In an extraction turbine, pressure of the extracted steam and rotate speed of the rotor are two important controlled quantities. The traditional linear state feedback control method is not perfect enough to control th...In an extraction turbine, pressure of the extracted steam and rotate speed of the rotor are two important controlled quantities. The traditional linear state feedback control method is not perfect enough to control the two quantities accurately because of existence of nonlinearity and coupling. A generalized minimum variance control method is studied for an extraction turbine. Firstly, a nonlinear mathematical model of the control system about the two quantities is transformed into a linear system with two white noises. Secondly, a generalized minimum variance control law is applied to the system. A comparative simulation is done. The simulation results indicate that precision and dynamic quality of the regulating system under the new control law are both better than those under the state feedback control law.展开更多
Identifying the main results of scientific papers is an important and challenging task.Unlike traditional named entity recognition(NER)in general domains,NER in the scientific field primarily aims to identify and clas...Identifying the main results of scientific papers is an important and challenging task.Unlike traditional named entity recognition(NER)in general domains,NER in the scientific field primarily aims to identify and classify specific scientific entity types from scientific literature.This requires handling professional terminology,complex sentence structures,and domain-specific contextual information.Recently,generation methods based on large language models(LLMs)have attempted to co-train models on multiple datasets while using prompt instructions to accurately extract target entities.However,there is currently insufficient exploration of document-level NER methods in the scientific field.To further fill this gap,we propose a document-level information extraction method for scientific literature based on LLMs.Specifically,the method first constructs a document-level labeled dataset to train a binary classifier based on BERT,which determines the relevance of each sentence to the target entities.In the information extraction phase,specific prompt templates are used to guide LLMs in entity recognition and extraction,while performance is enhanced through fine-tuning with the Lo RA framework.Extensive experimental evaluations were conducted on two public datasets,covering zero-shot and supervised conditions.The results show that this method significantly improves the performance in document-level information extraction in the scientific field,surpassing traditional methods.展开更多
A new type of superconductive true random number generator (TRNG) based on a negative-inductance superconducting quantum interference device (nSQUID) is proposed. The entropy harnessed to generate random numbers comes...A new type of superconductive true random number generator (TRNG) based on a negative-inductance superconducting quantum interference device (nSQUID) is proposed. The entropy harnessed to generate random numbers comes from the phenomenon of symmetry breaking in the nSQUID. The experimental circuit is fabricated by the Nb-based lift-off process. Low-temperature tests of the circuit verify the basic function of the proposed TRNG. The frequency characteristics of the TRNG have been analyzed by simulation. The generation rate of random numbers is expected to achieve hundreds of megahertz to tens of gigahertz.展开更多
Depth migration can image complex structures with high accuracy,thereby stimulating the increasingly urgent demands for developing depth-domain inversions and interpretations in industry.The well-seismic calibration i...Depth migration can image complex structures with high accuracy,thereby stimulating the increasingly urgent demands for developing depth-domain inversions and interpretations in industry.The well-seismic calibration in the depth domain serves as a crucial cornerstone for these interpretations and inversions.Well data provide a partial cognition of underground media.Seismic data must be accurately calibrated with well data to expand this cognition outward.Depth-domain seismic data are non-stationary,transforming traditional,mature time-domain well calibration methods unsuitable for direct application to depth-domain seismic data.Therefore,researchers usually adopt a domain transformation strategy to complete well-seismic calibration in the time domain and then convert the calibration results into the depth domain.However,this method inevitably introduces additional error accumulation caused by domain transformation.On the basis of a comprehensive review of previous research,we propose a direct depth-domain well-seismic calibration method.This method is based on the synthesis of the depth-domain seismic records and the extraction of the depth-domain generalized seismic wavelets.We introduce constrained dynamic warping with maximum stretch depth constraint and directly match seismic data with well data in the depth domain.The actual processing results show that the method improves the efficiency of the depth-domain well-seismic calibration and produces a reliable relationship between seismic and well depths after two to four iterations.展开更多
文摘In an extraction turbine, pressure of the extracted steam and rotate speed of the rotor are two important controlled quantities. The traditional linear state feedback control method is not perfect enough to control the two quantities accurately because of existence of nonlinearity and coupling. A generalized minimum variance control method is studied for an extraction turbine. Firstly, a nonlinear mathematical model of the control system about the two quantities is transformed into a linear system with two white noises. Secondly, a generalized minimum variance control law is applied to the system. A comparative simulation is done. The simulation results indicate that precision and dynamic quality of the regulating system under the new control law are both better than those under the state feedback control law.
基金supported by the National Natural Science Foundation of China(No.62276026)。
文摘Identifying the main results of scientific papers is an important and challenging task.Unlike traditional named entity recognition(NER)in general domains,NER in the scientific field primarily aims to identify and classify specific scientific entity types from scientific literature.This requires handling professional terminology,complex sentence structures,and domain-specific contextual information.Recently,generation methods based on large language models(LLMs)have attempted to co-train models on multiple datasets while using prompt instructions to accurately extract target entities.However,there is currently insufficient exploration of document-level NER methods in the scientific field.To further fill this gap,we propose a document-level information extraction method for scientific literature based on LLMs.Specifically,the method first constructs a document-level labeled dataset to train a binary classifier based on BERT,which determines the relevance of each sentence to the target entities.In the information extraction phase,specific prompt templates are used to guide LLMs in entity recognition and extraction,while performance is enhanced through fine-tuning with the Lo RA framework.Extensive experimental evaluations were conducted on two public datasets,covering zero-shot and supervised conditions.The results show that this method significantly improves the performance in document-level information extraction in the scientific field,surpassing traditional methods.
基金Supported by the State Key Program for Basic Research of China under Grant No 2011CBA00304the National Natural Science Foundation of China under Grant No 60836001the Tsinghua University Initiative Scientific Research Program under Grant No 20131089314
文摘A new type of superconductive true random number generator (TRNG) based on a negative-inductance superconducting quantum interference device (nSQUID) is proposed. The entropy harnessed to generate random numbers comes from the phenomenon of symmetry breaking in the nSQUID. The experimental circuit is fabricated by the Nb-based lift-off process. Low-temperature tests of the circuit verify the basic function of the proposed TRNG. The frequency characteristics of the TRNG have been analyzed by simulation. The generation rate of random numbers is expected to achieve hundreds of megahertz to tens of gigahertz.
基金supported by the National Natural Science Foundation of China(No.U23B20158)CNOOC major technology project(KJGG2022-0104)。
文摘Depth migration can image complex structures with high accuracy,thereby stimulating the increasingly urgent demands for developing depth-domain inversions and interpretations in industry.The well-seismic calibration in the depth domain serves as a crucial cornerstone for these interpretations and inversions.Well data provide a partial cognition of underground media.Seismic data must be accurately calibrated with well data to expand this cognition outward.Depth-domain seismic data are non-stationary,transforming traditional,mature time-domain well calibration methods unsuitable for direct application to depth-domain seismic data.Therefore,researchers usually adopt a domain transformation strategy to complete well-seismic calibration in the time domain and then convert the calibration results into the depth domain.However,this method inevitably introduces additional error accumulation caused by domain transformation.On the basis of a comprehensive review of previous research,we propose a direct depth-domain well-seismic calibration method.This method is based on the synthesis of the depth-domain seismic records and the extraction of the depth-domain generalized seismic wavelets.We introduce constrained dynamic warping with maximum stretch depth constraint and directly match seismic data with well data in the depth domain.The actual processing results show that the method improves the efficiency of the depth-domain well-seismic calibration and produces a reliable relationship between seismic and well depths after two to four iterations.
