Search

A knowledge-based system (KBS) which provides an integrated operational planning environment for distribution systems is presented. Four functions are incorporated: remedial control; protective device coordination; loss reduction; and service restoration. A real-time KBS development environment is used, which provides a high level user interface and acceptable processing speed. A systematic method is used to optimize the object-oriented data structure. The KBS has been tested with data from an electric utility. >

Establishing a smart power grid utilizing advanced agent information technology has raised much attention recently. In this paper we propose a method that is based on coordination and negotiation among a set of distributed local power committees who could together reach a consensus of “global” optimal power restoration solution in cope with local power failures after a fault is detected in a power grid. The optimality criteria can be specified by a human power manager in terms of actual operation costs such as the minimum number of switch actions. We show how local distributed agent committees whose members are essentially switch agents and power generator agents are connected locally as neighbors and can find a global optimal solution to the power restoration problem by localizing their decisions in the local committee and by communicating and negotiating via neighbors of their committee members.

Localization technology is one of the most important research areas in wireless sensor networks. This paper proposes a set of RF modulation methods to improve the accuracy of localization by utilizing the Receive Signal Strength Indication (RSSI). Two different RF power adjust methods are proposed in this paper. Genetic Algorithm (GA) and Artificial Bee Colony Algorithm (ABC) are used to optimize the location estimation in the indoor environment. Comparisons of the performance of these two methods with two different RF power adjustments are also conducted.

The novel feature extraction method for nonintrusive load monitoring (NILM) systems was proposed in this paper. In order to monitoring the status of each load, a sensor is installed for each load traditionally. The status of the load is transmitted to the main controller such that the status of each load can be monitored. On the other hand, the NILM is able to detect the status of loads by analyzing the current signals that is picked-up by a current sensor installed at the main electrical panel. Traditional NILM methods used real power, reactive power, harmonic contents of power signatures and transient energy to determine the status of appliances. These methods are very complex and require a lot of computation. In this paper, a novel method that integrates artificial intelligent recognition technique and load current acquisition method for NILM is proposed. The proposed method uses timedomain information. This approach is different from traditional NILM methods. The proposed method is able to detect the energization and de-energization of loads by applying back-propagation neural networks (BP-ANNs). The overall correct rate for this method is above 98.75%. This result shows that the proposed method is able to determine the operation status of loads with proper robustness.

CRAFT (customer restoration and fault testing) is a rule-based expert system capable of locating the faulted section of a power transmission line based on the status of circuit breakers an automatic switches. CRAFT has been operating online in the Puget Sound Power and Light control center since July 1988. The online operational experience is reported. Many of the developed expert systems will be maintained or modified by utility engineers. This study proposes a methodology for maintaining rule-based expert systems. The following capabilities are incorporated: detecting the relations of conflict, redundancy, etc., to maintain consistency of the rule base, tracing the reasoning process to understand the problem solving approach, and reducing the size of test cases after changes are made to the rules. The proposed method has been implemented for CRAFT. >

The configuration of a power distribution system can be altered by changing the status of switches on the feeders. When a fault occurred on a distribution system feeder, the faulted section is isolated by opening the switches adjacent to the fault. When the fault is isolated, the system is reconfigured so the outage areas due to the fault can be re-energized. In this paper, a BDI (Believe-Desire-Intension)-based multi-agent architecture is proposed. The agents can response to the changes of the environment and take necessary conversation with other agents to reach the final goal, supplying services to all de-energized customers. These agents act independently but co-operatively. Decision tree is used in the proposed agent architecture for making restoration plan. Decision trees are built during the restoration process based on the fault location and load conditions through communications between agents. By using the decision tree, a more robust decision can be achieved by agents. Simulation results show that the proposed BDI-Based agent architecture is capable of proposing adequate restoration plan as expected. communication bandwidth can be reduced. Furthermore, no information synchronization is needed since the information stored in each agent are local information. The proposed intelligent agent architecture not only can perform group restoration, a better zone restoration plan can also be identified by utilizing the restoration decision tree. Since agents are running in parallel, the zone restoration process can be performed simultaneously to improve the performance. In the proposed architecture, agents are installed in switches and feeders. The tasks of switch agents include monitoring and recording the operation conditions, responding and requesting information during an occurrence of a fault, etc. Feeder agents monitor the total connected loads and response to the requests from switch agents. Information exchange is important for the agent-based feeder service restoration since the operation condition of feeders may change. In order to make a correct restoration plan, acquiring the operation condition is very critical. In the proposed intelligent agent architecture, every agent is capable of initiating restoration process autonomously when the agent detects an occurrence of a fault. A restoration leader agent will be identified first by exchanging the information after a fault has occurred. This agent will initiate the necessary communication and try to find an adequate restoration plan based on the acquired topology and loading conditions. The leader agent is also responsible of making restoration plan based on received information.

Neural networks play an important role in artificial intelligence application domains. In most of applications, neural networks are often implemented in software form. Although the software implementation of neural networks provides flexibility, the computation speed is limited due to the sequential machine architecture. In most applications using artificial neural networks, the learning procedure is carried off-line. A large amount of mathematic operations are needed when learning task of neural networks is performed. The software implementation of neural network systems can only work well using high performance computers. The learning performance is not adequate when it is implemented on embedded systems. Following the development of modern semiconductor technologies, people attempt to realize the neural networks by hardware in order to improve the performance. Designs utilizing special architectures and parameters to achieve the performance were proposed in the past. This paper proposes a high efficiency and generic neural network hardware architecture. The architecture uses the toroidal series multiple data stream to process the back propagation neural network operations, which has the full function of recall and learning capabilities. Users can adjust the number of processor elements (PEs) in the system based on the requirement of the applications by setting the values in registers. Since the proposed system is developed in hardware, it can be integrated into embedded systems easily. The experimental results show that the system can reach much higher performance by using fewer logical elements while maintaining flexibility

The implementation and operation of an expert system in a control center environment is discussed. CRAFT (customer restoration and fault testing) is a rule-based expert system which is capable of locating the faulted section of a transmission line equipped with automatic switches and circuit breakers. In this study, the capabilities of CRAFT are extended by adding two new knowledge bases with rules to determine the status of unsupervised automatic switches and to detect the presence and location of a manual switch with an incorrect status, following a fault. An approach to appending a separate expert system computer to the SCADA system is developed. Software and hardware interfaces are designed and implemented. The completed implementation is operating online. The implementation represents a step toward practical application of expert systems for real time operation of power systems. >

The objective of feeder reconfiguration on distribution system during normal operation is to find the best switch operation plan for reducing power losses and enhancing service reliability. By changing on/off status of sectionalizing-switches and tie-switches on a distribution system, feeder reconfiguration can be achieved. Also, the configured system must satisfy operation conditions as well as restrictions. Since the reconfiguration is done by changing the status of switches, it can be categorized as a discrete combinational optimization problem. Particle swarm optimization (PSO) is the one of the methods that can be used to solve optimal problems. Typical PSO is designed for continuous functions optimization, it is not designed for discrete functions optimization. Therefore, the operators of PSO algorithm must be reviewed and redefined to fit the application field of distribution feeder reconfiguration. This paper proposed a method which modifies the operators of PSO's formula based on the characters of both status of switches and shift operator to construct the binary coding particle swarm optimization for feeder reconfiguration. The test results show that the proposed method can apply to feeder reconfiguration effectively.

Using soft computing for solving distribution reconfiguration problems were studied for many years. Genetic algorithm (GA) is one of the most popular technologies in the soft computing area for solving distribution system problems. However, due to the radial structure of power distribution system, traditional GAs may encounter some difficulties when searching for the optimal solution. Evolutionary programming (EP) was also being used to solve some distribution system problems, for example, loss minimization, service restoration, capacitor placement and many others. Hence, the EP is applied in this paper in order to overcome the weakness of traditional GAs (Fudou et. al, (1997); Miranda et al., (1994); Nara et al., (2003); Ying-Tung Hsiao, (2004), Back et al., (2004); Ying-Tung Hsiao and Ching-Yang Chien, 2000). One of the differences between GA and EP is that the weighting of chromosomes is used for selection operator. The weighting calculation of this paper is based on the characteristics of feeder losses and load balancing on distribution feeders. The results show that the proposed EP with adapted weight calculation performs better than traditional GAs

This work compares several genetic algorithm reproduction methods for distribution system loss reduction and load balancing problems. Traditional GA crossover procedures use one-point or two-point crossover methods to form chromosomes for the next generation. Although, the chromosomes can be easily obtained, a repair algorithm is needed in case of invalid chromosomes are formed. In this paper, asexual reproduction method is proposed. Several different reproduction methods are compared. From the results, the asexual reproduction methods require less generations to reach the optimal solution than gamogenesis.

In distribution automation, restoration process is done by switching actions of sectionalizing switches and tie switches on the system. The determination of the state of these switches is mainly logical reasoning with less mathematical process. Thus, knowledge-based systems are ideal tools to accomplish the task. In this research, CLIPS is used to develop the proposed system by taking advantage of its object-oriented feature. By applying object-oriented technique, the actual status of a distribution system component can be reflected properly by updating the attributes of all the objects. By referring to the empirical rules of previous work, the distribution system restoration procedure can be divided into three stages: group restoration, zone restoration and load transfer. The unique feature of the proposed expert system is that the variation of load currents is considered. By doing so, the restoration strategies can be more accurate, practical and effective. A simulation on a small distribution system is performed, the results are compared with the same system which has constant load currents.

Applications of information technology to power systems have gained more attention. The concepts of computer network and open systems are affecting the development of power system operation and analysis tools. Among various concepts in the information technology, two major issues, data exchange and distributed system, are the keys which influence the tool development.

The authors discuss the implementation and operation of an expert system in a control center environment. Attention is given to CRAFT (Customer Restoration And Fault Testing), a rule-based expert system which is capable of locating the faulted section of a power transmission line equipped with automatic switches and circuit breakers. In this study, the capabilities of CRAFT are extended by adding two new knowledge bases with rules to determine the status of unsupervised automatic switches and to detect the presence and location of a manual switch with an incorrect status following a fault. An approach involving the appending of a separate expert system computer to the existing SCADA system is developed. Software and hardware interfaces are designed and implemented. The completed implementation is operating online. >

A hybrid expert system which utilizes model-based and rule-based knowledge is developed for processing alarms generated from the SCADA computer of the power supply system of the rapid transit system of Taipei. The heuristic rules are used to perform message synthesis while the model-based knowledge is used to find the relationship among alarms. The traditional rule based alarm processing expert system has the disadvantages of incompleteness of the knowledge base. This problem can be resolved by using a model-based approach. On the other hand, message synthesis cannot be modeled easily. Therefore, heuristic knowledge is applied. In this paper, a hybrid expert system is developed to take advantages of both types of knowledge bases.

A new method for estimating fault sections in a power substation is proposed. In this paper, the knowledge representation and inference procedures based on the cause-effect networks (CE-Nets) and Boolean rule matrix transformation techniques are presented. A CE-Net is a graphic modeling tool for representing the causality between faults and actions of protective devices. By transforming the established CE-Nets into matrix forms, the possible fault sections can be estimated through simple matrix operations. The method is superior to existing production systems in the inference speed and the process of implementation. Since the proposed reasoning methods require only simple matrix operations in a parallel manner, it is well suitable for on-line applications. The proposed method has been tested on a typical Taiwan Power Company (Taipower) secondary substation. From the experimental results, it is found that the proposed method can estimate fault sections very efficiently.

This paper discusses a computer algorithm which can be used to detect the conditions that will separate the power system into two or more islands. In order to provide an early warning mechanism for the dispatchers in the control center such that some actions can be taken before an islanding condition occurs, the proposed algorithm converts the power system topology into an undirected graph first. Based on the graph theory, the algorithm will indicate that outages occur on one or more of the transmission lines will lead to an islanding situation. The Tai-Power Company 345 kV and 161 kV systems will be used to validate the algorithm and evaluate the performance.

A new method for estimating fault sections in power substations is proposed. In this paper the knowledge representation and inference procedures based on the cause-effect networks (CE-Nets) and rule matrix transformation techniques are presented. The matrix based inference procedures are proposed to estimate the possible fault sections through simple matrix operations by transforming the established CE-Nets into matrix forms. The method is superior to existing production systems in the inference speed and the process of implementation. By testing on a typical Taiwan Power Company's (Taipower) secondary substation, the experimental results show that the effectiveness of the proposed method even for the fault domains involving multiple faults and failure operations of protective devices. Moreover, it is easy to implement and transplant into different substations.