Li Qingsheng

Research on automatic testing based model validation framework for distribution automation system

With the development of energy and information technologies, smart distribution system has become an emerging area for the development of smart grid. Self-healing and active distribution system are the hot research topics of smart distribution network. The Distribution Automation System (DAS) is the fundamental system for integration and realization of such technologies, which integrates computer science, modern communication technologies and distribution automation technologies. The reliability of DAS has an profound influence on the application of emerging smart distribution network. After rigorous testing, DAS itself has been certified and authorized for higher security, reliability. As the model of distribution system evolves frequently, the correctness of the network model will seriously affect the applicability and reliability of distribution automation system. The network model management and validation are the key technologies to improve the reliability of DAS and applicability of advanced applications for smart distribution system. At present, there are relative researches on model validation technologies, such as basic grammar examination, model name and topological relation validation. Although these approaches can improve the automation and reliability level of distribution network model management to a certain extent. However, there are still some shortcomings: lack of automatic verification ways for interactive validation based on both diagram and model for DAS, which is heavily dependent on the system operators and there may be remaining potential security risks; lack of automatic testing tools, which increases the workload for the systematic validation of network model, and reduces efficiency and reliability at the same time. Compared with traditional manual test approach, automatic testing technology utilities emerging test approach to realize fully automatic test or semi automatic testing, which could effective reduce manual intervention, accelerated testing cycle and enhance testing efficiency by reduction of the repetitive and redundant testing process. Besides, with carefully designed test cases, automatic testing technology could guarantee the reliability of complex control system. Automatic test has been applied in the development of modern complex automation system. This paper presents a model validation framework based on automatic testing technology for distribution system. Firstly, an automatic testing framework for network model validation is proposed. Then the procedure and test approach are developed based on the automatic testing framework. The approach proposed in this paper has been applied in a practical distribution automation system. Actual examples are presented to prove that the proposed approach can effectively improve the efficiency of model management and validation for distribution system and greatly reduce the potential model related risks which could otherwise downgrade the reliability and correctness of DAS. The proposed approach could be applied and provide foundation for the further development of smart distribution system.

Model predictive control algorithm of flexible load in consideration of both electricity cost and comfort

This paper proposes a novel model predictive control algorithm based on intelligent terminal, which considers both the economic benefit and the customer comfort. In this algorithm, novel models for air-conditioner and water-heater are created, which obviously simplify the algorithm. In the new models, the multivariable thermodynamic process is separated into several independent univariate sub-processes, which can be solved easily. In addition, the algorithm obtains the model parameters by parametric regression from history data. By this way, the optimal results are independent of the outdoor temperature and other confounding factors. Thus, the complexity of the algorithm is significantly reduced. More, considering both electricity cost and customer satisfaction, a dual-objective optimization algorithm is implemented to get the control strategy, which is solved by particle swarm optimization (PSO) and linear programming. At last, a simulation result on MATLAB shows the economic benefits and load shifting effect.