Haoming Liu

MPPT of PV system under partial shading condition based on adaptive inertia weight particle swarm optimization algorithm

This paper adopts an adaptive inertial weight particle swarm optimization (AIWPSO) algorithm to improve the maximum power point tracking (MPPT) capability for photovoltaic (PV) system under partial shading condition. Partial shading is a common phenomenon in PV generation system, it causes imbalance and decreases for output power of PV array. Under partial shading condition, output characteristics of PV system will change and the P-V characteristic curve contains more than one peak, which makes the conventional algorithm for MPPT is difficult to track the practical MPP. Particle swarm optimization (PSO) algorithm is often used in MPPT under partial shading condition, but PSO algorithm has the disadvantages of low convergence speed and search accuracy. In this paper, AIWPSO algorithm is proposed to solve these problems. In AIWPSO algorithm, a nonlinear dynamic inertia weight factor is introduced into the PSO evolution to improve global searching ability of PSO algorithm. Simulation results for constant partial shading and rapid changing partial shading show that the proposed algorithm can avoid premature convergence effectively and has good global searching capability.

Active and reactive power coordinated control strategy of battery energy storage system in active distribution network

Large penetration of intermittent renewable energy and complex loads in Active Distribution Network (ADN) has aggravated the fluctuation of voltage and increased power loss. Battery energy storage system (BESS) is a critical device in ADN, which are used to provide active power for the system. However, by connected with the grid using converter, battery has the ability to provide reactive power for the grid without other reactive compensation devices. To exploit the potential of BESS to regulate voltage fluctuations in ADN, in this paper, an active and reactive power coordinated optimization of BESS is proposed, with the goal of realizing lowest power loss as well as minimal voltage fluctuations in ADN. Then a specific solution to this optimal problem is given based on particle swarm algorithm. The results of a case study show that, by comparison between active power control strategy and active and reactive power coordinated control strategy, this paper has confirmed that the latter with reactive output of battery can have a better effect on reducing the power loss and stabilizing voltage fluctuation.

Ultra-short term hybrid power forecasting model for photovoltaic power station with meteorological monitoring data

In order to realize the real time tracking and prediction of ultra-short term power of PV power station, according to the actual monitoring data of the photovoltaic power station, an ultra-short term power forecasting model based on autoregressive moving average (ARIMA) and support vector regression (SVR) is established in this paper. The ARIMA model has good adaptability to the linear trend. The results show that the method has good tracking and prediction affection in the prediction of irradiance and temperature. It can be a reliable input to regression models. The SVR model realizes the highly nonlinear fitting of data mapping. The effectiveness of the ARIMA-SVR model is verified by the prediction of the ultra-short term output power of PV power plants under four different weather conditions.

Deployment strategy for battery energy storage system in distribution network based on voltage violation regulation

In order to improve the situation of voltage violation caused by the grid-connection of photovoltaic (PV) system in a distribution network, a bi-level programming model is proposed for battery energy storage system (BESS) deployment. The objective function of inner level programming is to minimize voltage violation, with the power of PV and BESS as the variables. The objective function of outer level programming is to minimize the comprehensive function originated from inner layer programming and all the BESS operating parameters, with the capacity and rated power of BESS as the variables. The differential evolution (DE) algorithm is applied to solve the model. Based on distribution network operation scenarios with photovoltaic generation under multiple alternative output modes, the simulation results of IEEE 33-bus system prove that the deployment strategy of BESS proposed in this paper is well adapted to voltage violation regulation invariable distribution network operation scenarios. It contributes to regulating voltage violation in distribution network, as well as to improve the utilization of PV systems.

Individual pitch control strategy on alleviating periodic load fluctuations

Considering the periodic load fluctuations of wind turbine caused by wind shear and tower shadow, the individual pitch control strategy is proposed in this paper. According to the wind model of wind shear and tower shadow, the individual pitch control based on azimuth angle and load feedback is discussed, the blade flapwise load feedback is superimposed on the basis of azimuth angle weight coefficient distribution to fine correct the pitch angle of each blade. The strategy is applied on a 1.5 MW wind turbine, the simulation results on Matlab platform show that the proposed strategy can significantly alleviate blade flapwise load fluctuations, which means the strategy holds good potential to reduce rotor unbalance loads and mitigate fatigue loads of wind turbine.

Parameter error identification method for multi-doubtful parameters of power grid

Parameter error affects the quality of state estimation and the application of other senior software in energy management system (EMS). When there are multiple bad data of remote measurement and parameter errors, it is important to ensure the effectiveness of state estimation. This paper proposes an effective and practical parameter error identification method for essential data of power grid. Firstly, percentage residuals of all measuring data are calculated to obtain the set of suspicious measuring points and further the set of suspicious branches. Then, correlation indices of suspicious branches are calculated and sorted to get a descending sequence. The suspicious parameter corresponding to a suspicious branch is adjusted based on the Newton downhill method. Case study on a simple 5-node network and a practical power grid shows that the method purposed can identify the network parameter error and the measuring data error simultaneously, thus improving the accuracy of state estimation on essential data of power grid operation.