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 be an effective tool for time-series data prediction. However, when the changes intensely, long-term sequence of multivariate may cause gradient vanishing (exploding) during training RNN, leading prediction results local optimum. Long short-term memory (LSTM) deep structure RNN. Due its special hidden...

Devising policies for a low carbon city requires careful understanding of the characteristics urban residential lifestyle and consumption. The production-based accounting approach based on top-down statistical data has limited ability to reflect total greenhouse gas (GHG) emissions from In this paper, we present survey-based GHG methodology consumption, apply it in Xiamen City, rapidly urbanizing coastal southeast China. Based this, main influencing factors determining at household community...

The short-term forecasting of photovoltaic (PV) power generation ensures the scheduling and dispatching electrical power, helps design a PV-integrated energy management system, enhances security grid operation. However, due to randomness solar energy, output PV system will fluctuate, which affect safe operation grid. To solve this problem, high-precision hybrid prediction model based on variational quantum circuit (VQC) long memory (LSTM) network is developed predict irradiance 1 hour in...

The open-circuit (OC) fault is one of the most prevalent failures in two-level inverter used traction systems. Reliable diagnosis method can improve security inverter. During process, thresholds are very important and accuracy depends on thresholds. However, inaccurate may lead to false or missed alarms. To reduce dependence thresholds, a threshold-free OC proposed this paper, which based finite control set-model predictive (FCS-MPC). According analysis inverter, estimated values cost...

In order to build a local electricity market (LEM), community members can trade peer-to-peer (P2P) with their neighbors. This paper proposes Hierarchical Bidding and Transaction Structure based on blockchain (HBTS). First, combined the multi-agents, each microgrid corrects estimated cost probability distribution of other microgrids by Bayesian theorem, making its closer accurate probability. Second, for maximize benefits microgrid, this uses Nash equilibrium in Cournot model find optimal...

In order to solve the problem of insufficient control performance various traditional strategies in complex environment grid-connected inverters, active disturbance rejection (ADRC) strategy based on virtual synchronous generator (VSG) is proposed. The mathematical model a photovoltaic inverter VSG built. proposed provides with more attenuation and rotational inertia. estimates compensates total generates reference power reactive by ADRC. converts three-phase voltage current outputs into...

This article proposes an adaptive, optimal, data-driven control approach based on reinforcement learning and adaptive dynamic programming to the three-phase grid-connected inverter employed in virtual synchronous generators (VSGs). takes into account unknown system dynamics different grid conditions, including balanced/unbalanced grids, voltage drop/sag, weak grids. The proposed method is value iteration, which does not rely initial admissible policy for learning. Considering premise that...

Energy storage plays a significant role in improving the stability of distributed energy, power quality and peak regulation micro-grid system, which is great significance to sustainable development energy. In grid-connected mode, energy mainly used reduce operating costs micro-grid. Real-time price arbitrage an important source revenue. It feasible design strategies using Q-learning algorithm. Due overestimation Q learning algorithm, this paper proposes strategy method based on Double-Q...

Current studies on microplastic pollution mainly focus marine systems. However, few have investigated microplastics in an urban lake. This research intends to use lake (Songshan Lake) as example explore the characteristics of and principal component well heat map analysis discuss relationships between different shapes microplastics. According this study, average abundance surface water sediments Songshan Lake were, respectively, 2.29 ± 0.98 items/m3 244 121 items/kg; thin films were major...

The traction rectifier plays a key role in high-speed trains. Unexpected failure often occurs the sensors of rectifier, which may affect control performance electric and even cause serious deterioration to A sensor fault diagnosis method is presented this paper, considering three kinds common types. It can not only locate fault, but also identify Based on influences faults, thresholds be calculated quantitatively. No additional hardware required. First, model established, estimator built....

Solar irradiance data include temporal information and geospatial information, so solar prediction can be regarded as a spatiotemporal sequence problem. However, at present, most of the research is based on time series models, studies spatial-temporal models are relatively few. Therefore, it necessary to integrate construct model for research. In this paper, graph convolutional network (GCN) long short-term memory (LSTM) was established prediction. model, radiation observatories were modeled...

Decomposition of the signal on orthogonal or nonorthogonal basis space is traditional method for fault feature extraction in field inverter diagnosis. These analysis methods make result decomposition not sparse and they are self-adaptive. In recent years, representation has received considerable attention processing because can overcome shortcomings by decomposing an over-complete dictionary instead basis. This paper proposes a combination support vector machine (SVM) diagnosis neutral point...

Aiming at improving the quality of output current when inverter is connected to grid, this paper proposes a control strategy neutral point clamped (NPC) grid-connected based on radial basis function neural network (RBFNN) multi-step predictive control. Firstly, model for NPC constructed by αβ coordinate transformation. Then future values are predicted each voltage vectors, using RBFNN prediction model. After obtained, cost f calculated vectors and optimal vector which minimizes selected. The...

The toxicity of nanomaterials to microorganisms is related their dose and environmental factors. aim this study was investigate the shifts in microbial community structure metabolic profiles evaluate factors a laboratory scale intertidal wetland system exposed zinc oxide nanoparticles (ZnO NPs). Microbial assemblages were determined using 16S rRNA high-throughput sequencing. Community-level physiological Biolog-ECO technology. Results showed Proteobacteria predominant (42.6%-55.8%) phylum...

Although the output of a photovoltaic power generation system is significantly positively correlated with solar irradiance, latter variable intermittent, random, and volatile. Volatility in irradiance particularly marked when weather conditions are complex, so, this factor has proved to be difficult predict. A neural network (NN)-based approach applied for short-term predictions study based on timescale that encompasses amount each hour throughout next day. Thus, backpropagation NN (BPNN),...

It is difficult for a single energy storage to meet both power and requirements in the island micro-grid because of randomness wind solar irradiation. A reasonable way use hybrid micro-grid. For management optimization control systems, there are various problems with traditional methods, such as large computational complexity dynamic programming. Q-learning has recently been applied optimal systems. Due limitations algorithm state space, this article uses Double deep (DQN) design strategy an...

Constructing accurate spatiotemporal correlations is a challenging task in joint prediction of multiple photovoltaic sites. Some advanced algorithms for incorporating other surrounding site information have been proposed, such as graph neural network-based methods, which are usually based on static or dynamic graphs to build spatial dependencies between However, the possibility simultaneous existence not considered. This paper establishes model hybrid network. In this model, we apply...