Yingchen Liao

Distribution network reconfiguration considering the random character of wind power generation

A chance constrained programming formulation for distribution reconfiguration with wind power generator (WPG) is proposed that aims at minimum power loss and increment voltage quality. A scenario analysis method is applied to describe the random output of WPG through the scenario probability and scenario output. A multiple objective particle algorithm (MOPSO) is employed to solve the multi-objective discrete nonlinear optimization problem. The dedicated particle encoding with the mesh information of the distribution network can effectively avoid producing a large amount of invalid solutions. With MOSPO, it is possible to obtain the optimum solution set of each objective while helping the operator to choose the most appropriate plan for reconfiguration. Application of the model and MOPSO algorithm to the 69 distribution network has verified their feasibility and correctness.

Optimal bidding strategies with risks for LSEs in competitive electricity markets

In competitive electricity markets which are open for both generation and distribution sides, load serving entities (LSEs) are required to participate in the market competition with risks, since their profits usually reduced because of the shortage of power supply, and their purchasing prices were fluctuant with uncertainties due to the unforeseen bidding. This paper develops the optimal bidding strategies with risks for LSEs to participate in competitive electricity markets, in which stepwise bidding functions and pay-as-bid settlement protocols are utilized. A normal probability distribution function is used to describe the bidding behaviors of rivals, and the issue of constructing optimal bidding strategies for LSEs is formulated as a multi-objective stochastic optimization model, then Monte-Carlo approach is applied to get the corresponding solutions. The proposed bidding strategies have been tested on an electricity market with 3 power companies (Gencos) and 4 LSEs, and relevant analyses have been presented.

Short-term load forecasting based on time series reconstruction and support vector regression

The load curve presents certain randomness for reasons such as human social activities, the load of electric vehicle charging and discharging and so on, which covers up the regularity of load sequence. This paper proposes an approach which can restore the original feature of the loads. In this approach, firstly, the bad data is excluded; Secondly, the characteristics of the time series are extracted by the time series reconstruction; At last, the loads are forecasted by support vector regression and the forecasting sequences are taken as training samples of the next prediction. After repeat predictions, the final series are obtained then restored, and precise prediction result is obtained. This paper uses the load data of a city in Jiang Su. Compared with traditional support vector regression method, this approach ensures a higher prediction accuracy.

Combined approach for short-term wind power prediction: A case study of the east coast of China

Power systems with high wind power experience increased variability and uncertainty. Therefore accurate forecast technique is essential to cope with the uncertainty in day-ahead electricity market. Statistical forecast model is considered as a powerful technique for wind-power forecast, however, the forecast accuracy significantly drops as the forecast horizon grows. Focusing on wind speed persistence, a hybrid multi-step-ahead (HMS) method is proposed, which composed of direct multi-step-ahead prediction (DMS) approach and indirect multi-step-ahead prediction (IMS) approach in time-series forecasting. To validate the effectiveness of the proposed method, one-month period of real data, which is collected from five operating wind farms in the east coast of China, is used for test. The persistence model (PM) is used as a benchmark with the new forecast method. Test results show that the proposed method is accurate and effective.

The variable weight combination load forecasting based on grey model and semi-parametric Regression Model

Grey model using dimensional information update technology always contains the latest information from sample data, it guarantees the accuracy of mid-long term load forecasting. Semi-parametric regression model which combines the advantages of parametric model and nonparametric model, it can fully reflect the complexity and uncertainties of the load change. This paper put forward an improved method which combines grey model with semi-parametric regression model by time-varying weight for load forecasting, the proposed method would make full use of data information and consider its inherent regularity completely, which makes prediction more realistic. At last, a comparison of the error has been made between the single model and the combination model. The test example results show that this method has higher precision.

Source-grid-load interaction research in power system

The large-scale access of flexible equipment and the in-depth development of electricity market form a two-way transmission of energy and information among source side, grid side and load side in future power system, which leads to an extensive source-grid-load interaction. A well-established source-grid-load interaction research will provide a theoretical basis to control the future power system and to guide its development. In this paper, the interactive system is further analyzed, the mathematical models of interaction excitation and interaction cost are proposed, the power balance in power system is expressed by probabilistic indexes, and eventually an optimal interaction framework is formed, a typical source-grid-load interaction process analysis shows that this framework can be used to find the optimal interaction strategy.

Fast service restoration of distribution system with distributed generations

A fast, effective service restoration method is designed for distribution systems with the high penetration of distributed generation (DG), which is formulated as a constrained multi-objectives optimization problem. This paper regards the nodes connected with DG as the power source nodes which are encoded by the depth encoding technology, then searches the islands which include DGs, and establishes islanding scheme after the failure of distribution system. On this basis, the island transition load indexes are established, and are used to evaluate the reserve capacity of distribution networks with DGs. Those island transition load indexes combined with the index of the reserve capacity of tie switches are optimizated to indicate an optimal path to restore service for non-fault areas according to heuristic rules. One target is that DG and tie switches are fully used to restore all the loads and minimize the number of required switch operations as few as possible. Finally, the example of IEEE-69 indicates that this algorithm not only fully uses DGs to improve the reliability of power supply, but also greatly narrows the search space and reduces the search time of optimization. This method is mainly applied to service restoration in the distribution network with DGs for real-time operation.

Service restoration study of distribution system with distributed generators based on particle swarm optimization

After the occurrence of fault in a distribution network, the loads get disconnected and are left unsupplied. This paper studies service restoration in distribution network with distributed generation (DG), Proposed a particle swarm algorithm based on two-dimensional depth-coded to solve the service restoration in the event of a large-scale blackout in a distribution network with DG. Pre-fault system using two-dimensional depth-coded, the range of blackout can be obtained quickly. Using depth-first search for Blackout area, renumbered the nodes which in eare of nlackout using two-dimensional depth-coded. According to load the importance to sort in level, and taking into account the adjust ability of distribution genertion, Solved the maximum supply range of DG. Contact switch to be deal with as a distributed generater according its reserve capacity. Finally, particle swarm algorithm for solved the maximum supply range. Particle has a two-dimensional, one expressed as the number of layers to service restoration, another expressed as the serial number of load in its level. To search for the maximum supply range though update the particle position. Numerical results of an IEEE69-bus example. It means the proposed method can solve the service restoration of a distribution system with DG after widespread blackout effectively and with fast convergence speed and strong stability.

Determining reserve requirements in systems with significant stochastic generation capacity using copulas

The large-scale integration of stochastic generation requires additional operating reserves to cope with the uncertainty in power system operation, due to the large forecast error with existing methodologies. Previous research shows that allocating power reserves dynamically according to conditional distribution of power forecast error would benefit the generation scheduling in day-ahead electricity market. Therefore, this paper applies the empirical copula model to estimate the dynamic operating reserves, via formulating the joint probability distribution of forecast errors for multiple wind farms and photovoltaic (PV) plants. Simulation results demonstrate that the proposed probabilistic method can facilitate more accurate scheduling of dynamic reserves for stochastic generation, in the electricity market with large shares of wind and solar power.

Optimal layout of energy consumption monitoring points based on the evaluation of energy efficiency

Industrial users are big energy consumers, and most of them install monitoring devices on all transmission lines and load branches of the network during energy consumption monitoring. As a result, there is no direction in energy consumption monitoring, and there is a lot of the phenomenon of repeated monitoring which results in a great waste of resources. This paper presents the concept of fluctuation coefficient of energy efficiency. By comparing the size of fluctuation coefficient of each production line, we can realize the main monitoring of some production lines in industrial users. Using binary quantum particle swarm optimization (BQPSO) to further optimize the arrangement of energy consumption monitoring points, we can further reduce the number of energy monitoring points in the premise of making the whole network observed. Doing the simulation of a textile enterprise, the number of energy monitoring points can be reduced from the original 118 to 50 through optimization. The results show that the optimization method can significantly decrease the number of energy monitoring points in the premise of making the whole network observed and can reduce the cost of energy consumption monitoring for industrial users effectively.