Search

Power load forecasting plays an important role in power systems, and the accuracy of load forecasting is of vital importance to power system planning as well as economic efficiency. Power load data are nonsmooth, nonlinear time-series and “noisy” data. Traditional load forecasting has low accuracy and curves not fitting the load variation. It is not well predicted by a single forecasting model. In this paper, we propose a novel model based on the combination of data mining and deep learning to improve the prediction accuracy. First, data preprocessing is performed. Second, identification and correction of anomalous data, normalization of continuous sequences, and one-hot encoding of discrete sequences are performed. The load data are decomposed and denoised using the double decomposition modal (LVMD) strategy, the load curves are clustered using the double weighted fuzzy C-means (DBFCM) algorithm, and the typical curves obtained are used as load patterns. In addition, data feature analysis is performed. A convolutional neural network (CNN) is used to extract data features. A bidirectional long short-term memory (BLSTM) network is used for prediction, in which the number of hidden layer neurons, the number of training epochs, the learning rate, the regularization coefficient, and other relevant parameters in the BLSTM network are optimized using the influenza virus immunity optimization algorithm (IVIA). Finally, the historical data of City H from 1 January 2016 to 31 December 2018, are used for load forecasting. The experimental results show that the novel model based on LVMD-DBFCM load c1urve clustering combined with CNN-IVIA-BLSTM proposed in this paper has an error of only 2% for electric load forecasting.

In this paper, a new metaheuristic optimization algorithm is proposed which is named as Influenza Virus Immune Algorithm (IVIA). The algorithm is derived by studying the process of influenza virus transmission and immunization in the population. Different infection strategies are developed by classifying the contact distance between the infected person and other individuals in the population into different situations. The process of an individual being infected is divided into three states: uninfected, infected, and immune. The immune individual will not be infected when confronted with the virus again. The population has herd immunity when the number of immune individuals in the population exceeds 60% of the total population. In order to test the optimization effect of IVIA, this paper tests IVIA with 23 sets of standard test functions and compares it with 7 optimization algorithms which were proposed in recent years,such as TFWO, GEO, PPA, RSO, GWO, PSO and WOA. The results of the tests show that IVIA converges faster with higher accuracy, optimizes the function better, and finds the minimum value of the function in fewer iterations. Then IVIA is applied to short-term power load forecasting in order to verify its practical application performance. Least Squares Support Vector Machine(LSSVM) and Long Short-Term Memory(LSTM) are used as the base prediction models respectively, and then different optimization algorithms are used to find the best initial parameters for them.The example results show that the model of IVIA-LSSVM has the smallest error evaluation index and the best overall forecasting effect among all the combined forecasting models. it can meet the accuracy requirements of load prediction results. Therefore, it is proved that IVIA has better optimization effect and it can be applied to practical engineering field.

As more and more swarm intelligence algorithms are proposed, these optimization algorithms are used in different combined forecasting models to improve the problem of low prediction accuracy of a single model. This paper proposes SSALSSVM short-term load forecasting model, through the good overall optimization ability and fast convergence characteristics of the Sparrow Search Algorithm(SSA) chose the best initial parameters for Least Squares Support Vector Machine(LSSVM).Finally, an examples verify the usability and good prediction effect of the proposed model, by comparing with the PSO-LSSVM model, it is found that the SSA-LSSVM model has higher prediction accuracy and shorter time, which can better meet the performance requirements of the short-term load forecasting model.

In order to optimize the ejector performance, the internal flow field of steam ejector is simulated by using Fluent software. In this paper, by using the single factor variable method, we will study nozzle’s structure to improve ejector performance and to reduce the ejector energy loss, which make the ejector work normally. When studying the nozzle, the other structure of the ejector should remain unchanged. The results indicate that there is an respectively optimum Figure of the nozzle convergent length, the nozzle throat length and the nozzle divergent length to make ejection coefficient reach its maximum value. In addition, the nozzle throat diameter have also an optimum value, which makes ejector performance reach its tip.

In remote areas far from the grid, wind/PV/storage generating system is relatively a good choice, whatever in resource configuration, performance or prices. For the independent hybrid power system, the output models of wind turbines, photovoltaic arrays and batteries are built in this paper, and based on the objectives of the capacity configuration optimal model, constraints used in the process of capacity configuration are analyzed. These provide convenient conditions and theoretical basis for the optimal capacity configuration of independent wind/PV/storage system.

Main steam temperature control system has a big delay in the control object, the accuracy and stability must follow the requirements. It is very difficult to get a good running result by PID control system. Smith predictor brings a link of predictive compensation at the basic of PID feedback control system, so it can perfect the big delay, and the controlled quality of the system can be greatly improved. The adaptive gain control system is developed at the basic of smith predictor control, it is able to overcome the instability of the control system model errors.

The wind energy resource is rich and superior in the Inner Mongolia area. Wind power industry reach a preliminary scale. But with the wind power industry rapid development, appear some critical problems such as power grid lagged, equipment failure. The problems affect the wind power industry healthy and rapid development. To enhance the policies and measures about power grid construct, scientific development, equipment estimate is the guarantee of Inner Mongolia wind power industry medium & long-term development.

Wind power industries begin to take shape in Inner Mongolia, pitch control is commonly used in a variety of wind turbines. Fuzzy adaptive control by seeking the three parameters of PID controller and the relationship between e and ec, make the best adjustment for PID parameters on the pitch system automatically. This method can make the pitch mechanism with high performance to ensure the output power of wind turbine is more stable.

Renewable energy which independent power generation systems from the nets is increasingly extensive research and application. This paper is in connection with ac distribution problems of village, it is with the foundation of multiple national's standard and field operation experience. From lines in-station, transmission and distribution line, distribution cabinets and other power distribution equipment structure, function, capacity selection, and so on, this paper has a positive research summary and elaboration, can provide the beneficial reference for relevant standards.