factor, CF) on transmission operation constraints, thereafter on congestion management in electricity markets. From the perspective of Independent System Operators (ISO's) and DIStribution COmpanies (DISCO's), it is crucial to address such impacts in order to identify the "desirable" and "undesirable" DR's. This paper introduces a straightforward-partitioning space, ITDA, developed by a fuzzy c-means clustering approach based on accurate CF's derived from AC power flow. Case studies with 6-bus and 30-bus test systems demonstrate that the proposed technique is simple and applicable.

In this paper, the customer information in customer information systems (CIS) and information of customer and distribution transformers in outage management information systems (OMIS) in Taiwan Power Company (Taipower), are used to determine the daily load patterns of service areas, sectionalizing switches, distribution feeders, and main transformers. During system normal operation conditions, the feeder reconfiguration for load balancing among distribution feeders is obtained by the G-Nets inference mechanism to enhance the operation performance of distribution systems. For distribution contingencies, such as feeder overloading and/or short-circuit fault, the G-Nets inference mechanism with operation rules is applied to derive the optimal switching operation decision for service restoration to perform the optimal load transfer among distribution feeders after the fault has been identified and isolated. To determine the effectiveness of the proposed methodology, a practical Taiwan power distribution system with daily load patterns derived by load survey study is selected to perform the computer simulation. It is found that the G-Nets inference mechanism approach can enhance the solution process of fault restoration with proper load transfer and improve feeder load balancing for distribution systems by considering the load characteristics of the service customers.

A high power low distortion static VAr compensator was proposed by the authors in 1997. The harmonics are eliminated by incorporating a low power IGBT based controlled current auxiliary converter in conjunction with a high power GTO based converter. The current reference required by the auxiliary converter is calculated by an involved procedure which makes the overall compensation process complicated. In this paper, a simplified control scheme for the VAr compensator is proposed. The effectiveness of the proposed scheme is investigated for a case where the VAr compensator is specifically used for load compensation and also for the case where it is used as a general purpose VAr generator. Detailed simulation studies of the schemes are presented. Experimental results obtained from a laboratory prototype are provided to demonstrate the viability of the proposed method.

This paper undertakes the problem of determining the load capability of distribution networks, or, equivalently, the amount of load a feeder, specific area or circuit of a large-scale unbalanced distribution network can withstand before violating an operational constraint. A new formulation is given. A solution algorithm suitable for large-scale unbalanced distribution networks with capacitor control actions is developed and test results on a NYSEG 394-bus distribution network are included.

The multiscale transient event detector (TED) introduced by Leeb (1993) uses measurements of current and voltage made at the electric utility service entry of a building to help determine the operating schedule of all of the significant loads in the building. This paper describes a TED algorithm implemented on a custom multiprocessing machine with several inexpensive processors performing tasks in parallel. A parallel version of the serial TED algorithm is developed, and the hardware design for a multiprocessing TED is presented. The performance of the algorithm is illustrated with results from a prototype.

This paper presents a formulation for the optimal design of power distribution systems taking cold load pickup into consideration. The optimization problem is applied to an actual distribution feeder and the solution is obtained using a genetic algorithm. Being concerned about voltage drop and overloading problems, transformer load in time during cold load pickup is determined and a power flow is run for the maximum sustained part of the load. Corrective actions are suggested.

Optimal capacitor allocation determines the size, type, and location of capacitors to be installed on a radial distribution feeder that will reduce peak power and energy losses while minimizing the costs of investment and installation of the capacitor banks. Because of lack of information and experimental data about the real power and reactive power demands at the various nodes of the distribution feeder, it has been customary to assume uniform load characteristics at constant power factor for the loads along the feeder. This paper presents an optimal capacitor allocation method that uses sensitivity factors and a characterization scheme for the load demands at the various feeder nodes. Sensitivity factors are used to reduce the number of solution alternatives to be evaluated. A distribution-analyser-recorder (DAR) has been developed to collect data from the feeder. The data collected is processed into load multipliers that are used as input data for the optimization process. The proposed method has been applied to two test feeders, and simulation results appear very promising.

In response to deregulation, many UK regional electricity companies (RECs) are considering redesigning their communication to meet this and other business requirements. This paper presents a proposed communication infrastructure for South Western Electricity plc. The company services a wide variety of customers in the south west of England. The supporting technology, REC and customer benefits, together with valued added services (VAS) are addressed. The SWEB communication system consists of a microwave radio network which is used for telephony, VHF radio, and data traffic. The applications of the communication system include automated meter reading, load management, quality monitoring and billing.

Large scale system implementation of demand side management (DSM) and distribution automation (DA) intended for an electric utility network serving a large customer base (one million or larger) spread over a large geographical area requires a two-way communication system. Unlike pilot projects where testing and feasibility studies are highly structured and controlled, many of the results from the pilot tests are not readily transportable into the design of very large systems. In a multitasking environment, system data throughput, response time and communication bandwidth utilization are not only functions of communication baud rate and performance, but are also highly dependent on the system infrastructure, local geography, the customer base and the regulatory environment. In this paper, three generic communication infrastructures are analyzed and expressions for the system response time and message throughput rate are developed. The system message rate and response time requirements are discussed and their relationship to the communication infrastructure and the meter reading application throughput is analyzed. The impact of poor communication performance to the system response time and throughput is presented.

This paper introduces integrated models of distribution transformers and their loads for three-phase power flow analyses. All transformer connections can be easily included, such as single-phase, open wye, open delta and three-phase. For an existing three-phase power flow program without rigorous transformer models, only a slight modification of this program is needed to analyze distribution systems in more detail by using these proposed models. For those with rigorous transformer models, the rigorous transformer models usually make the program converge with difficulty, or even diverge. The convergence characteristics of these program can be dramatically improved if proposed integrated models are used instead of the rigorous transformer models. Moreover, these models can be easily applied by some functions of advanced distribution management systems or automatic mapping and facility management systems, such as transformer load management and feeder load management, to evaluate the individual phase loads along a feeder.

This paper describes the theoretical foundation and prototype implementation of a power system transient event detector for use in a nonintrusive load monitor (NILM). The NILM determines the operating schedule of the major electrical loads in a building from measurements made at the electric utility service entry. The transient event detector extends the applicability of the NILM to challenging commercial and industrial sites. A spectral preprocessor for use in the transient event detector is introduced first. Then, the transient event detection algorithm is developed. The performance of the algorithm is illustrated with results from a prototype event detector. >

An approach to the design of digital algorithms based on multi-function correlators applied to current and voltage samples to measure power consumption characteristics is described in this paper. Based on this approach, a new technique for algorithm design is developed. The main advantage of this approach is the determination of all power consumption characteristics by the same algorithm. The algorithm has a high accuracy and regular structure. The regular structure is a very important property for the development of specialized chips. >

This paper describes a study conducted by the Andhra Pradesh State Electricity Board (APSEB), India, and ECC, Inc., USA for the design of a distribution automation system (DAS). Distribution automation functions applicable to an Indian electric utility were identified. Communications technologies that are capable of meeting DAS requirements were evaluated, and a communications system suitable for India was defined. A benefit/cost analysis was performed; first for a basic DAS and then for additional distribution automation functions. >

A fuzzy reasoning approach is proposed for the service restoration of a distribution system. After the location of a fault has been identified and the faulted zone has been isolated, it is important for the operators to reach a proper service restoration plan in order to restore the electricity service outside the faulted zone. The operators tend to use their past experience and heuristic rules to devise such a restoration plan because it must satisfy a lot of practical needs and objectives. In addition, the operators' needs and heuristic rules are often expressed in imprecise linguistic terms. In this paper, fuzzy set notation is employed to deal with these imprecise linguistic variables and a set of fuzzy reasoning procedures are developed to implement the operators' heuristic rules. These procedures can be employed to solve the multiple-objective problem of service restoration described in imprecise linguistic variables. To demonstrate the effectiveness of the proposed fuzzy reasoning approach, service restoration on a distribution system within the service area of Taipei West District Office of Taiwan Power Company is examined. It is found that a proper restoration plan can be reached very efficiently by the proposed approach. >

Previous work by the authors (see Proc. First Internat. Forum on Appls. of Neural Nets to Power Systs., Seattle, p.205-9, 1991) has shown the viability of using artificial neural networks (ANNs) for the early detection and control of overloaded lines. This paper focuses on the development of an artificial neural network based scheme for intelligent load shedding. The load shedding scheme consists of localized detectors that consider parameters of the power system during an emergency. The research was developed using the IEEE 30 bus test system which incorporates a design of net import and export regions. The approach is implemented in the import area of the system. The purpose of the scheme is to detect overloaded lines and make intelligent decisions about where within the topology of the system the load should be dropped and how much load to shed. The proposed scheme is designed to avoid unintentional separation of the system by fast, proactive, and adaptive control. >

Service restoration following a major power failure can occasion voltage problems on weak EHV systems from the point of view of both steady-state and temporary operation. This is due to the use of long EHV transmission links between the production sites and the load centres, which can cause large amounts of reactive power to be generated, and low short-circuit level, leading to low-frequency resonance, further excited by the presence of transformers. This paper describes an approach to service restoration which should speed up the process of restoring service to customers while at the same time limiting the risks of damage to equipment.

Determining open switch positions for the minimum-loss configuration of a radial distribution system is a discrete optimization problem. This paper presents a global optimality condition of the problem and two algorithms: one is based on the uniformly distributed load model and the other on the concentrated current (or power) demand model. The derivation of the optimality condition relies on a concept of basic current profiles of feeder sections and a transformation of the optimization problem. In the first algorithm, a closed form of the minimum-loss open point of a feeder pair can be found since the loss function is piece-wise parabolic. The algorithm obtains the optimal solution when the minimum is obtained for every feeder pair. In the second algorithm, a similar procedure is performed by moving open points, one at a time, from an actual switch position to another until no further loss reduction can be achieved. The proposed algorithms can be utilized as an on-line aid to distribution system operators.