Search

Under the impact of frost heave and thaw settlement of seasonal frozen soil, freezing pulling and horizontal tilt of piles often occurs. Aim to the above problems, this paper proposed a material using ceramsite replacement concrete as the foundation to reduce heat exchange with atmosphere. The results showed that when the replacement rate of ceramsite is 80%, the density of the blocks is 1896 kg/m3, which is 19.3% lower than that of plain concrete. Also, the addition of ceramsite significantly reduced the thermal conductivity of concrete. The minimum thermal conductivity of tested blocks is 0.524 W/(m.K), which is 58.6% lower than that of ordinary concrete. Meanwhile, the strength of concrete decreased the addition of ceramsite. When the replacement rate is 80%, the compressive strength and splitting tensile strength are 18.1 and 2.7 MPa, respectively, which are 54.7% and 42.6% lower than that of plain concrete. Taking the climatic environment in the Hexi area as the boundary conditions, a numerical simulation on the design and optimization of the thickness of the insulation system was carried out. It was found that the installation of 2 m insulation concrete can effectively increase the temperature of the soil in extreme environments and avoid frost heave.

The autoregressive model algorithm (ARA) of time series analysis and reverse transmission neural network algorithm (BPA) of machine learning were used in this paper, and applied the load forecasting of electric vehicle (EV), through studying and analysising of different algorithms, aiming at certain domestic city electric vehicle charging station has carried on the short-term load forecast, compared of prediction results of the two ways, the simulation results show that different algorithm also have different effects in the short-term load forecasting, therefore, according to different system to select appropriate learning algorithm is particularly important.

A new algorithm was proposed in this paper. Based on the principle that the single model prediction results with small errors accounted for a large proportion of the combined results, the BP neural network and the Autoregressive Moving Average (ARMA) Model were improved and combined to form a new combined model algorithm (NEW ARMA-BP) to predict PV output to get more accurate predictions. Taking the output power collected by a 30 MW photovoltaic power plant as an input sample, based on the ARMA algorithm and BP neural network, these two prediction models were built in sequence under the Matlab environment to predict the short-term PV output. Using the “normal test chart of error” in probability theory, we judged that the prediction error level based on the BP neural network was closer to the normal distribution, indicating that the prediction was more accurate. Based on the prediction errors of the ARMA algorithm and BP neural network algorithm, new methods were used to calculate the weights, and the prediction results were recombined based on the weights to obtain new prediction values. Finally, based on the three kinds of error assessment indicators and Pearson coefficient, the error levels of NEW ARMA-BP algorithm and uni-prediction model were compared to verify the applicability of the combined forecast in the field of PV output prediction.