Research On Power System Economic Dispatch For Wind Power Consumption

As a renewable energy source,wind resource has attracted much attention from the academia circles due to its pollution-free and low-cost features.However,due to the randomness and fluctuation of wind power,large-scale grid connection will bring huge instability to the system,which will cause wind curtailment in the case of light and load shedding in the case of heavy.In recent years,the characteristics of peak storage and valley filling of electric energy storage systems and the electric-thermal coupling of electric boilers have made them widely used in the fields of promoting wind power consumption.In addition,in order to better accommodate wind power,the dispatching model of power systems containing wind power has become more and more complex,and higher requirements have been placed on the optimization algorithm.Therefore,how to establish a reasonable dispatching model to promote wind power consumption and how to quickly and accurately solve the dispatching model have become hot research topics in this field.Based on this,the research content of this article is as follows:(1)Aiming at the problem of dynamic economic dispatch of power system,a modified Anti-Predatory Particle Swarm Algorithm is proposed.The algorithm adds a social avoidance inertia weight to the original anti-predator particle swarm velocity update formula with evasion behavior,whose size is determined by the distance between the global worst particle and other particles,and the distance is normalized to use the Sigmoid function.The characteristics and linear decreasing method control the inertia weight in the reasonable range,which improves the ability of particles to avoid the worst solution.Then,according to the characteristics of the acceleration coefficient which can adjust the local and global searching ability of particles,a non-linear strategy is proposed to improve the search ability of the algorithm.Finally,the improved algorithm is used to solve the benchmark function and the economic dispatching model,which verifies the effectiveness and superiority of the proposed algorithm.(2)In order to solve the problems caused by wind power forecast errors,a dynamic wind curtailment penalty factor based on wind power forecast error estimation is proposed.Firstly,several parameters related to wind power prediction error are studied,and the correlation coefficients between each parameter and prediction error are analyzed.The parameters with better correlation coefficient are used as the estimation index of wind power prediction error,and calculates the weightvalue of each indicator by analytic hierarchy process,and obtains the estimated value of the wind power forecast error as the basis for adjusting the dynamic wind curtailment penalty factor.Then,the dynamic penalty factor of wind abandonment is introduced into the scheduling model,and the penalty factor of wind abandonment is adjusted in real time by the estimated prediction error value to reduce the impact of wind power prediction error on the power system.Finally,an example analysis verifies the accuracy of the forecast error estimation method and the rationality of the scheduling model.(3)To cope with the problems caused by the fluctuation characteristics of different frequency bands of wind power,a multi time scale scheduling model of integrated electric heating system based on variable mode decomposition(VMD)is proposed.First,In the day-ahead dispatch model,VMD is used to divide the day-ahead forecast value of wind power into three components: long term,medium term and short term.The long term fluctuation component of the first level wind power is used to draw up the unit's start-up and shutdown,the initial output of the unit and the electric boiler.In the intra-day dispatch model,VMD is used to divide the intra-day forecast value of wind power into three layers,and the long-term fluctuation component of wind power on the first layer is used to correct the initial output of units and electric boilers.In the real-time dispatch model,the real-time prediction value of wind power is divided into three layers.The long-term fluctuation component of the first layer is used to revise the output of the unit and the electric boiler.The medium-term fluctuation component of the second layer is suitable for the response speed of energy storage device(storage battery)to determine the charging and discharging plan of the energy storage device.The third layer wind power short-term component is suitable for the power storage device(super capacitor)to make its charge and discharge plan.Finally,the rationality of the scheduling model is verified by an example.There are 27 figures,18 tables and 94 references in this thesis.