83results about How to "Without loss of accuracy" patented technology

A quality control accessory (QC accessory) is provided which is adapted for use in linear accelerator quality control or in verification of an arbitrary isocentric radiation treatment plan of a patient or radiation sensitive body. The accessory includes an absorber, a detector and an optional orientation device. The absorber is rotationally symmetric, preferably spherical or hemispherical. The detector is adapted to be fixed in a stationary spatial relationship with respect to the absorber. The optional orientation device is adapted to maintain the two dimensional detector (2d detector) and the absorber in a fixed relative spatial relationship with respect to the beam focus of the linear accelerator, when the gantry is rotated, so that the central axis of the beam is essentially orthogonal to the 2d detector and the phantom axis of symmetry is parallel to or aligned with the central axis of the gantry rotation axis.

Provided are a tunnel geology comprehensive advanced forecasting expert system and an implementation method thereof. The system comprises a tunnel geology comprehensive advanced forecasting expert main module, a tunnel hole inner geology survey device, a tunnel hole inner geophysical prospecting device and an advanced horizontal drilling device. The tunnel geology comprehensive advanced forecasting expert main module is connected with the tunnel hole inner geology survey device, the tunnel hole inner geophysical prospecting device and the advanced horizontal drilling device respectively through data reading and outputting interfaces. The tunnel geology comprehensive advanced forecasting expert main module comprises a tunnel geology macroscopic forecasting sub-module, a tunnel hole inner geology survey analysis sub-module, a geophysical prospecting method selecting and detecting result analysis sub-module, a tunnel face front unfavorable geology forecasting sub-module, an advanced horizontal drilling and result analysis sub-module and a generating processing measurement sub-module. According to the system and method, a tunnel geology advanced forecasting worker can be helped to carry out tunnel geology forecasting working, and accuracy of forecasting is not lowered.

A high-speed measuring electronic digital outside micrometer allowing a user to make measurements quickly and effectively compared to conventional electronic outside micrometers with rotary thread spindles, which have limited measuring speed. This high-speed measuring electronic digital outside micrometer features: a linear motion digital sensor, a linear scale digital sensor, a linear motion digital sensor base, a digital display unit, an electronic processing unit, a micrometer body, a linear spindle, a linear spindle movement unit, a spindle control unit, a micrometer anvil, and retractable springs. In a stationary position, the measuring surface of the micrometer anvil and the linear spindle are in contact to the other. By operating the spindle control unit, it will create distance between the micrometer anvil and the linear spindle measuring surface. This distance is a measurable distance of the micrometer. By set free of the spindle control unit, the retractable springs will re-set the micrometer to its stationary position.

The invention discloses a method for distributed type cognitive radio network routing based on cross-layer design, which belongs to the technical field of cognitive radio network. The method comprises the following steps: establishing a coloring multi-graph model according to useable frequency spectrum information acquired by node perception; transferring a cross-layer design routing algorithm to search the routing between a source node and a target node, and simultaneously selecting communication signal channels between adjacent nodes; and updating the topology of coloring multi-graph and interface number of the nodes. The routing method adopts the cross-layer design on the routing selection of a network layer and an MAC layer, and concretely optimizes routing hop number and adjacent hop interfere under the condition of lower time complexity. The method is applicable to radio network, next generation heterogeneous network and so on.

The invention relates to an image recognition method and device. The method comprises the following steps that: the features of a target image are extracted; dimensionality reduction is performed on the features of the target image according to a principal component analysis method; and the target image which has been subjected to dimensionality reduction is recognized according to a pre-trained image model. With the image recognition method adopted, a terminal can perform dimensionality reduction on the features of the target image according to the principal component analysis method, and therefore, the number of the features of the target image can be decreased, and the key features of the target image can be reserved; the terminal can recognize the target image which has been subjected to dimensionality reduction according to the pre-trained image model; and therefore, the efficiency of image recognition can be improved, and the accuracy of image recognition will not be reduced, and the use experience of a user can be improved.

The invention discloses a PCB secondary and multiple accurate punching method based on visual tracking. After the image information of the PCB is obtained through the vision system, the image information is subjected to binaryzation through a cvThreshold function; calculating a communication domain which exists on the processed image by utilizing a FindBlobs function, and filtering the communication domain which is smaller than the area; accurately positioning the edge by means of a Zernike moment edge operator in the possible edge points after all possible edge points are quickly found through a Sobel operator, and calculating to obtain the center position coordinates of the Mark points or the prefabricated hole according to the edge coordinates. The data volume of the points needing to be calculated is greatly reduced, and the speed of determining the center coordinates is increased, and the accuracy is not lost; by adopting a multi-time punching mode, the center coordinates of the prefabricated hole are positioned again during each punching, and the deviation problem possibly caused by single positioning is corrected, the drill bit is a grinding process of the edge of the prefabricated hole punched previously during multiple times of punching, and the precision is far higher than that of the single punching.

The invention discloses an LLVM-based automatic performance prediction method for a scientific calculation program, and belongs to the technical field of program performance prediction. The automatic performance prediction method aims at achieving automatic analysis of the scientific calculation program and improving the accuracy of static analysis, and meanwhile program prediction time can be given finally. The automatic performance prediction method is characterized by including the steps of converting the predicted source program into an intermediate bitcode in an LLVM; analyzing the intermediate bitcode to obtain a calling time number, a circulating time number and a static branch probability for identifying a communication command MPI; carrying out mixed pile penetrating on the intermediate bitcode; carrying out code deletion on the intermediate bitcode obtained after mixed pile penetrating to carry out optimizing; running the optimized intermediate bitcode to obtain llvmprof.out files; analyzing the llvmprof.out files, and predicting execution time in cooperation with command time. The automatic performance prediction method is suitable for predicting the performance of the scientific calculation program.

The invention relates to a hydraulic clutch and a method for determining an adaptive clutch fill volume or return spring pressure. The application discloses a method for learning a characteristic filling volume of a hydraulic clutch (22). The method comprises iterative steps of applying a pressure pulse to the hydraulic clutch (22) when the clutch is in a disengaged state and determining an inflection event at an input or at an output of a torque path in which the hydraulic clutch (22) is situated. A characteristic filling volume of the hydraulic clutch (22) is derived from the determined inflection event. Furthermore, the application discloses a method for learning a characteristic return spring pressure of the hydraulic clutch (22), for engaging the hydraulic clutch (22) and corresponding devices for carrying out the methods according to the application.

The invention relates to a fine-grained image recognition method, a convolutional neural network and a training method thereof, computer equipment and a computer readable storage medium. The method comprises the steps of obtaining an original image; extracting global features of the original image by adopting a convolutional network; determining a plurality of candidate regions in the original image, and determining a feature value of each candidate region according to the global feature; sorting the plurality of feature values to obtain a first sorting result, and determining at least one candidate region corresponding to the maximum N feature values in the first sorting result; extracting features of the at least one candidate region to obtain at least one local region feature, each candidate region corresponding to one local region feature; cascading the global feature and the at least one local region feature to obtain a cascading feature; classifying the original images accordingto the cascade features to obtain the classification result, so that the problems of complexity and high labeling cost of a fine-grained image recognition method are solved, and the fine-grained imagerecognition method is simplified.

The invention provides a parawing four-freedom-degree semi-physical simulation platform which comprises a casing body, a data acquisition system, a motion control system, an umbrella-mounted control system and a model state solving system. The data acquisition system acquires the position information of the simulation platform. The umbrella-mounted control system receives the position information, and acquires the corresponding control amount according to a homing control algorithm. The model state solving system receives the control amount, receives an interference instruction from an interference device, decodes a status instruction and feeds back the status instruction to the umbrella-mounted control system to control the motion control system. The platform moves based on different control instructions, and can simulate action caused by a wing umbrella system for different trailing edge parachute cord manipulated variables in actual air-drop. The data acquisition system and the umbrella-mounted control system form closed-loop control. The control effect is visually reflected in the motion law of the platform, and the position and attitude can be measured accurately. The effectiveness of the control algorithm can be verified before air-drop, which greatly reduces the experimental cost and the experimental risk.

The invention provides a federated learning privacy protection system and method based on hierarchical aggregation and a block chain. The system comprises a trusted mechanism, the block chain, a federated learning module and an IPFS distributed storage system. The block chain is used for recording various data generated by the federal learning module and model addresses corresponding to training models of various versions iterated all the time in the training process; the federated learning module is used for realizing federated learning training; the user terminal obtains a current global model as a local model, and trains the local model on a local data set to obtain a new round of local model; and the IPFS distributed storage system is used for storing local model parameters and global model parameters. Based on the system provided by the invention, the technical problems of low efficiency, reduced model accuracy and large storage overhead of a federal learning privacy protection mechanism in the prior art are solved, the federal learning efficiency can be improved, and the global model accuracy can be improved.

The invention relates to a power injection model of a power flow controller under a direct current power flow model. In a direct current power flow, an effect of the power flow controller on a system is equivalent to voltages of nodes i and j and power injection equivalence is performed so as to acquire the power injection model under the direct current power flow; shift-to power of the nodes i and j is defined in the description, and xij is an impedance between the nodes i and j. The model and the method have advantages that the method provided in the invention is used to carry out power flow controller addressing so that an addressing speed can be increased and simultaneously accuracy can be achieved; and the model and the method are especially suitable for a UPFC addressing problem of a large power grid.

The invention relates to an indicator and a method for cable core correction and transformer tripping outlet correction, and belongs to the technical field of power supply maintenance. According to the technical scheme, an internal detection circuit (102) is formed by parallelly connecting N identical relay action keeping modules, the relay action keeping modules are independent from each other, and each relay action keeping module comprises a relay action part and a switching control keeping part; peripheral equipment comprises a power input part (101) and an operation indicating part (103); and the operation indicating part comprises an operation panel with N pairs of sockets and an indicating lamp, wherein the pair quantity N of the sockets is matched with the quantity N of the relay action keeping modules, and the operation indicating part is used for wiring and indicating the closure conditions of contacts after relays act. By adopting the indicator and the method, the turn-on and turn-off condition of a cable core and the insulation condition of other wire cores can be directly learnt; in tripping matrix test, the action result can be intuitively and accurately judged by viewing the indicating lamp of the external indicating panel; and the indicator is easy to operate, convenient to use and quick and accurate in detection and correction.

The invention provides a method for predicting a horizontal electromagnetic scattering mean of a ship loaded with a cylindrical whip antenna. The method comprises the following steps: 1), according to geometric information of the cylindrical whip antenna loaded on the ship, computing single station RCS (radar cross section) values in the horizontal direction of the cylindrical whip antenna by a physical optical estimating formula; 2), establishing a geometrical model of the ship except the cylindrical whip antenna, and computing the mean of the single station RCS values in the horizontal direction of the geometrical model by a conventional ship electromagnetic scattering predicting method; and 3), superposing an RCS algebraic sum of the single station RCS values in the horizontal direction of the cylindrical whip antenna and the mean of the single station RCS values in the horizontal direction of the geometrical model of the ship except the cylindrical whip antenna by adopting a random phase summing method, so as to obtain the horizontal electromagnetic scattering mean of the ship loaded with the cylindrical whip antenna. The cylindrical whip antenna and the other components of the ship are computed separately, so that the grid number and the corresponding computation amount are maintained on a magnitude when the whip antenna is not loaded, and the difficulty that computation cannot be carried out by a planar element method can be avoided.

The invention provides a user defined simulation method for an electrical power system based on mixed variable step size stimulation. A user defined function of an analyzing packaged program of the electrical power system is optimized through the pretreatment, initialization and stimulation calculation of a user defined model; the redundancy of the user defined model is reduced when the dynamic process of the system is stable; the calculation speed is increased while the acceptable accuracy of the engineering calculation is not lost; the fine accurate calculation is performed in the to-be-concerned period; the mixed step size stimulation function in the electromechanical-electromagnetic transient and electromechanical-medium and long term processes of the analyzing packaged program of the electrical power system is achieved; the variable step size stimulation is realized to some extent; the calculation convergence, accuracy and calculation efficiency of the analyzing packaged program of the electrical power system are increased; the use scope of the analyzing packaged program of the electrical power system is widened.

The invention discloses a signal delay elimination method for a silver self-powered detector based on H infinity filtering, which comprises the following steps: 1)establishing a nuclear reaction model of silver and thermal neutron; 2)employing decoupling transformation to establish a discrete state equation corresponding to the nuclear reaction model; 3)determining the instantaneous response share of the current of the silver self-powered detector; and 4)using a H infinity filter for delay elimination of the current signal of the silver self-powered detector. The method is capable of performing delay elimination on the current signal of the silver self-powered detector, noise can be effectively inhibited, normal usage of the silver self-powered detector can be carried out under transient condition of a reactor, due to the H infinity filter, prediction of the statistics characteristic of external disturbance input signal is not required while delay elimination is carried out.

The invention discloses a rhodium self-powered detector signal delaying eliminating method based on IIR filtering. The method comprises the following steps: step 1. establishing a rhodium reaction model of rhodium and thermal neutrons; step 2. establishing a rhodium self-powered detector signal system model; step 3. establishing a delaying eliminating module and a noise inhibition module respectively; and step 4. finishing a parameter design of a delaying eliminating system before the delaying eliminating system is operated, and operating the delaying eliminating system in real time to realize the delaying elimination of a signal. The rhodium self-powered detector signal delaying eliminating method has a whole simple procedure; and the signal delaying elimination and the noise inhibition are separated so that the method is convenient to realize, popularize and apply. By virtue of the rhodium self-powered detector signal delaying eliminating method, the delaying eliminating problem of the signal of a rhodium self-powered neutron detector in a reactor is solved; and the rhodium self-powered detector current signal is directly used for a subsequent link of an advanced reactor core measurement system, and the accuracy is not lost.

The invention provides a scrambling appraisal system used for appraisal or study, comprising a scrambler module and a terminal presenting a scrambling result, wherein the scrambler module can scramble a test question sequence and/ or a choice question option sequence, and the terminal presenting the scrambling result is used for providing a scrambled test question information to a person who participates in appraisal or studies. In this way, the achieved effects are as follows: actually, the same test paper can derive multiple test papers with different answers by scrambling, thus avoiding cheating phenomena, and the invention has the important meaning for purifying examination room culture and embodying the righteousness and fairness of an examination.

A stabilizer device (100) for a firearm, apt to control the movement performed by the firearm barrel (11) in the instant in which a bullet is ejected, comprising a support element (1) for the firearm barrel (11), removably fixed or fixable to the barrel (11) and apt to be made slidable along a main direction of extension of the barrel (11) with respect to the stock (13) of the firearm that can be gripped by a shooter, and control means for controlling the sliding motion of the support element (1) with respect to the stock (13) of the firearm, having a first (14; 140) and a second (15; 150) control element constrained or constrainable respectively to the support element (1) and to the stock (13) and apt to cooperate to limit and control the extent of such sliding.

The invention discloses a method and a device for estimating a channel. The method comprises the following steps of: presetting extracting multiple of at least two stages, and dynamically determining the current extracting multiple according to link quality; according to the current extracting multiple, extracting a received frequency domain demodulation reference mark and a local frequency domain demodulation reference mark; and figuring out the current frequency domain chain estimated value by using the extracted received frequency domain demodulation reference mark and the local frequency domain demodulation reference mark. The device comprises an extracting multiple judging module, a pilot frequency extraction module and a channel estimation module, wherein the extracting multiple judging module is used for presetting the extracting multiple of the at least two stages, and dynamically determining the current extracting multiple according to the link quality; the pilot frequency extraction module is used for extracting the received frequency domain demodulation reference mark and the local frequency domain demodulation reference mark according to the current extracting multiple; and the channel estimation module is used for figuring out the current frequency domain chain estimated value through the extracted received frequency domain demodulation reference mark and the local frequency domain demodulation reference mark. The method and the device have the advantages of reducing the complexity of a receiver and reducing the processing time delay.

The invention discloses a delay elimination method of a signal of a rhodium self-powered detector based on Luenberger form H2/H infinity hybrid filtering, which comprises the following steps: 1)establishing a nuclear reaction model of rhodium and thermal neutron; 2)employing decoupling transformation to establish a discrete state equation corresponding to the nuclear reaction model; 3)determining the instantaneous response share of the current of the rhodium self-powered detector; and 4)using a H infinity filter for delay elimination of the current signal of the rhodium self-powered detector. The method is capable of performing delay elimination on the current signal of the rhodium self-powered detector, noise can be effectively inhibited, normal usage of the rhodium self-powered detector can be carried out under transient condition of a reactor. The method requires that error variance of filtering corresponding to a measured error channel has an upper bound, so that when the input signal is an indeterminate signal possessing limit energy, the normal application of the rhodium self-powered detector can be carried out.

The invention provides a picture character recognition method. The picture character recognition method comprises the following steps of: reading a text area, on a picture, frame-selected by a user in a current picture form; respectively selecting each pixel character in the frame-selected text area; comparing the first pixel character in the frame-selected text area with each matrix character in matrix libraries corresponding to various conventional fonts so as to find matrix characters with the highest similarity; and recording the matrix libraries of the matrix characters with the highest similarity, defining the matrix libraries as standard matrix libraries, respectively finding the matrix characters with the highest similarity in the standard matrix libraries, and outputting the matrix characters with the highest similarity in an editable character form. According to the method, the character information in pictures can be correctly and efficiently recognized.