Abstract: A load cell with a Roberval structure having an anti-offset function, and a weighing device are disclosed. The load cell includes strain detection elements mounted on an elastic element which includes a loading portion for receiving a load from a connector, a fixing portion for fixing the elastic element, a strain generation portion for converting the load into deformation, and a parallel guide beam for transferring the load and keeping the loading portion translational when loaded. The strain generation portion includes at least one hollow cavity body. The loading portion and the fixing portion are respectively located on left and right sides of the strain generation portion. The parallel guide beam is located at upper and lower portions of the strain generation portion A center of rotation and a geometric center of the strain generation portion coincide when the load cell with the Roberval structure is loaded.

Abstract: Interferometer displacement measurement system and method are disclosed, wherein, a measurement light is processed by a first polarization splitting prism, a first ¼ wave plate, a first splitting prism, an optical waveguide component and a reflector, and then is returned to a first photoeletric detector and a second photoeletric detector. The reference light is processed by a second polarization splitting prism, a second ¼ wave plate, a second splitting prism and a reflecting mirror, and then is returned to the first photoeletric detector and the second photoeletric detector. The first photoeletric detector forms a measurement signal according to the processed measurement light and the processed reference light, and the second photoeletric detector forms a reference signal according to the processed measurement light and the processed reference light. Displacement information of the object to be detected is determined according to the measurement signal and the reference signal.

Abstract: Techniques that facilitate machine learning optimization are provided. In one example, a system includes a computational resource component, a batch interval component, and a machine learning component. The computational resource component collects computational resource data associated with a group of computing devices that performs a machine learning process. The batch interval component determines, based on the computational resource data, batch interval data indicative of a time interval to collect data for the machine learning process. The machine learning component provides the batch interval data to the group of computing devices to facilitate execution of the machine learning process based on the batch interval data.

Abstract: A strain gauge (12, 21A, 21B, 25A, 25B, 31, 35, 41, 45) and method of manufacturing a strain gauge (12, 21A, 21B, 25A, 25B, 31, 41, 35, 45) against moisture penetration comprises or includes the step of producing a coated base or cover layer (14, 34, 44) by forming a moisture barrier coating (17) on the surface the latter.

Abstract: The present invention provides a method for determining the prognosis of cancer in a subject. The method comprises measuring the amount of megakaryocytes in a sample from the subject. Usually, the sample is a blood sample. The method may also comprise measuring the number of circulating tumour cells (CTCs) in the sample, and in some embodiments a comparison of the number of megakaryocytes and CTCs in the sample. The present invention also provides methods of treatment for cancer in a patient for whom a poor prognosis is predicted using a method of prognosis of the invention.

Abstract: A neutron-capture therapy system includes a charged particle beam generation unit, a beam transmission unit and a neutron beam generation unit. The charged particle beam generation unit includes an ion source and an accelerator. The accelerator accelerates charged particles generated by the ion source, so as to obtain a charged particle beam of the required energy. The neutron beam generation unit includes a target, a beam shaping body and a collimator. The charged particle beam irradiates onto the target through the beam transmission unit to generate neutrons, which sequentially pass through the beam shaping body and the collimator to form a neutron beam for therapy. The neutron-capture therapy system is accommodated in a concrete building including an irradiation room, an accelerator chamber and a beam transmission chamber. The neutron beam generation unit is at least partially accommodated in a partition wall of the irradiation chamber and the beam transmission chamber.

Abstract: Some embodiments may be associated with a cloud computing environment. A computer processor of a traffic prediction server may retrieve performance stack trace logs from a traffic performance stack trace log repository that stores traffic information of the cloud computing environment. The traffic prediction server parses the performance stack trace logs as an objects list including parent/child object relationships and stores the parsed objects list in a graph database. The traffic prediction server may then transform the graph database into training data including spatial and temporal information and use the transformed training data to train a transformer model. According to some embodiments, the traffic prediction server also provides previous traffic input data to the transformer model when generates predicted traffic output data based on the previous traffic input data (e.g., to facilitate cloud load management).

Abstract: Methods, systems, and computer-readable storage media for a multimodal machine learning image and text combined search method. One example method includes processing items that each have an associated image and a textual description. A first image feature vector is generated by processing a first image using a first machine learning model. A first textual feature vector is generated by processing a first textual description using a second machine learning model. The first image feature vector and the first textual feature vector are combined to generate a first combined feature vector for a first item. Similarity lists of similar items are generated for the first item based on similarities between the first image feature vector, the first text feature vector, the first combined feature vector and respective corresponding vectors of other items. The similarity lists for the first item are combined to generate a combined similarity list for the first item.

Abstract: Various examples are directed to systems and methods for characterizing natural language text units. A plurality of text units may be used to train a bidirectional model. A bidirectional model may be applied to a set of annotated text units to generate a plurality of span context vectors. The plurality of span context vectors may be used to train a span prediction model. The span prediction model may be applied to at least a portion of the plurality of text units to generate a plurality of span characterizations, a first span characterization corresponding to a first span indicating that the first span describes a first job skill.

Abstract: A 5G distributed system includes a near-end unit, a remote device, and an active antenna unit. The near-end unit transmits signals of multiple modes including 5G signal to the remote device. The remote device performs frequency conversion processing on 5G signal and does not perform frequency conversion processing on signals of other modes, such as 4G, and then outputs to the active antenna unit for coverage. In the present disclosure, the 5G signal is converted by frequency conversion at the remote device, and signals of other modes other than 5G signal are transmitted to the remote active antenna unit without being processed by frequency conversion for signal coverage, which satisfies the coverage requirements of 5G indoor distribution.

Abstract: An assemblable modular and disc coreless permanent magnet superconducting ultra-efficient motor comprises a junction box, a housing, bearings, a rotating shaft, rotors, a stator, and air gaps; the housing comprises a left and a right end cover, the rotors comprise a left and right rotor; both left and right rotor comprise 2N pairs of left or right rotor poles and rotor yokes; the stator comprises a stator armature disc and stator windings, the bearings comprise a left and right bearing, and the air gaps comprise a left and right air gap. The rotor's magnetic field is replaced by high-temperature resistant rare earth high-intensity permanent magnets, and the rotor's electrically excited magnetic field is replaced by permanent magnets, eliminating rotor's copper and iron losses. The winding superconducting wires eliminate stator's copper loss, and stator iron core removal eliminates stator's iron and magnetic saturation loss, improving the efficiency and torque output.

Abstract: A MEMS gas sensor (A) and array (B) thereof, a gas detection and preparation method. The gas sensor (A) comprises a first substrate (A2) with a cavity (A1) provided in a first surface, and a gas detection assembly (A3) arranged at an opening of the cavity The gas detection assembly comprises: a supporting suspension bridge (A31) erected on the opening of the cavity, and a gas detection part (A32) arranged on the supporting suspension bridge. The gas detection part comprises a strip-shaped heating electrode part (A321), an insulating layer (A322), a strip-shaped detection electrode part (A323) and a gas-sensitive material part (A324), which are sequentially stacked. The strip-shaped detection electrode part comprises a first detection electrode part (A323-1) and a second detection electrode part (A323-2), with a first opening (A325) provided between the A323-1 and A323-2; the gas-sensitive material part is arranged at the position of the first opening.

Abstract: A unified open-circuit fault-tolerant control method for a vector control (VC) drive system and a direct torque control (DTC) drive system of a five-phase permanent magnet fault-tolerant motor are provided. The control method adopts a unified open-circuit fault-tolerant control strategy. The unified open-circuit fault-tolerant control strategy includes: obtaining a predetermined torque, obtaining predetermined direct-axis and quadrature-axis voltages, analyzing a fault-tolerant mechanism to obtain fault-tolerant currents, obtaining winding phase voltages in a fault mode based on the fault-tolerant mechanism, and obtaining fault-tolerant voltages based on a back-electromagnetic force (EMF). The unified open-circuit fault-tolerant control strategy suitable for the VC drive system and the DTC drive system is proposed based on chaotic pulse width modulation (CPWM).

Abstract: Provided are tuning circuit and communication device. The tuning circuit includes tuning switch, ground patch point, and at least two sets of patch units. Each set of patch units at least includes a first patch point and a second patch point; the tuning switch comprises a standard port and at least one conversion port, and each set of patch units is connected to the corresponding conversion port; a first end of a first patch point and a first end of a second patch point in the same set of patch units are connected to the same conversion port; second ends of first patch points in different patch units are respectively connected to different contact points of an antenna; the second ends of all the second patch points are connected to the ground; and both ends of the ground patch point are respectively connected to the standard port and the ground.

Abstract: A system comprises one or more processors. The system further comprises memory having computer readable instructions stored thereon that, when executed by the one or more processors, cause the system to: obtain a reference tree of nodes and linkages for a reference three dimensional volumetric representation of a branched anatomical formation in a reference state; based on the reference tree, generate a reference three dimensional geometric model; based on measurements from a sensor system, detect deformation of the branched anatomical formation; based on the detected deformation, obtain a deformed tree of nodes and linkages; and generate a three dimensional deformation field that represents the detected deformation of the branched anatomical formation.

Abstract: A method and device for providing real-time data service are disclosed. The method includes creating a message engine for receiving real-time data generated by a real-time data source, creating a data processing engine for acquiring the real-time data in the message engine and processing the real-time data to obtain target data, creating a storage engine for storing the target data and providing the target data according to requirement of a real-time data service, and containerizing the message engine, the data processing engine and the storage engine, and constructing a real-time data processing architecture with a containerized message engine, a containerized data processing engine and a containerized storage engine to provide the real-time data service. Embodiments of this disclosure can realize an overall architecture of real-time data reception, real-time data processing, real-time data storage, and real-time data service.

Abstract: Constructing a feature segment-based ensemble can include generating a data structure for each element of an initial set of training data. Multiple strongly correlated features of the elements can be identified as well as weakly correlated features. For each strongly correlated feature, a feature segmentation training set can be generated, each training set's elements each containing one of the strongly correlated features and excluding other strongly correlated features. One or more machine learning algorithms can be selected from a software library. The one or more machine learning algorithms can be applied to the feature segmentation training sets to train multiple machine learning models. Each machine learning model that improves the predictive accuracy of the feature segment-based ensemble can be integrated in the feature segment-based ensemble.

Abstract: A conductive terminal, a conductive terminal assembly and a connector are provided. The conductive terminal assembly includes a casing and at least one row of conductive terminals. The at least one row of conductive terminals includes a plurality of conductive terminals arranged side by side. Each conductive terminal includes a contact arm, a contact pin, an extension part and a tenon structure. The contact arm extends along a first direction, the contact pin extends along a second direction, and the first direction is perpendicular to the second direction. The extension part is connected between the contact arm and the contact pin, and the extension part is bent and extended relative to the contact pin. The tenon structure is connected to one of the contact arm and the contact pin. The tenon structure extends in parallel along the first direction or the second direction.

Abstract: Systems and methods are presented for providing a response to a user query. Reception of a user query is detected. An augmentation machine learning model is utilized to determine one or more variations of the user query that correspond to a semantic meaning of the user query. A plurality of response candidates is determined that correspond to the user query by comparing the user query and the one or more variations of the user query to a plurality of documents. A final response candidate is determined from the plurality of response candidates based on utilizing a semantic machine learning model to perform a semantic comparison between the plurality of response candidates and at least the user query.

Abstract: The present disclosure provides an electronic device and a method for controlling data updates of the electronic device, the electronic device comprising: a control portion including a storage configured to store data and a controller configured to control data updates in the storage; a power supply portion; a power storage portion configured to be connected with the power supply portion, receive and store electric power of the power supply portion, and supply the electric power to the control portion; and a power supply detection portion configured to detect whether the power supply portion supplies power normally, the storage includes a first storage portion configured to store start-up data for electronic device start-up and a second storage portion configured to store running data for electronic device running after start-up; in the process of updating the start-up data of the first storage portion, if the power supply detection portion detects the power supply portion has a power supply abnormality, the