Complicated weather conditions lead to intermittent, random and volatility in photovoltaic (PV) systems, which makes PV predictions difficult. A recurrent neural network (RNN) is considered to be an effective tool for time-series data prediction. However, when the weather changes intensely, the long-term sequence of multivariate may cause gradient vanishing (exploding) during the training of RNN, leading the prediction results to local optimum. Long short-term memory (LSTM) network is the deep structure of RNN. Due to its special hidden layer unit structure, it can preserve the trend information contained in the longterm sequence, which is allowed to solve the problems of RNN and improve performance. An LSTM-based approach is applied for short-term predictions in this study based on a timescale that encompasses global horizontal irradiance (GHI) one hour in advance and one day in advance. Inaccurate forecasts usually occur on cloudy days, and the results of ANN and SVR in the literature prove this. To improve prediction accuracy on cloudy days, the clearness-index was introduced as an input data for the LSTM model and to classify the type of weather by k-means during the data processing, where cloudy days are classified as the cloudy and the mixed(partially cloudy). NN models are established to compare the accuracy of different approaches and the cross-regional study is to prove whether the method can be generalizable. From the results of hourly forecast, the R 2 coefficient of LSTM on cloudy days and mixed days is exceeding 0.9, while the R 2 of RNN is only 0.70 and 0.79 in Atlanta and Hawaii. From the results of daily forecast, All R 2 on cloudy days is about 0.85. However, the LSTM is still very effective in improving of RNN and more accurate than other models.

No abstract

Devising policies for a low carbon city requires a careful understanding of the characteristics of urban residential lifestyle and consumption. The production-based accounting approach based on top-down statistical data has a limited ability to reflect the total greenhouse gas (GHG) emissions from residential consumption. In this paper, we present a survey-based GHG emissions accounting methodology for urban residential consumption, and apply it in Xiamen City, a rapidly urbanizing coastal city in southeast China. Based on this, the main influencing factors determining residential GHG emissions at the household and community scale are identified, and the typical profiles of low, medium and high GHG emission households and communities are identified. Up to 70% of household GHG emissions are from regional and national activities that support household consumption including the supply of energy and building materials, while 17% are from urban level basic services and supplies such as sewage treatment and solid waste management, and only 13% are direct emissions from household consumption. Housing area and household size are the two main factors determining GHG emissions from residential consumption at the household scale, while average housing area and building height were the main factors at the community scale. Our results show a large disparity in GHG emissions profiles among different households, with high GHG emissions households emitting about five times more than low GHG emissions households. Emissions from high GHG emissions communities are about twice as high as from low GHG emissions communities. Our findings can contribute to better tailored and targeted policies aimed at reducing household GHG emissions, and developing low GHG emissions residential communities in China.

In order to solve the problem of insufficient control performance of various traditional control strategies in the complex environment of grid-connected inverters, the active disturbance rejection control (ADRC) strategy based on the virtual synchronous generator (VSG) is proposed. The mathematical model of a grid-connected photovoltaic inverter based on the VSG is built. The proposed control strategy provides the inverter with more disturbance attenuation and provides rotational inertia. The control strategy estimates and compensates the total disturbance and generates the reference active power and reactive power by ADRC. The control strategy converts the three-phase voltage and current outputs into positive and negative sequences on the dq reference frame. The VSG control module generates a reference voltage command and outputs it to the dual closed loop PI feedforward decoupling control. The PWM signal is finally obtained by the PI feedforward decoupling control. The ADRC strategy based on the VSG does not change the original control characteristics of the VSG; it retains the characteristics of the synchronous generator, and it also provides inertia and damping for the power grid. The simulation shows that the ADRC strategy based on the VSG applied to the inverter can attenuate disturbances. Under the unfavorable conditions of the unstable reference power, such as the unbalanced three-phase voltage and the random disturbance, the output power matches the international electricity standard. INDEX TERMS Virtual synchronous generator (VSG), active disturbance rejection control (ADRC), power control, grid-connected inverter, positive and negative separation. The associate editor coordinating the review of this manuscript and approving it for publication was Huanqing Wang.

In order to build a local electricity market (LEM), community members can trade electricity peer-to-peer (P2P) with their neighbors. This paper proposes a Hierarchical Bidding and Transaction Structure based on blockchain (HBTS). First, combined with the multi-agents, each microgrid corrects the estimated cost probability distribution of other microgrids by Bayesian theorem, making its probability closer to the accurate probability. Second, for maximize the benefits of the microgrid, this paper uses the Nash equilibrium in the Cournot model to find the optimal quotation and output of different bidding strategies for the microgrid under different power demand conditions. Then the exchange of electricity translates into an exchange of digital proof of electricity purchases and sales of electricity on the Hyperledger Fabric, ensuring the security of the transaction process and the irreparable modification of ledgers. Finally, we verify the effectiveness of the bidding strategy through experiments, and analyze the transaction process.

No abstract

No abstract

This paper proposes an original data-driven intelligent control solution to the cooperative output regulation problem of discrete-time multi-agent systems. Based on the combination of internal model principle and reinforcement learning, a distributed suboptimal controller is learned realtime via online input-state data collected from system trajectories. Rigorous theoretical analysis guarantees the convergence of the proposed algorithm and the asymptotic stability of the closed-loop system. Numerical results validate the effectiveness of the proposed control methodology.