Modelling of shallow landslides with machine learning algorithms

This paper introduces three machine learning(ML)algorithms,the‘ensemble'Random Forest(RF),the‘ensemble'Gradient Boosted Regression Tree(GBRT)and the Multi Layer Perceptron neural network(MLP)and applies them to the spatial modelling of shallow landslides near Kvam in Norway.In the development of the ML models,a total of 11 significant landslide controlling factors were selected.The controlling factors relate to the geomorphology,geology,geo-environment and anthropogenic effects:slope angle,aspect,plan curvature,profile curvature,flow accumulation,flow direction,distance to rivers,water content,saturation,rainfall and distance to roads.It is observed that slope angle was the most significant controlling factor in the ML analyses.The performance of the three ML models was evaluated quantitatively based on the Receiver Operating Characteristic(ROC)analysis.The results show that the‘ensemble'GBRT machine learning model yielded the most promising results for the spatial prediction of shallow landslides,with a 95%probability of landslide detection and 87%prediction efficiency.

Designing and Performance Evaluation of Biochar Production in a Top-Lit Updraft Up-scaled Gasifier

The Original Belonio Rice Husk Gasifier (OBRHG), initially of height of 0.6 m, diameter of 0.15 m and thickness of 0.025 m was tested for biochar production through air gasification of rice husk (RH) and the design was upscaled to height of 1.65 m, diameter of 0.85 m and thickness of 0.16 m. A total of 27 experiments were conducted to monitor the gasifier performance and the system can operate with the centrifugal blower operating at a power input of 155 W and a maximum flow rate of 1450 m3/hr regulated according to the air requirement. Building the UBRHG is simple and inexpensive to fabricate and with the fairly satisfactory performance and ease of construction along with the convenience of operation, the UBRHG with RH as feed would find abundant avenues of applications in a rural setting for biochar production alongside thermal, mechanical and electrical energy delivery.

Improved Gasification of Rice Husks for Optimized Biochar Production in a Top Lit Updraft Gasifier

Biochar is a solid material obtained from the carbonization of biomass. If properly produced, it is useful for soil application to enrich plant values. Rice husk (RH) waste, an abundant agricultural by-product, was gasified in a top-lit updraft Belonio rice husk gasifier with a biochar yield of 29.0% ± 1.9%. The equivalence ratio (ER) for optimum biochar production was identified and its effect on biochar properties such as pH, volatile matter (VM), fixed carbon (FC) and ash content (AC), electricity consumption, biochar yield, specific gasification rate (SGR) as well as reactor temperature investigated and statistically analyzed. As ER increased from 0.292 ± 0.005 to 0.442 ± 0.016, the SGR decreased from 85.4 ± 4.5 kg/(m2hr) to 51.6 ± 2.4 kg/(m2hr) whereas reactor temperature increased linearly with ER. The original VM content of RH was found to be 76.1% ± 1.2% and decreased with increasing ER from 14.1% ± 0.2% to 10.6% ± 0.3%. The original FC and AC of 5.49% ± 0.22% and 9.10% ± 1.23% increased with ER from 50.5% ± 0.7% to 51.3% ± 0.4% and 33.7% ± 0.4% to 36.7% ± 0.1% respectively. The biochar pH at low, medium and high ER was 9.36 ± 0.11, 9.64 ± 0.03 and 9.42 ± 0.01, respectively. Results revealed a significant change in biochar yield and proximate values as ER changes from low to high.

Sustainable Technologies for Small-Scale Biochar Production—A Review

Charcoal has found enormous application in both agriculture (AKA biochar) and other sectors. Despite its potential benefits, small scale technologies relevant for its production remain a challenge. Technologies striking a balance between user friendliness, energy efficiency, ease of adaptation and limited emissions could easily be integrated into the local community for the sustainable production of biochar answering both technical and socio-economic aspects. These technologies can be customized to recover the produced heat alongside biochar and the producer gas. The purpose of this work is to review the state of the art in small scale technologies, their associated risks and challenges as well as research gaps for future work. Factors affecting biochar production have been discussed and temperature is known to heavily influence the biomass to biochar conversion process. Based on the reviewed work, there is a need to develop and promote sustainable and efficient technologies that can be integrated into biochar production systems. There is also further need to develop portable, economically viable technologies that could be integrated into the biochar production process without compromising the quality of produced biochar. Such technologies at midscale level can be channeled into conventional small scale farmer use in order that the farmers can process their own biochar.

The presence and partitioning behavior of flame retardants in waste,leachate,and air particles from Norwegian waste-handling facilities

Flame retardants in commercial products eventually make their way into the waste stream.Herein the presence of flame retardants in Norwegian landfills, incineration facilities and recycling sorting/defragmenting facilities is investigated. These facilities handled waste electrical and electronic equipment（WEEE）, vehicles, digestate, glass, combustibles, bottom ash and fly ash. The flame retardants considered included polybrominated diphenyl ethers（∑BDE-10） as well as dechlorane plus, polybrominated biphenyls, hexabromobenzene,pentabromotoluene and pentabromoethylbenzene（collectively referred to as ∑FR-7）. Plastic,WEEE and vehicles contained the largest amount of flame retardants（∑BDE-10： 45,000–210,000 μg/kg; ∑FR-7： 300–13,000 μg/kg）. It was hypothesized leachate and air concentrations from facilities that sort/defragment WEEE and vehicles would be the highest. This was supported for total air phase concentrations（∑BDE-10： 9000–195,000 pg/m^3 WEEE/vehicle facilities, 80–900 pg/m^3 in incineration/sorting and landfill sites）, but not for water leachate concentrations（e.g., ∑BDE-10： 15–3500 ng/L in WEEE/Vehicle facilities and 1–250 ng/L in landfill sites）. Landfill leachate exhibited similar concentrations as WEEE/vehicle sorting and defragmenting facility leachate. To better account for concentrations in leachates at the different facilities, waste-water partitioning coefficients, Kwastewere measured（for the first time to our knowledge for flame retardants）. WEEE and plastic waste had elevated Kwastecompared to other wastes, likely because flame retardants are directly added to these materials. The results of this study have implications for the development of strategies to reduce exposure and environmental emissions of flame retardants in waste and recycled products through improved waste management practices.

边坡位移智能预测算法

目的:边坡位移预测是实现滑坡灾害预报的有效手段,对降低滑坡灾害导致的损失具有重要意义。本文针对三峡库区广泛分布的"阶跃型"滑坡,采用三种不同的机器学习算法:长短期记忆(LSTM)神经网络、随机森林(RF)算法和门控递归单元(GRU),预测三个不同的三峡库区边坡位移,并对比三种算法的预测精度,从而选择适用于边坡位移预测的机器学习算法。创新点:1.建立了基于时间序列分解和机器学习算法的动态预测模型,并能够准确预测边坡位移。2.对比了不同的机器学习算法预测边坡周期项位移的精度。方法:1.基于时间序列分解原理,将边坡累积位移分解为趋势项位移和周期项位移。2.利用多项式拟合对边坡趋势项位移进行预测。3.基于位移影响因素采用三种机器学习模型(LSTM、GRU和RF)预测边坡周期项位移。结论:1.本文提出的基于时间序列分解和机器学习算法的动态预测模型可以准确预测三峡库区"阶跃型"边坡位移。2.LSTM和GRU算法可以充分利用滑坡历史信息,精确预测边坡位移的周期项。

Soft computing approach for prediction of surface settlement induced by earth pressure balance shield tunneling

Estimating surface settlement induced by excavation construction is an indispensable task in tunneling,particularly for earth pressure balance(EPB)shield machines.In this study,predictive models for assessing surface settlement caused by EPB tunneling were established based on extreme gradient boosting(XGBoost),artificial neural network,support vector machine,and multivariate adaptive regression spline.Datasets from three tunnel construction projects in Singapore were used,with main input parameters of cover depth,advance rate,earth pressure,mean standard penetration test(SPT)value above crown level,mean tunnel SPT value,mean moisture content,mean soil elastic modulus,and grout pressure.The performances of these soft computing models were evaluated by comparing predicted deformation with measured values.Results demonstrate the acceptable accuracy of the model in predicting ground settlement,while XGBoost demonstrates a slightly higher accuracy.In addition,the ensemble method of XGBoost is more computationally efficient and can be used as a reliable alternative in solving multivariate nonlinear geo-engineering problems.

Earthquake Disaster Risk Reduction in Iran:Lessons and "Lessons Learned" from Three Large Earthquake Disasters——Tabas 1978,Rudbar 1990,and Bam 2003

This article addresses three large earthquake disasters in Iran: Tabas in 1978, Rudbar in 1990, and Bam in 2003. Lessons and 'Lessons Learned' from these three earthquake disasters were investigated together with their contributions over time towards earthquake disaster risk reduction in Iran. Many lessons from 1978 Tabas, 1990 Rudbar, and 2003 Bam did not become 'Lessons Learned' and they were identified again within the dramatic context of other earthquake disasters in various places of Iran. Both lessons and 'Lessons Learned' from Tabas, Rudbar, Bam,and other earthquake disasters in Iran require a sustainable long-term framework—an earthquake culture.