A Distributed Self-Healing Method for Active Distribution Systems

In order to overcome the disadvantages of a traditional gas distribution system of G-M type pulse tube cryocoolers, a novel active gas distribution system with a middle stage reservoir is proposed. Different from the traditional intake and exhaust processes, the refrigeration cycle with the active gas distribution system consists of an intake process from the middle stage reservoir, an intake process from the compressor, an exhaust process to the middle stage reservoir and an exhaust process to the compressor. Theoretical analysis shows that the active distribution system may significantly reduce the power consumption of the compressor due to the reduction of irreversible losses during the intake and exhaust processes. The self-made pulse tube cooler with the active gas distribution system was tested. Experimental results show that the active gas distribution system decreases the power consumption by 27% and entropy generation by 37% when the proportion of opening time of the middle stage reservoir is 30%. The cryocooler reaches a lower refrigeration temperature with minor decrease of cooling power. Besides, the active gas distribution system may increase the life time for valves.

Considering that the conventional synchronous splitting (MSS) method for power flow calculation of coupled transmission and active distribution (TDC) systems has limitations in efficiency and convergence due to the penetration of distributed generations (DGs) and loop operation, this paper proposes an asynchronous forward-backward-splitting power flow algorithm of TDC systems. In this algorithm, the transmission system operator (TSO) acts as a coordinator and processes forward steps while the distribution system operators (DSOs) act as agents and process backward steps. Two first-in-first-out queues are introduced to realize an asynchronous mechanism, and thus, the TSO does not have to wait for the slowest DSO to complete its power flow calculation and power injection update before the next iteration can proceed. This asynchronous mechanism has a higher efficiency compared with a synchronous one when communication delays exist. In addition, the conventional fixed-point iteration in the MSS method is replaced with the Krasnosel’skii-Mann iteration. The proposed algorithm with appropriate settings will achieve better convergence, especially when active distribution networks are considered, where volt-walt-control-typed and volt-var-control-typed DGs are penetrated. The linear convergence of the proposed algorithm is proved under several assumptions and the parameter tuning issue is discussed with numerical experiments.

An essential aspect that contributes to the resilience of radial electrical distribution systems (EDS) is the deployment of efficient and fast service restoration methods. In the presence of a permanent fault, the service restoration minimizes the amount of de-energized consumers by reducing the area that needs to be disconnected to isolate the faulty section of the network. To do so, the service restoration algorithm can make use of automated switching devices and dispatchable energy resources allocated throughout the network to transfer loads among feeders, considering the system's operational constraints. Depending on the size of the system and the complexity of the telecommunication infrastructure, the application of centralized restoration systems may neither be practical nor flexible. Thus, in this paper, a distributed service restoration method is proposed, using the alternating direction method of multipliers (ADMM) algorithm to solve the optimization problem. As such, the original service restoration model is distributed among the nodes and branches of the system that exchange limited information with neighboring nodes and branches. Radiality constraint is also distributed by the proposed ADMM algorithm. A 44-node test system was used to conduct case studies. Results demonstrate that the proposed method efficiently solves the service restoration in active EDS without using a centralized Intelligence.

For an active distribution system (ADS) that integrates high levels of distributed generators (DGs), the control dimension of such a wide and dynamic set of resources would become overwhelming. This study proposes a novel pinning group consensus (PGC)-based distributed coordination control to simplify ADS control by using the virtual clusters building block concept. A given ADS is rethought as coupled small virtual clusters, including virtual microgrid clusters and virtual power plant clusters; accordingly, the ADS can be coordinated by controlling a small number of selected pinning agents instead of a huge number of DGs. The predefining of the PGC values for these pinning agents, which comprehensively considers both clusters features and DG capacities, is the most distinguishing work for the proposed control scheme and can lead to effective global coordination in a distributed manner. Simulation cases under normal/disturbed/emergency conditions verify the effectiveness and advantages of the proposed scheme.

Volt/Var control problem of distribution systems has been extensively investigated in literature. Many control models and algorithms have been proposed to achieve better system quality, security, reliability, efficiency, loadability, and cost effectiveness. Early distribution systems are built based on centralized power generation which is named passive distribution system (PDS), where power flow is unidirectional. Nowadays, the topology of the distribution system allows for bidirectional power flow which is named active distribution system (ADS) due to the presence of active resources, such as distributed generations (DGs). The complexity of controlling each system depends on the topology and size of the network, as well as the control devices used. However, in general there are mainly two main control strategies used to control power networks: centralized and decentralized. This paper provides a review for both control strategies in the distribution system based on Volt/Var control techniques. It introduces the most commonly used techniques and algorithms in the literature for passive and active distribution systems. Moreover, it provides the reader with a comprehensive review on the common optimization techniques and the different objective functions used in terms of loss minimization, voltage deviation, and minimum control variable operation.

This paper proposes a multi-objective benefit function for operation of active distribution systems considering demand response program (DRP) and energy storage system (ESS). In the active distribution system, active network management (ANM) is applied so that the distribution system equipment is controlled in real-time status based on the real-time measurements of system parameters (voltages and currents). The multi-objective optimization problem is solved using ε-constraint method, and a fuzzy satisfying approach has been employed to select the best compromise solution. Two different objective functions are considered as follows: benefit maximization of distribution company (DisCo); benefit maximization of distributed generation owner (DGO). To increase the benefits and efficient implementation of distributed generation (DG), DGO has installed battery as energy storage system (ESS) in parallel with DG unit. Consequently, DGO decides for the battery charging/discharging. DisCo has the ability to exchange energy with the upstream network and DGO. Also, DisCo focuses to study the effect of demand response program (DRP) on total benefit function and consequently its influence on the load profile has been discussed. This model is successfully applied to a 33-bus radial distribution network.

This paper proposes a cyber-physical active distribution system (CPADS) planning method, which considers the economic aspect and overall performance in the context of distributed generations (DGs) interconnection. First, the composition of CPADS is introduced for analyzing the effect of cyber system on the planning goals and constraints, based on which the dependence of the physical system on the cyber system is investigated. Then, the framework of CPADS planning is proposed by comparing it with active distribution system (ADS). Second, the planning is modeled to search for the appropriate DGs capacities, communication mode, and voltage control frequency with the constraints including performance, construction investment, and out-of-limit voltage probability, aiming at minimizing the construction and operation cost of cyber-physical systems and maximizing the DG penetration level. After that, the bi-layer multi-objective programming without considering the random failure of cyber system is adopted to derive the optimal solution set. Monte Carlo Simulation is then performed to solve for the out-of-limit voltage probability due to cyber failures, based on which the final scheme is obtained. Simulations are conducted on the modified IEEE 33-node distribution system. The results verify the effectiveness of the proposed method.

In this paper, a distributed dynamic state estimation (DDSE) based on particle filter (PF) for real-time monitoring of the active distribution system is proposed. The active distribution system is partitioned into several areas with local estimators. Local estimations based on PF are performed with the collected measurement information from phasor measurement units (PMUs) within their own area and forecasted loads, after which the posterior estimates of boundary buses are exchanged between neighboring areas. Re-estimations will be performed until the differences of the estimates of boundary buses from different estimators don’t exceed the given limits. In addition, the influences of distributed generators (DGs) are also considered, the outputs of which are approximated by Gaussian mixture model (GMM). Simulation results based on 119-bus distribution system demonstrate the effectiveness of the proposed method.

The active nature of next generation distribution systems requires a highly accurate real-time monitoring for state estimation and situational awareness. This paper considers the problem of robust placement of a limited number of voltage magnitude meters and phasor measurement units for state estimation in an active distribution system comprising topological reconfigurations. The trace of the inverse of the Fisher information matrix is chosen as criterion for the estimation accuracy. For meter placement in a single configuration distribution system, the greedy approach provides a near-optimal solution as well as a theoretical approximation guarantee. However, for meter placement in an active distribution system, a robust approach is required to optimize the worst case estimation accuracy among all possible configurations of the system. In this paper, a simple robust algorithm, called submodular saturation algorithm, for meter placement in active distribution systems is proposed. Numerical results on three different active distribution systems show that the submodular saturation algorithm outperforms the greedy algorithm and the genetic algorithm in most cases, and provides competitive results in other cases.

A recloser is a protection device used to mitigate long power interruptions during transient faults in passive distribution systems. Passive distribution systems are becoming active because of the integration of distributed generation. This generation introduces a reverse power flow and consequently a reverse fault current. The pick-up and operation time of the recloser can be affected because of the change in the fault current level. It is essential to study and compare the recloser's pick-up activity and operation time before setting the recloser in an active distribution system. This study, therefore, develops a virtual single-shot reclosing scheme and its performance tested in a passive and active distribution system. Matlab/SimulinkiSimscape power systems environment is used to simulate cases. Simulations are conducted for four case studies to identify issues in the performance of the recloser when there is distributed generation. The results show that the fault contribution of the distributed generation causes false tripping, protection blinding, and recloser-fuse misco-ordination.

Risk-averse decision under worst-case continuous and discrete uncertainties in transmission system with the support of active distribution systems

Protection blinding is a challenging issue in renewables‐penetrated distribution grids and refers to a situation where a circuit breaker may not trip due to fault current contribution from distributed generation. This research addresses how the distributed generation location and capacity impact the operation of the circuit breaker in terms of the response time of the circuit breakers. The relative electrical distances of the faults and distributed generation to the circuit breakers are considered. The impact of distributed generation capacity considering the fault location is characterized using a new index called the heterogeneity index. The electrical distance between distributed generations and circuit breakers and the electrical distance between fault and circuit breaker is considered by a second new index called the electrical distance ratio. Data analysis on simulation results shows that these indices capture the phenomena of protection blinding caused by distributed generation. Results show that a higher distributed generation penetration and faults that are electrically further away from a circuit breaker show severe cases of protection blinding captured by the indices. Furthermore, it is demonstrated how these indices can identify the worst impacted locations in the distribution grid. A key result is that protection blinding does not necessarily occur solely due to the presence of distributed generation between a circuit breaker and a fault, but is dependent on factors such as distributed generation location in the distribution grid, fault level, fault level distribution across the generation units and fault location.

Distributed generation (DG) brought significant changes to the planning and operation of distribution systems (DS). The presence of DG, allied with advanced technologies of measurement and control related to Smart Grid concepts, converts the DS into active networks, subject to bidirectional power flows. As a consequence, aiming to improve the efficiency and reliability of the power supply, active networks begin to operate more often as closed-loop circuits or even as fully meshed topologies.This context points to the need to update traditional computational methods related to DS analysis to incorporate the consequences of technological advances. At the same time, such updated methods must still consider intrinsic invariable characteristics of the distribution networks, like the low-voltage levels and the low X/R ratio of DS lines.This chapter presents a fast-decoupled power flow (FDPF) formulation combined with the complex per unit (cpu) normalization technique that extends the well-known efficiency of FDPF algorithm to active distribution system analysis. The formulation bases of power flow calculation and cpu technique are fully described. Simulation results with DS under distinct operational conditions are presented to demonstrate the effectiveness and versatility of the cpu-based FDPF approach.

Active power dispatch and reactive power optimization problems are usually handled separately in active distribution systems, aiming at minimizing the total generation cost or transmission losses. However, the separate optimization cannot achieve a global optimum scheme in distribution system operations. Moreover, the significant relationship between the active power and reactive power may pose great challenges to distribution system operations due to the uncertain nature of load demands and intermittent renewable energy resources. In this paper, using the branch flow model-based relaxed optimal power flow, we formulate a robust coordinated optimization problem for active and reactive powers as a mixed integer second-order cone (SOC) programming problem. Furthermore, in order to address the uncertainties, a two-stage robust optimization model is proposed to coordinate the on load tap changer ratios, reactive power compensators, and charge–discharge power of energy storage system to find a robust optimal solution. Then the column-and-constraint generation algorithm is applied to solve the proposed robust two-stage optimization model. In the relaxed optimal power flow, a stricter cut is added to speed up the computation process of the SOC relaxation in order to guarantee the exactness for representative cases, such as those with high penetration of distributed energy resources. Numerical results based on the 33-bus and 69-bus systems verify the effectiveness of the proposed method.

Distributed dynamic state-input estimation for power networks of Microgrids and active distribution systems with unknown inputs