Vol. 05, No. 11 [November 2019]

With the rapid economic and social development of Province A, the province's electricity consumption level has improved significantly. During the summer and winter peak hours, the province has adopted various levels of orderly electricity use measures, and the power supply support capacity has reached the limit. Hunan's energy endowment is insufficient, and the province's power supply and demand has been unable to achieve self-balance. With the commissioning of the UHV DC project, the scale of purchasing electricity outside the province has grown rapidly. The role of foreign calls in ensuring the power supply of the province, reducing the price of electricity in the province, and improving the ecological environment are becoming more and more obvious. In order to solve the above problems, we first analyze the inter-provincial power trading varieties and the factors affecting the purchase of electricity, and analyze the problems according to the purchase situation of the province in 2018, and put forward the problems to be solved. Then, the outsourcing electricity model is established. Finally, the average electricity price of the province is analyzed by various uncertain scenarios. Under the premise of satisfying the constraints, the optimal power purchase plan is obtained, which is formulated for the province A. Provide a reference for annual and monthly trading plans.
Keywords:Outsourcing Eletricity Model, Monthly Trading Plans, Clean Energy Consumption, Smart Grid

The porosity is a defect in a weld, which results in weakening the strength of the joint and corrode the metal parts by depositing impurities like moisture and atmospheric contaminant. This defect is caused due to disturbance of external impurities like oxygen, nitrogen, hydrogen, etc., in the molten weld pool. In fact, most of the industrial and manufacturing companies play a key role in finish products. Besides, there is substantial use of welding in building construction and therefore the quality of welding effects the strength of welded structures. In general, the entire product quality depends on the weld quality, irrespective of the type of weld being used. The many researchers have examined the different mechanisms to reduce the porosity such as adjusting the pulse frequency, a set of arc current, depth of penetration, proper application of welding parameters, enough duration to solidification process, etc. This review paper will disseminate the preventive measures to reduce the porosity as a collective idea in a small amount to count for big volume.
Keywords: porosity, impurities, contaminant, depositing and penetration.

The study includes an analysis of data from 2011 to 2018, it was proposed to obtain the best or best Probability Distribution Functions that Model PMCO Concentrations in Mexico City, using the following PDF, Gama Distribution Function, Function of Distribution Extreme Value Distribution and Nakagami Distribution Function, to obtain the estimators, the Maximum Probability and Moments Method was used and aided by the Matlab 2017 program, for the evaluation of the forecast model, RMSE, Determination Coefficient, Prediction Approach and Index were used Approach, in turn an analysis is made to observe its trend within the period using the Bayesian Point Change Detection method to observe the most notable variations of the trend, the database used is from the official air page From Mexico City.
Keywords: Particulate Material Coarse Fraction, Probability Distributions, Adjustment Indicators.

Present paper studies hybrid flow shop (HFS) scheduling problem with stage skipping and adjustable processing time in steelmaking- continuous casting (SCC) production process. The SCC scheduling problem is solved to determine the machine allocations, starting times and ending times for all operations of all sms charges (jobs). Considering the production objectives and constraints related to stageskipping and adjustable processing time, a new SCC scheduling model is built. Genetic algorithm (GA) is applied to address the scheduling problem.
Keywords: scheduling, hybrid flow shop, stage skipping, adjustable processing time, genetic algorithm

This paper is on the design of milk churning mechanism manually for the benefit of farmers. This efficient and affordable mechanism is expected to support most of the farmers in the rural areas in the country. It is learnt that most of farmers are depending on conventional tool which requires physical effort and not efficient in the production process. Although milk production provides economic value to farmers, demanding physical strength and consumption of time in the current practice of milk production has led to shortcoming in other daily farming activities. To mitigate this challenges and to strengthen economic of rural farmers, the design and development of this milk churning mechanism is initiated as research project. The outcome is intended to serve as resourceful tool in the milk production.
Keywords: churner, energetic, exponential, agriculture, robust.

Question classification has become an important task in the research of automatic question answering but remain challenging for complexity of problem, which can be analyzed from syntax and semantic subspace generally. Consequently, an integrated algorithm that mixes multiple deep learning models is proposed to capture the syntactic information and deep semantic subspace information of travel-related textsin this paper. On the one hand, a word-level Convolutional Neural Network (CNN) can acquires low-level subspace structure information from a travel-related text sequence to improve the characterizing ability of text-level spatial structure information.On the other hand, a sentence-level Bi-directional Long Short-Term Memory Network (Bi-LSTM)obtains the global deep semantic and syntactic information of the text from the text sentence, and complements the low-level spatial structure information acquired by the word-level CNN. Furthermore, the multi-headed self-attention mechanism is used to integrate the both networksabove, and assign weights of attention to the captured features. Finally, accurate classification of texts in the tourism field is achieved through the SoftMax classifier at the end of the model. The experimental results indicate that the proposed integrated algorithm improves the accuracy of travel text classification, compared with the traditional deep learning model.
Keywords: Bidirectional long short-term memory network, Convolutional neural network, Deep abstract semantic information, Low-level subspace structure information, Multi-headed attention