Victor Levi

A new decomposition based method for optimal expansion planning of large transmission networks

A new method is presented for long-range transmission network expansion planning, based on the decomposition principle. The overall transmission expansion planning task is divided into two problems, the first one dealing with investments, and the second with operations. The investment problem is specified as the minimum cost problem of network programming, leading to the decomposition into the model of minimum load curtailment and the proposed model of the (local) marginal network. The operation problem is solved by applying the Monte Carlo simulation, with suitable control strategies and additional reliability constraints. The final result is the software package, verified on test examples, as well as on the real transmission network of the power pool of eastern Yugoslavia. >

Isolated Chargers for EVs Incorporating Six-Phase Machines

This paper considers two isolated solutions for fast charging of electric vehicles (EVs). The isolation is located on the grid side (off board), whereas the rest of the charging apparatus is placed on board the EV, and it entirely consists of the existing power electronics components that would be otherwise used only for propulsion. Thus, substantial savings on space, weight, and cost are achieved. The considered configurations fully incorporate either a symmetrical or an asymmetrical six-phase machine, as well as a six-phase inverter, into the charging process. Due to the nature of the connections, torque production is avoided during the charging/vehicle-to-grid (V2G) modes of operation. Thus, the machines do not have to be mechanically locked, and their rotors naturally stay at standstill. Control schemes for both configurations are elaborated, and theoretical results are validated by experiments for the two configurations in both charging and V2G modes.

Optimal Demand Response Scheduling With Real-Time Thermal Ratings of Overhead Lines for Improved Network Reliability

This paper proposes a probabilistic framework for optimal demand response scheduling in the day-ahead planning of transmission networks. Optimal load reduction plans are determined from network security requirements, physical characteristics of various customer types, and by recognizing two types of reductions, voluntary and involuntary. Ranking of both load reduction categories is based on their values and expected outage durations, while sizing takes into account the inherent probabilistic components. The optimal schedule of load recovery is then found by optimizing the customers’ position in the joint energy and reserve market, while considering several operational and demand response constraints. The developed methodology is incorporated in the sequential Monte Carlo simulation procedure and tested on several IEEE networks. Here, the overhead lines are modeled with the aid of either static-seasonal or real-time thermal ratings. Wind generating units are also connected to the network in order to model wind uncertainty. The results show that the proposed demand response scheduling improves both reliability and economic indices, particularly when emergency energy prices drive the load recovery.

Integrated methodology for transmission and reactive power planning

This paper presents the comprehensive methodology for solving the transmission reinforcement and reactive power planning problems. The static approach is applied to both problems and they are separated into corresponding investment and operation subproblems. The solution of the transmission reinforcement investment problem is obtained through the further decomposition to two independent subproblems, while the operation problem is dealt with by the Monte Carlo simulation. Both the investment and operation problems of the reactive power planning are studied with the optimum power flow models which are extended to encompass the voltage stability.

Modelling of Ageing Distribution Cable for Replacement Planning

The paper proposes a new reliability model of ageing distribution cable and simulation methodology for prioritization of cable replacement. The proposed reliability model combines IEC cable thermal model, Arrhenius aging model and Weibull probability distribution. Here, Weibull scale parameter is defined through the Arrhenius aging model, in which conductor temperature is governed by the cable thermal model. A novel estimation method for the calculation of Arrhenius model parameters is proposed. The cable reliability model is then incorporated into the developed sequential Monte Carlo simulation procedure and reliability indicators and loss-of-lives are calculated. An industry based cable ranking scheme is further proposed to prioritize individual cable due for replacement. The whole methodology is tested on several medium voltage distribution networks.

Cable Replacement Considering Optimal Wind Integration and Network Reconfiguration

This paper proposes a methodology for optimal cable replacement based on utility financial benefits that are calculated by considering the entire network in the planning period. The overall problem is set within a sequential Monte Carlo simulation framework and uses novel reliability modeling of ageing distribution cable using the “IEC-Arrhenius-Weibull” model. Optimal cable replacement schemes are developed by considering the connection of new wind sources and network reconfiguration to reduce cable ageing and defer its replacement. The core concepts of the proposed methodology are two new mixed-integer non-linear optimization models. The first optimizes the connection of new wind sources by minimizing the connection and reinforcement costs, as well as minimizing the cost of the cable thermal-loss-of-lives in the planning period. In the operations stage, the network is optimally reconfigured to minimize the cost of losses, the thermal-loss-of-life of cable, and reliability. Both optimization models are applicable to radially operated medium voltage networks that are backfed from several normally open points. The final outputs are proposed cable ranking lists for replacement and reliability and cost metrics.

Integrated Planning of Distribution Networks Considering Utility Planning Concepts

This paper presents a methodology for integrated planning of real-life medium-voltage networks based on the utility planning concepts. The research is motivated by the need to develop a methodology that would line-up with utility day-to-day businesses and could be applied in real-life. Its core is a two-stage optimization process, where the first stage solves the static investment optimization and the second stage considers operational problem. A probabilistic decision tree approach is proposed for the solution of the entire problem to consider uncertainties in the planning period. The overall formulation is given first, which is followed by details of the investment model and outlines of the proposed operation planning. The novelty of the investment problem, which determines optimal network reinforcements, is explicit modeling of network security constraints of radially operated networks, whilst considering different operating regimes. Additional novel features include modeling of real-life supply restoration rules through network reconfiguration and optimal placement of new switching devices, as well as consideration of “customer flows” on the network. Connection of new distributed generation and demand centers and construction of circuits on new corridors are also included. Two investment models are formulated as mixed-integer nonlinear optimization problems, tested on several MV networks and compared with established methods. The proposed operational problem is solved in two stages, quality-of-supply and operation cost optimization. Computational aspects are also presented.

Maximization of Wind Energy Utilization Through Corrective Scheduling and FACTS Deployment

The paper proposes a probabilistic methodology for minimizing wind spillage and maximizing capacity of the deployed wind generation, whilst improving system reliability. Capacities of the connected wind units are initially determined by using a method developed by the industry. A probabilistic approach is applied for the day-ahead planning to find maximum deployable wind sources so that the prescribed wind spillage is not exceeded. This is done using the optimum power flow, where wind spillages are prioritised with the probabilistic “cost coefficients.” Further improvement of wind energy utilization is achieved by installing FACTS devices and making use of real-time thermal ratings. Two ranking lists are developed to prioritize location of SVCs and TCSCs, and they are then combined into a unified method for best FACTS placement. The entire methodology is realized in two sequential Monte Carlo procedures, and the probabilistic results are compared with the state enumeration ones. Results show improved wind utilization, network reliability, and economic aspects.

Expansion planning of medium voltage distribution networks

The paper proposes a new methodology for optimal planning of medium voltage distribution networks. The overall model is presented as a two-stage optimization model, where the first stage deals with the investment part and the second stage focuses on the operation part. The novelty of the proposed model is demonstrated through the explicit incorporation of network security constraints in the mathematical model, in line with the UK planning standards, while taking into consideration different operating regimes with changing load and generation profiles, integration of new distributed generation units, construction of circuits on new corridors and optimal network reconfiguration. The highly complex mixed-integer nonlinear optimization model is applied to a few practical medium-voltage test systems of up to 120 nodes using a dedicated commercial software package. Some of the characteristics results are examined and highlighted in this paper. Additionally, sensitivity analyses are done to determine critical parameters affecting the development and performance of medium voltage distribution networks.