34 results about "Differential transformation" patented technology

The invention provides an LC-DSM algorithm (low-complexity differential detection algorithm). Firstly, a symbolic matrix is constructed at the transmitting end of a differential spatial modulation system by applying a bit filling scheme and a transmission bit calculation formula, and a receiving signal is obtained at the receiving end through differential transformation and transmitting. Then a search mode of detection by column is adopted on a receiving matrix based on the equation structure of a differential maximum likelihood detection algorithm to perform signal decision column by column so that a transmitting symbol and the sequence number of a transmitting antenna are obtained. Finally information bits are obtained through inverse mapping according to the corresponding relation of the bit and the transmitting symbol and the bit and the sequence number of the transmitting antenna. The method is obtained by combination of the low-complexity detection algorithm under spatial modulation and the equation structure formed by the maximum likelihood detection algorithm under the differential spatial modulation system. The research indicates that complexity is greatly reduced relative to the differential maximum likelihood detection algorithm by using the LC-DSM algorithm under the premise of guaranteeing no loss of algorithm performance.

The invention relates to a sequence overlapping differential direction protection method for a power transmission line with a static synchronous series compensator, and belongs to the technical field of relay protection of power systems. The method includes the steps that when a failure occurs in the power transmission line with the static synchronous series compensator, line mode voltage transient quantity ul and line mode current transient quantity il with fault phases are collected and calculated through a direction relay R1 at the measurement end M, four-order sequence overlapping differential transformation is conducted on ul and il, and then a voltage differential sequence Sul and a current differential sequence Sil are acquired; the power differential sequence SPl is acquired by multiplying the voltage differential sequence by the current differential sequence, and the direction of the failure is judged according to whether the polarity of a first nonzero mutation in the SPl is positive or negative; if the polarity is negative, the failure is judged as the forward failure, or else the failure is judged as the reverse failure. According to the method, the power differential sequence acquired by multiplying the voltage differential sequence by the current differential sequence, the direction of the failure is polarized, and judgment is reliable and effective.

This invention discloses a control method for PFC bus bar voltage-sharing of half-bridge type and its device, in which the positive and negative bus bar differential voltage is underground a proportion-integral-differential transformation, then the signal is directly superinposed on the duty ratio signal not to be front end of current loop. Even if the current loop exists, its variation would notinfluence the gain of equalizer ring, since the adjuster parameters of the internal current loop are adjusted independently and the parameters of the equalizer ring need to be adjusted once, and the equalizer ring would not influence current loop to generate little distortion to current order and THD of the current would not become large because of the qualizer ring.

The invention discloses a fast and real-time video coding and decoding compression algorithm. The coding algorithm comprises the following steps of S01, inputting a bitmap image sequence, and performing the brightness color difference signal transformation on the bitmap image sequence; S02, performing the differential transformation on the bitmap image sequence after the brightness color difference signal transformation process; S03, performing the fast frequency domain transformation on the bitmap image sequence after the differential transformation process; S04, performing the rapid quantization on the bitmap image sequence after the fast frequency domain transformation process; S05, conducting the entropy compression on the fast quantized bitmap image sequence, and outputting the coded data of a compressed video. In addition, the invention discloses the specific steps of the decoding algorithm. According to the invention, the information of a high-definition image can be compressed at about 8-16 megabytes per second based on software. In this way, the transmission and storage problem during the video real-time coding and decoding process of a video processing host and partly embedded equipment without special hardware processing in the prior art is solved. The method can be widely applied to the fields of remote control, remote monitoring and the like for computer equipment.

The invention provides a method and system for calibrating the pose error of a mechanical arm. The method includes: performing equivalent differential transformation on the pose error of the tail endof the mechanical arm to obtain an equivalent differential motion vector, decomposing the equivalent differential motion vector into the equivalent differential motion vector of each joint shaft, during independent motion, of the mechanical arm, and performing calibration and decoupling transformation on the pose error of each joint shaft, during independent rotation sampling, of the mechanical arm to calibrate the pose error of the tail end of the mechanical arm. The method and system has the advantages that equivalent different transformation description is performed on the pose error at thetail end of the mechanical arm to obtain the equivalent differential motion vector, the equivalent differential motion vector is decomposed into the equivalent differential motion vector of each connecting rod coordinate system, and decoupling calculation and transformation are performed on the connecting rod coordinate systems and joint motion coupling relations to achieve high-precision and high-reliability synchronous calibration of the absolute pose error of the mechanical arm in a full working space range.

The invention relates to a distribution network single-phase ground protection method based on zero-sequence current sequential overlapping derivative transformation direction, belonging to the technical field of power system relay protection. For a resonant earthed system, when single-phase earth fault occurs on one feed line, firstly, a bus zero-sequence voltage in a short time window and zero-sequence currents of feed lines after the fault are intercepted and subjected to sequential overlapping derivative transformation respectively; then, a bus zero-sequence voltage sequential overlapping derivative transformation result is multiplied by zero-sequence current sequential overlapping derivative transformation results respectively to obtain a series of wave fronts on a time axis; finally, the amplitude of the first wave front in the series of wave fronts is used as a basis for protective start; when the amplitude of the first wave front is less than zero to determine positive-direction fault occurs on the feed line, a protective element acts; when the amplitude of the first wave front is larger than zero to determine reverse-direction fault occurs on the feed line, and the protective element does not act. The distribution network single-phase ground protection method is low in using difficulty and high in protection reliability.

The invention discloses a space Jacobian matrix construction method for robot parameter identification, including S100: constructing a robot kinematics model; S200: analyzing the differential motion characteristics of robot joints based on the robot kinematics model, and establishing a coordinate system in the case of differential motion The homogeneous transformation matrix between them; S300: Assuming that a joint coordinate system undergoes differential motion, on this basis, establish a virtual coordinate system corresponding to the joint, and construct a transformation matrix between the two, and then calculate the robot end relative to the base The actual pose of the coordinate system is compared with the original theoretical pose to obtain the pose error of the robot end relative to the base coordinate system caused by the motion error of the joint coordinate system, and construct the space Jacobian matrix of the robot. The method of the present invention is a robot space Jacobian matrix construction method based on the differential transformation principle and the virtual coordinate system method, has lower time consumption, and can quickly construct the robot space Jacobian matrix.