Electricity load prediction is the primary basis on which power-related departments to make logical and effective generation plans and scientific scheduling plans for the most effective power utilization. The perpetual evolution of deep learning has recommended advanced and innovative concepts for short-term load prediction. Taking into consideration the time and nonlinear characteristics of power system load data and further considering the impact of historical and future information on the current state, this paper proposes a Seq2seq short-term load prediction model based on a long short-term memory network (LSTM). Firstly, the periodic fluctuation characteristics of users’ load data are analyzed, establishing a correlation of the load data so as to determine the model’s order in the time series. Secondly, the specifications of the Seq2seq model are given preference and a coalescence of the Residual mechanism (Residual) and the two Attention mechanisms (Attention) is developed. Then, comparing the predictive performance of the model under different types of Attention mechanism, this paper finally adopts the Seq2seq short-term load prediction model of Residual LSTM and the Bahdanau Attention mechanism. Eventually, the prediction model obtains better results when merging the actual power system load data of a certain place. In order to validate the developed model, the Seq2seq was compared with recurrent neural network (RNN), LSTM, and gated recurrent unit (GRU) algorithms. Last but not least, the performance indices were calculated. when training and testing the model with power system load data, it was noted that the root mean square error (RMSE) of Seq2seq was decreased by 6.61%, 16.95%, and 7.80% compared with RNN, LSTM, and GRU, respectively. In addition, a supplementary case study was carried out using data for a small power system considering different weather conditions and user behaviors in order to confirm the applicability and stability of the proposed model. The Seq2seq model for short-term load prediction can be reported to demonstrate superiority in all areas, exhibiting better prediction and stable performance.
With the integration of highly permeable renewable energy to the grid at different levels (transmission, distribution and grid-connected), the volatility on both sides (source side and load side) leading to bidirectional power flow in the power grid complicates the control mechanism. In order to ensure the real-time power balance, energy exchange, higher energy utilization efficiency and stability maintenance in the electric power system, this paper proposes an integrated application of blockchain technology on energy routers at transmission and distribution networks with increased renewable energy penetration. This paper focuses on the safe and stable operation of a highly penetrated renewable energy grid-connected power system and its operation. It also demonstrates a blockchain-based negotiation model with weakly centralized scenarios for “source-network-load” collaborative scheduling operations; secondly, the QoS (quality of service) index of energy flow control and energy router node doubly-fed stability control model were designed. Further, it also introduces the MOPSO (multi-objective particle swarm optimization) algorithm for power output optimization of multienergy power generation; Thirdly, based on the blockchain underlying architecture and load prediction value constraints, this paper puts forward the optimization mechanism and control flow of autonomous energy coordination of b2u (bottom-up) between router nodes of transmission and distribution network based on blockchain.
This paper puts forward a Bi-LSTM attention mechanism model based on voltage phasor for electric power system transient stability assessment. The Bi-LSTM attention mechanism is used to map the relationship between voltage phasor and power system transient stability, and by establishing a sample matrix of transient stability information in the initial stage of the perturbation level, the extracted features are more robust, effectively reducing the false and missing samples, thus improving the generalization ability and evaluation performance of the model. Furthermore, by adjusting the network structure parameters of the best evaluation indicator. The mapping model between input features and transient stability is established to further reduce false positives and sample loss, and to improve the accuracy of network model evaluation. The improved model combines Bi-LSTM feature extraction layer and attention mechanism to form a hybrid model for a transient stability classification model, and the IEEE-39 bus New England test system is used to verify the accuracy of the model, and the wide-area noise is introduced into the generated data to evaluate the robustness of the system. Finally, the method is used to realize the transient stability evaluation of the electric power system based on voltage phasor data, and the validity of the proposed model is authenticated by comparative analysis of the proposed hybrid model with another deep learning models.
Increasing interconnections to fulfill the rising electrical demand and wide application of power electronics devices has led to an increase in transient stability criterion. This leads to the requirement of high ended computational features for Transient Stability Assessment (TSA). On failing to assess of transient stability may cause severe failures as blackouts and some other limitations in the power system. The deployment of different sensors, PMUs, and other measuring devices has increased for its advanced features and performance capabilities with time synchronism. The conventional methods of TSA are unable to process huge data from these devices. Therefore, requires deep learning methods for data mining and feature extraction for transient stability assessment. This paper proposes the hybrid CNNLSTM model for feature extraction of voltage phasor for transient stability assessment. CNN-LSTM hybrid model improves accuracy mapping the non-linear relationship between input and output data. This model possesses the features of both CNN and LSTM, therefore, improves the TSA ability with enhanced performance. The proposed hybrid model is applied to the IEEE 39-bus New England Power System to authenticate its accuracy. And also, a comparison between the proposed hybrid model and the other one is included demonstrating the efficacy of the proposed model.
In order to better realize the classified management of the massive user electricity data and do a better job for peak shifting and averting of user side, this paper presents a hybrid model of deep belief network and extreme learning machine (DBN-ELM) for the analysis of electricity consumption behavior of users. Firstly, the total load curve is calculated. Then, the load curve of resident users is compared with the total load curve to make the classification rules of user electricity consumption behavior. Furthermore, the DBN-ELM hierarchy is optimized, and by traversing the number of hidden nodes, a reasonable number of hidden layer nodes are selected considering the simulation accuracy of different nodes. Finally, load data from Irish users are used as data sources, through comparative experiments with ELM and DBN networks, the results show that the proposed DBN-ELM algorithm for electricity consumption behavior analysis can extract high-level features better from the underlying features, and can better analyze the user's electrical consumption behavior, to improve the effect of peak shifting and averting work.
With the rapid development of the power industry and Internet of Things technologies and their industries, society's dependence on electricity and power supply reliability are higher. The increasing number and types of access devices makes the power grid change its behavior dramatically making it more complex. The specification and requirements for safe operation of the grid has increased. In order to cope with the challenges of the future power system, the security management and control architecture of ubiquitous electric power internet of things (UEP-IoT) based on Edge-Cloud Computing Collaboration Technology (ECCC) is proposed around the national power grid "Three-type and Two-network" world-class energy Internet enterprise construction requirements. The architecture is committed for solving the current security protection, information interaction, data security and offsite backup of the power system through edge cloud collaboration. By building UEP-IoT, the grid will be safer to operate, leaner in management, more accurate in investment, and better in service.