In order to improve the precision of soil organic carbon (SOC) estimates, the sources of uncertainty in soil organic carbon density (SOCD) estimates and SOC stocks were examined using 363 soil profiles in Hebei Provin...In order to improve the precision of soil organic carbon (SOC) estimates, the sources of uncertainty in soil organic carbon density (SOCD) estimates and SOC stocks were examined using 363 soil profiles in Hebei Province, China, with three methods: the soil profile statistics (SPS), GIS-based soil type (GST), and kriging interpolation (KI). The GST method, utilizing both pedological professional knowledge and GIS technology, was considered the most accurate method of the three estimations, with SOCD estimates for SPS 10% lower and KI 10% higher. The SOCD range for GST was 84% wider than KI as KI smoothing effect narrowed the SOCD range. Nevertheless, the coefficient of variation for SOCD with KI (41.7%) was less than GST and SPS. Comparing SOCD’s lower estimates for SPS versus GST, the major sources of uncertainty were the conflicting area of proportional relations. Meanwhile, the fewer number of soil profiles and the necessity of using the smoothing effect with KI were its sources of uncertainty. Moreover, for local detailed variations of SOCD, GST was more advantageous in reflecting the distribution pattern than KI.展开更多
A machine-learning approach was developed for automated building of knowledgebases for soil resources mapping by using a classification tree to generate knowledge from trainingdata. With this method, building a knowle...A machine-learning approach was developed for automated building of knowledgebases for soil resources mapping by using a classification tree to generate knowledge from trainingdata. With this method, building a knowledge base for automated soil mapping was easier than usingthe conventional knowledge acquisition approach. The knowledge base built by classification tree wasused by the knowledge classifier to perform the soil type classification of Longyou County,Zhejiang Province, China using Landsat TM bi-temporal images and CIS data. To evaluate theperformance of the resultant knowledge bases, the classification results were compared to existingsoil map based on a field survey. The accuracy assessment and analysis of the resultant soil mapssuggested that the knowledge bases built by the machine-learning method was of good quality formapping distribution model of soil classes over the study area.展开更多
Due to the insufficiency of utilizing knowledge to guide the complex optimal searching, existing genetic algorithms fail to effectively solve excavator boom structural optimization problem. To improve the optimization...Due to the insufficiency of utilizing knowledge to guide the complex optimal searching, existing genetic algorithms fail to effectively solve excavator boom structural optimization problem. To improve the optimization efficiency and quality, a new knowledge-based real-coded genetic algorithm is proposed. A dual evolution mechanism combining knowledge evolution with genetic algorithm is established to extract, handle and utilize the shallow and deep implicit constraint knowledge to guide the optimal searching of genetic algorithm circularly. Based on this dual evolution mechanism, knowledge evolution and population evolution can be connected by knowledge influence operators to improve the conflgurability of knowledge and genetic operators. Then, the new knowledge-based selection operator, crossover operator and mutation operator are proposed to integrate the optimal process knowledge and domain culture to guide the excavator boom structural optimization. Eight kinds of testing algorithms, which include different genetic operators, arc taken as examples to solve the structural optimization of a medium-sized excavator boom. By comparing the results of optimization, it is shown that the algorithm including all the new knowledge-based genetic operators can more remarkably improve the evolutionary rate and searching ability than other testing algorithms, which demonstrates the effectiveness of knowledge for guiding optimal searching. The proposed knowledge-based genetic algorithm by combining multi-level knowledge evolution with numerical optimization provides a new effective method for solving the complex engineering optimization problem.展开更多
Knowledge graph technology is widely applied in the domain of general knowledge reasoning with an excellent performance.For fine-grained professional fields,professional knowledge graphs can provide more accurate info...Knowledge graph technology is widely applied in the domain of general knowledge reasoning with an excellent performance.For fine-grained professional fields,professional knowledge graphs can provide more accurate information in practical industrial scenarios.Based on an aviation assembly domain-specific knowledge graph,the article constructs a joint knowledge reasoning model,which combines a named entity recognition model and a subgraph embedding learning model.When performing knowledge reasoning tasks,the two models vectorize entities,relationships and entity attributes in the same space,so as to share parameters and optimize learning efficiency.The knowledge reasoning model,which provides intelligent question answering services,is able to reduce the assembly error rate and improve the assembly efficiency.The system can accurately solve general knowledge reasoning problems in the assembly process in actual industrial scenarios of general assembly and component assembly under interference-free conditions.Finally,this paper compares the proposed knowledge reasoning model based on knowledge representation learning and the question-answering system based on large-scale pre-trained models.In the application scenario of system functional testing in general assembly,the joint model attains an accuracy rate of 95%,outperforming GPT with 78%accuracy and enhanced representation through knowledge integration with 71%accuracy.展开更多
基金Project supported by the Knowledge Innovation Project in Leading Edge Fields, Chinese Academy of Sciences(No. ISSASIP0201), the National Key Basic Research Support Foundation of China (No. G1999011810) and the KnowledgeInnovation Project in Resource and
文摘In order to improve the precision of soil organic carbon (SOC) estimates, the sources of uncertainty in soil organic carbon density (SOCD) estimates and SOC stocks were examined using 363 soil profiles in Hebei Province, China, with three methods: the soil profile statistics (SPS), GIS-based soil type (GST), and kriging interpolation (KI). The GST method, utilizing both pedological professional knowledge and GIS technology, was considered the most accurate method of the three estimations, with SOCD estimates for SPS 10% lower and KI 10% higher. The SOCD range for GST was 84% wider than KI as KI smoothing effect narrowed the SOCD range. Nevertheless, the coefficient of variation for SOCD with KI (41.7%) was less than GST and SPS. Comparing SOCD’s lower estimates for SPS versus GST, the major sources of uncertainty were the conflicting area of proportional relations. Meanwhile, the fewer number of soil profiles and the necessity of using the smoothing effect with KI were its sources of uncertainty. Moreover, for local detailed variations of SOCD, GST was more advantageous in reflecting the distribution pattern than KI.
基金Project supported by the National Natural Science Foundation of China(Nos.40101014 and 40001008).
文摘A machine-learning approach was developed for automated building of knowledgebases for soil resources mapping by using a classification tree to generate knowledge from trainingdata. With this method, building a knowledge base for automated soil mapping was easier than usingthe conventional knowledge acquisition approach. The knowledge base built by classification tree wasused by the knowledge classifier to perform the soil type classification of Longyou County,Zhejiang Province, China using Landsat TM bi-temporal images and CIS data. To evaluate theperformance of the resultant knowledge bases, the classification results were compared to existingsoil map based on a field survey. The accuracy assessment and analysis of the resultant soil mapssuggested that the knowledge bases built by the machine-learning method was of good quality formapping distribution model of soil classes over the study area.
基金supported by National Natural Science Foundation of China(Grant No.51175086)
文摘Due to the insufficiency of utilizing knowledge to guide the complex optimal searching, existing genetic algorithms fail to effectively solve excavator boom structural optimization problem. To improve the optimization efficiency and quality, a new knowledge-based real-coded genetic algorithm is proposed. A dual evolution mechanism combining knowledge evolution with genetic algorithm is established to extract, handle and utilize the shallow and deep implicit constraint knowledge to guide the optimal searching of genetic algorithm circularly. Based on this dual evolution mechanism, knowledge evolution and population evolution can be connected by knowledge influence operators to improve the conflgurability of knowledge and genetic operators. Then, the new knowledge-based selection operator, crossover operator and mutation operator are proposed to integrate the optimal process knowledge and domain culture to guide the excavator boom structural optimization. Eight kinds of testing algorithms, which include different genetic operators, arc taken as examples to solve the structural optimization of a medium-sized excavator boom. By comparing the results of optimization, it is shown that the algorithm including all the new knowledge-based genetic operators can more remarkably improve the evolutionary rate and searching ability than other testing algorithms, which demonstrates the effectiveness of knowledge for guiding optimal searching. The proposed knowledge-based genetic algorithm by combining multi-level knowledge evolution with numerical optimization provides a new effective method for solving the complex engineering optimization problem.
基金supported by the National Natural Science Foundation of China(Grant Nos.52275020,62293514,and 91948301).
文摘Knowledge graph technology is widely applied in the domain of general knowledge reasoning with an excellent performance.For fine-grained professional fields,professional knowledge graphs can provide more accurate information in practical industrial scenarios.Based on an aviation assembly domain-specific knowledge graph,the article constructs a joint knowledge reasoning model,which combines a named entity recognition model and a subgraph embedding learning model.When performing knowledge reasoning tasks,the two models vectorize entities,relationships and entity attributes in the same space,so as to share parameters and optimize learning efficiency.The knowledge reasoning model,which provides intelligent question answering services,is able to reduce the assembly error rate and improve the assembly efficiency.The system can accurately solve general knowledge reasoning problems in the assembly process in actual industrial scenarios of general assembly and component assembly under interference-free conditions.Finally,this paper compares the proposed knowledge reasoning model based on knowledge representation learning and the question-answering system based on large-scale pre-trained models.In the application scenario of system functional testing in general assembly,the joint model attains an accuracy rate of 95%,outperforming GPT with 78%accuracy and enhanced representation through knowledge integration with 71%accuracy.
