Search

Highly accurate wind speed prediction is one of the effective ways to improve wind energy utilization. Therefore, this study proposes a novel double‐layer neural network (D‐NN) framework based on feedforward neural networks. First, a feedforward neural network with an extremely short training time is used as the basic learner to ensure that the D‐NN framework has a short training and optimization time. Then, the double‐layer neural network automatically segments and adjusts the length of the training set using an improved error correction method, thus improving the model's ability to capture sequence features and enhancing the model's generalization performance in a non‐directly‐connected superposition manner. Finally, a genetic algorithm is used to automatically optimize the hyperparameters of each sub‐neural network in the D‐NN and the optimal training set length of each sub‐neural network, resulting in the final optimized prediction model. The accuracy and reliability of the prediction models designed based on the D‐NN framework are verified by comparing several different models. The results of the five different prediction evaluation metrics show that the D‐NN framework can greatly improve the accuracy of wind speed prediction. [ABSTRACT FROM AUTHOR]

In recent years, with the promotion of policies and technologies, a large number of distributed generations are connected to the urban distribution network in various forms. At the same time, the flexible DC transmission technology has become mature and gradually applied to the distribution network. In this condition, the AC/DC hybrid distribution networks with flexible regulation ability will be the trend of urban distribution network. This paper accurately considers the loss of voltage source converter, and proposes a unified iterative power flow method for AC/DC hybrid distribution network, which takes the voltage source converter control modes into account. Because of the uncertainties of the distributed generations and loads, a random fuzzy power flow model is proposed, in which cumulant method is applied in the random stage. In the fuzzy stage, the fuzzy simulation technology is applied. Finally, an improved PG&E 69‐Bus system is taken as an example to verify the effectiveness of the unified iterative method and the random fuzzy power flow calculation, and put forward some suggestions for the operation of distribution network with distributed generations. [ABSTRACT FROM AUTHOR]

The stochastic economic dispatch problem of power system with multiple wind farms and pumped‐storage hydro stations is formulated as a specific stochastic dynamic programming (DP) model, i.e. stochastic storage model, it is impossible to obtain an accurate solution due to the curse of dimensionality. Based on the approximate DP (ADP) method, the stochastic storage model can be transformed into a series of mixed‐integer linear programming (MILP) models by describing the approximate value functions (AVFs) as convex piecewise linear functions in post‐decision states. The AVFs are first initialised using the results of the deterministic model under a forecast scenario of wind farm output and then trained by scanning stochastic sampling scenarios consecutively with the successive projective approximation routine algorithm. To obtain a near‐optimal day‐ahead dispatch scheme, the forecast scenario is substituted into the MILP models expressed by the trained AVFs and is solved forward through each time interval. The network constraints are incorporated by the while‐loop detection of critical lines. Test results on an actual provincial power system and the modified IEEE 39‐bus system, including the comparison among the ADP, DP, scenario‐based and chance‐constrained programming methods, demonstrate the feasibility and efficiency of the proposed model and algorithm. [ABSTRACT FROM AUTHOR]

This study focuses on future distribution networks featuring the integration of extra‐high distributed energy resources (DERs). In these distribution networks, the extra‐high DER penetration and the randomness and volatility of DER outputs increase the likelihood of voltage problems. Hence, it is necessary to design efficient solutions for DER output setpoints to regulate the voltages of distribution networks. However, the existing online distributed algorithms were designed based on different linearised technologies for power flow equations, which will cause imprecise results under extra‐high DER integration. In this study, the authors propose an increment‐correction method to improve the linearised power flow model. To obtain an efficient distributed manner, they introduced an indirect function into the end‐customer (the DER owner)'s objective function, stimulating them to provide voltage‐regulating service. In particular, two price incentive signals contained in the indirect function were transmitted from the distribution system operator to drive the DER owner to regulate output increments, which is suitable for cases where the DER belongs to different end‐customers. The proposed method can not only protect the private information of end‐customers but also enhance the robustness of the distribution network for the DER integration level. Numerical test results demonstrate the merits of the proposed method. [ABSTRACT FROM AUTHOR]

Calculation of the static voltage stability margin (SVSM) of a power system with high wind-power penetration requires consideration of the uncertain fluctuation of wind farms' (WFs') output. Using interval numbers to describe the uncertain WF output fluctuation and based on the interval optimisation theory, a new SVSM interval calculation method is proposed. The method establishes two bi-level optimal power flow (OPF) models to calculate the SVSM interval, including a min–min model for the upper bound and a max–min model for the lower bound. Through the dual theory of convex programming, the max–min model is transformed into a max–max model, which is solved by transforming it into a one-layer optimisation model. Using correlation angles to describe the correlation of different WF output intervals and using linear transformation to decorrelate them into independent intervals, two bi-level OPF models of calculating the SVSM interval considering different WFs' correlation are established. Take the IEEE-39 bus system and an actual 964-bus provincial power grid as examples, compared with the Monte-Carlo method and the continuous power flow and affine interval integrated method, the results indicate the proposed method obtains more accurate SVSM interval and has better calculation efficiency. [ABSTRACT FROM AUTHOR]

Calculation of the static voltage stability margin (SVSM) of a power system with high wind‐power penetration requires consideration of the uncertain fluctuation of wind farms' (WFs') output. Using interval numbers to describe the uncertain WF output fluctuation and based on the interval optimisation theory, a new SVSM interval calculation method is proposed. The method establishes two bi‐level optimal power flow (OPF) models to calculate the SVSM interval, including a min–min model for the upper bound and a max–min model for the lower bound. Through the dual theory of convex programming, the max–min model is transformed into a max–max model, which is solved by transforming it into a one‐layer optimisation model. Using correlation angles to describe the correlation of different WF output intervals and using linear transformation to decorrelate them into independent intervals, two bi‐level OPF models of calculating the SVSM interval considering different WFs' correlation are established. Take the IEEE‐39 bus system and an actual 964‐bus provincial power grid as examples, compared with the Monte‐Carlo method and the continuous power flow and affine interval integrated method, the results indicate the proposed method obtains more accurate SVSM interval and has better calculation efficiency. [ABSTRACT FROM AUTHOR]

This study focuses on the real‐time operation of a microgrid (MG). A novel approximate dynamic programming based spatiotemporal decomposition approach is developed to incorporate efficient management of distributed energy storage systems into MG real‐time operation while considering uncertainties in renewable generation. The original dynamic energy management problem is decomposed into single‐period and single‐unit sub‐problems, and the value functions are used to describe the interaction among the sub‐problems. A two‐stage procedure is further designed for the real‐time decisions of those sub‐problems. In the first stage, empirical data is utilised offline to approximate the value functions. Then in the second stage, each sub‐problem can make immediate and independent decision in both temporal and spatial dimensions to mitigate adverse effects of intermittent renewable generation in a MG. No central operator intervention is required, and the near optimal decisions can be obtained at a very fast speed. Case studies based on a six‐bus MG and an actual island MG are conducted to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Calculation of the static voltage stability margin (SVSM) of power system with high wind‐power penetration must consider the uncertain fluctuation of power output from wind farms (WFs). On the basis of a continuous power flow (CPF) and improved affine interval (IAI) algorithm that considers quadratic terms, a new method to calculate the SVSM interval of a power system that considers the uncertain fluctuation intervals of WF output is proposed. Considering the high penetration of wind power, the power variation of the active load, and wind‐power fluctuation are assigned by all conventional generators, except the swing generator in the CPF calculation model. In the proposed IAI algorithm, the non‐linear second‐order sensitivities are considered to obtain a more accurate SVSM interval. With the correlation of different WF output intervals described by relative angles, as well as the independent intervals obtained by decorrelation of the correlated WF output intervals, the SVSM interval is calculated by the CPF and IAI algorithms. Take the IEEE 39‐bus system and an actual 964‐bus provincial power grid as examples, and compared with Monte Carlo method, the results show that the SVSM interval calculated by the proposed method is more accurate, and the computation time is significantly reduced. [ABSTRACT FROM AUTHOR]

Large‐scale wind farms connected to the power grid have brought great challenges to power‐system dispatch. In this study, an improved two‐stage compensation stochastic optimisation algorithm based on recursive dynamic regression is proposed to solve the day‐ahead dynamic economic‐dispatching problem considering the high‐dimensional correlation of multiple wind farms. First, a copula function is used to describe the correlation of high‐dimensional wind farms. Second, a two‐stage compensation stochastic‐optimisation algorithm is proposed to convert the dynamic economic‐dispatching model to a two‐stage model with mutual iteration by decoupling the conventional and stochastic variables. In this decoupled model, the calculation of the expected compensation cost is critical and usually limited by the dimension of the correlated wind farms, which leads to inefficient convergence and long computation times. To solve those problems, a recursive dynamic multivariable linear regression method based on global least squares is proposed to improve the two‐stage stochastic optimisation algorithm. This improved two‐stage compensation algorithm overcomes the dimensional disaster of traditional stochastic optimisation methods and can solve the dynamic economic dispatching problem considering the high‐dimensional correlation of multiple wind farms. Finally, the practicability and efficiency of the proposed algorithm are verified by the examples of an IEEE118 system and an actual provincial system. [ABSTRACT FROM AUTHOR]

No study in the literature considers both randomness and fuzziness simultaneously, which actually coexist as the penetration of renewable energy in power system increases. In order to handle these two kinds of uncertain features simultaneously, a novel random fuzzy power flow (RFPF) calculation method for a distribution network based on random fuzzy theory is presented here. Firstly, the random fuzzy models of wind and photovoltaic (PV) generation, and loads are set up for the first time according to their features of randomness and fuzziness. Then, a two‐fold random fuzzy simulation is conducted to obtain the results of the RFPF calculations; the random simulation stage is based on the 2m + 1 scheme of the point estimate method. Finally, the proposed method is applied to two test systems. The results show that the proposed method is feasible and effective in identifying important areas in the power system affected by distribution generation and loads with these two uncertainties. [ABSTRACT FROM AUTHOR]

Multidisciplinary collaborative optimisation (MCO) is an effective theory to solve the design optimisation problems of complex systems. Here, power‐system day‐ahead dynamic economic dispatch with integrated wind power is studied. The MCO method is introduced to decouple the scenarios and collaboratively optimise the multiple scenarios to deal with the uncertainty of wind‐power generation. Based on this, the dynamic economic‐dispatch problem is divided into a system‐level model and multiple subdisciplinary optimisation models for the forecasting and error scenarios. A dynamic relaxation algorithm is then introduced to solve the system‐level optimisation model. The decoupled subdisciplinary models for the error scenarios are solved by a grid‐computation tool in parallel, which greatly improves computational efficiency. Finally, the established model and its corresponding solution method are applied to a 10‐machine, 39‐bus test system. It is shown that the proposed MCO‐based dynamic economic dispatching method performs much better in high‐dimensional scenarios, which are the inherent limitations of the traditional centralised multi‐scenario method. [ABSTRACT FROM AUTHOR]

The operation mode of microgrids influences the dispatching management of distribution system in the power market. The study uses game theory to study dispatching strategies between multi‐microgrid in the distribution system. For this purpose, a two‐level game model of multi‐microgrid dispatching in the electricity market is proposed. Firstly, the upper level of the model researches the answer of a basic question for a multi‐microgrid distribution system that what the boundary line between non‐cooperation mode and coalition mode is. Secondly, by the lower level of the model, multi‐microgrid decides the operation mode ultimately when it is uncertain in the upper level of model. The upper level is a non‐cooperative price game between multi‐microgrid and distribution system while the lower level is a cooperative trading loss cost game. In order to optimise multi‐microgrid dispatching, an algorithm is proposed to allow microgrids merge or split self‐adaptively based on NSGA‐II. Simulation result shows that the proposed algorithm can find the Nash equilibrium of the upper level of the model and the optimal operation mode for multi‐microgrid in the two‐level game, which yields a reduction of 52.7% in coalition mode compared to non‐cooperation mode. [ABSTRACT FROM AUTHOR]