查看更多>>摘要：In order to solve the problems of high average absolute error and long time consumption in traditional forecasting methods, a short-term load prediction method of electric vehicle charging stations based on conditional generative adversarial networks is proposed. This method involves the analysis of the initial charging time, initial state of charge, and battery characteristics of electric vehicles. Based on the analysis results, a conditional generative adversarial networks (CGAN) model is constructed to anticipate the short-term load of electric vehicle charging stations. In the CGAN model, the charging start time, initial state of charge, and battery characteristics of electric vehicles serve as conditional values. Through training, the model learns the relationship between these conditions and the target, generating accurate load forecasting results. Experimental findings reveal that the proposed method boasts a maximum average absolute error of merely 1.4% and a minimum prediction time of just 1.26 seconds, thus demonstrating its practicality.
查看更多>>摘要：In order to solve the problems of low accuracy and low efficiency of existing automatic classification methods of construction waste, an automatic classification method of construction waste based on machine vision was proposed. Firstly, the CCD camera is used to collect the image and enhance the image. Then, the maximum entropy method is used to obtain the optimal segmentation threshold of the image, and the construction waste image is segmented. Finally, the gradient information is used to obtain the image features of construction waste, and the automatic classification of construction waste is realised by combining with the SVM algorithm. The experimental results show that the classification accuracy of the proposed method is between 90% and 98%. When the number of construction waste images is 1,000, the classification time is 13 min, which indicates that the proposed method has high classification accuracy, high efficiency and good application performance.
查看更多>>摘要：To enhance the accuracy and recall rate of landscape ecological quality assessment, this study proposes a method based on remote sensing ecological index (RSEI). Firstly, data on landscape architecture's ecological quality are collected, and the RSEI is derived from remote sensing images. Secondly, ecological quality assessment component indicators are constructed, and principal component analysis (PCA) is employed to combine these indicators. Lastly, the grey correlation model is utilised to analyse the ecological environment situation, and the equidistant separation method is applied to categorise RSEI into five levels. The ecological quality assessment results are obtained using spatial statistical methods. The findings reveal that the prediction error rate of RSEI values in this method is controlled within 2.87%, achieving an accuracy of 99.2% and a recall rate of 99.9%. This indicates that the method has the potential to improve the comprehensiveness of landscape ecological quality assessment.
查看更多>>摘要：In order to improve accounting accuracy and shorten accounting time, the paper proposes a positive and negative balance accounting method for carbon emissions in parks based on K-nearest neighbour clustering algorithm. Firstly, collect and standardise the carbon emission data of the park. Then, K-nearest neighbour clustering algorithm is used to cluster the carbon emission management items. After reconstructing the data components, the reconstruction component with the highest sample entropy is decomposed twice to obtain the carbon emission coefficient of land use type. Finally, a carbon emission balance accounting model is constructed, and the accounting results are obtained by integrating various production factors in the model. The experimental shows that after applying this method, the accuracy and recall of carbon emission accounting can reach 96.09% and 99.6%, respectively. The time required for positive and negative balance accounting is only 2.5 minutes, indicating that this method has achieved the design expectations.
查看更多>>摘要：In response to the low accuracy and poor comprehensiveness of existing research on the factors affecting carbon emissions in regional tourism industry, this paper conducts research on the factors affecting carbon emissions in regional tourism industry based on the LMDI model. Firstly, carbon emission related data is collected from the tourism industry and the data is pre-processed using normalisation methods. Secondly, a bottom-up approach is adopted to estimate various energy consumption during the tourism process. Finally, the LMDI model is used for factor decomposition to study the factors affecting carbon emissions. Through experiments, it has been proven that the application of the LMDI model to analyse the influencing factors of carbon emissions can always be more than 90% comprehensive, and the error between the calculated carbon emissions and the actual carbon emissions is always less than 100,000 tons. The effect of analysing the influencing factors of carbon emissions is good.
查看更多>>摘要：In order to reduce the error of carbon emission peak prediction and shorten the prediction time, an expressway toll station carbon emission peak prediction method based on the GRA-LSTM model is proposed in the background of dual carbon. Firstly, analyse the dual carbon goals and the characteristics of sustainable development. Secondly, convert the energy consumption generated during the vehicle's payment process into the vehicle's carbon emissions data. Finally, use the grey correlation analysis (GRA) method based on the collected carbon emission data, to calculate the correlation degree between the factors affecting carbon emissions. Using the long short-term memory (LSTM) model to construct a carbon emission peak prediction model, and the output result is the carbon emission peak prediction result. The experimental results show that the proposed method can shorten the prediction time while reducing the prediction RSME.
查看更多>>摘要：To overcome the problems of low accuracy, long calculation time, and minimal carbon emission reduction in traditional carbon emission calculation and control methods, a new carbon emission calculation and control method of agricultural product supply chain under the background of energy conservation and emission reduction is proposed. The grey relational analysis and extended STIRPAT model are used to select the influencing factors of carbon emissions in the agricultural product supply chain, and the AOA-LSTM model is used to calculate the carbon emissions. The carbon emissions of the agricultural product supply chain under the background of energy conservation and emission reduction are controlled based on allocation adjustment factors, carbon emissions increment distribution ratios, and allocation quotas. The experimental results show that the accuracy of the proposed method varies between 94.9% and 97.9%, with a maximum calculation time of 1.03 s. The carbon emission reduction after nine months 1.238 ×106t.
查看更多>>摘要：The preparation of flood zoning maps is crucial in global urban development studies, serving as fundamental information for investment and project implementation assessments by relevant organisations. This study focused on hydraulic modelling and flood zoning in Sulaymaniyah urban basin using hydrologic model HEC-RAS and GIS. Given the escalating urban flood frequency, effective solutions are imperative. The 2D hydraulic model HEC-RAS was employed for a more accurate simulation of flow patterns, especially in flood-prone areas. Hydrological data, digital elevation model, and urban structure data were utilised for modelling, with the aim to identify vulnerable areas and propose damage reduction solutions. The study generated water surface profile maps, width, depth, and flow velocity for return periods of 2, 5, 10, 25, 50, and 100 years. With a 25-year return period, flood vulnerability was classified based on depth and flow velocity, revealing substantial risks in urban areas. H5 and H6 hazard zones covered 43.5% and 24.5% of these areas, posing threats to individuals, vehicles, and structures. The H4 hazard zone, comprising 8.8% of the flood-prone area, presented risks to people and vehicles. Also, the correctness of choosing Manning's roughness coefficient was evaluated. This research provides valuable insights for urban flood management and hazard mitigation.
查看更多>>摘要：Teak trees (Tectona grandis) are known for their excellent and expensive wood quality. However, teak trees have many contributions to the environment, such as maintaining soil stability and preventing erosion. Teak leaves also make an environmental contribution, namely producing oxygen and absorbing carbon. This study aims to create a plant computational model of teak trees using the functional-structural plant modelling (FSPM) method which is implemented on the growth grammar-related interactive modelling platform (GrolMP). This model can simulate the growth of teak trees morphologically and predict the contribution of teak trees to the environment such as the total oxygen produced, and the carbon absorption. The model simulated that one teak tree at the age of 20 can produce 1,800 grams of oxygen per day and absorb 670 grams of carbon per day, providing enough oxygen for three people in one day.
查看更多>>摘要：A carbon flow tracking method for the power system in an energy-saving and emission reducing environment is studied in order to accurately track the carbon flow of the power system and reduce the carbon footprint error rate. Firstly, carbon emission data is collected and features using Pearson correlation coefficients are extracted. Then, a carbon emission factor prediction model is established through neural networks, and the MDI method is used to calculate the carbon emission intensity of the power system. Finally, a DC power flow model is introduced with carbon emission intensity as input to achieve carbon flow tracking. The experimental results show that the carbon footprint error rate of the method proposed in this paper is 5.2%, the cost-effectiveness ratio of emission reduction is 80 yuan/ton of CO_2, and it has strong anti-interference ability against data noise, demonstrating good carbon flow tracking performance.
NETL
NSTL