16results about How to "The implementation process is simple" patented technology

The application provides a wind tunnel pressure regulating valve opening set fault diagnosis method based on a composite criterion, relates to the technical field of equipment fault diagnosis, and comprises the following steps: training a fault diagnosis model by using seen fault sample, extracting sample features by using an encoder, and obtaining classification information and reconstruction information of the sample by using a classifier and a decoder; then, according to the output result of the model on training data, a composite criterion is constructed in a feature space, a probability space and a sample space, and is used for identifying unknown faults; finally, when a new fault sample is input, the composite criterion is used to judge the sample, accurate diagnosis of known faults and effective identification of unknown faults are realized. The composite criterion method has a clear structure, a simple implementation process, can be extended on the basis of an existing fault diagnosis model, and has good engineering application value.

The application provides a method for observing rotor flux of an alternating current motor, and belongs to the technical field of motor control. The method comprises the following steps: calculating rotor observation flux based on a first voltage-type flux calculation model and stator voltage and stator current in a two-phase static coordinate system; obtaining an amplitude deviation by subtracting the amplitude of the rotor rated flux from the amplitude of the rotor observation flux, and obtaining a deviation compensation quantity by multiplying the amplitude deviation by a proportional integral operation; and compensating and correcting the first voltage-type flux calculation model based on the deviation compensation quantity to obtain a voltage-type flux calculation model with compensation correction, so that the amplitude of the rotor observation flux converges. The application converges the rotor observation flux to the rotor given flux by connecting a correction link based on the traditional voltage-type flux calculation model, and finally realizes accurate flux position observation.

The invention relates to a multi-task face sensing method and device, electronic equipment, a storage medium and a computer program product. The multi-task face sensing method comprises the steps that a to-be-detected face picture is acquired; a to-be-detected face picture is input into the multi-task face perception model, face features and positioning reference points are extracted from the face picture through the multi-task face perception model, and the positioning reference points are used for executing a positioning task about face key point detection; semantic reference points are calculated based on the positioning reference points through a multi-task face perception model, and the semantic reference points are used for executing semantic tasks about face state detection; and executing a positioning task and a semantic task based on the face features, the positioning reference points and the semantic reference points through a multi-task face perception model. Therefore, by explicitly utilizing the physical representation of the face key points, effective supervision for the semantic reference points is realized, the guiding effect of the face key points on the face perception task is enhanced, and the prediction effect of the face perception model can be improved.

The invention discloses a method for rapidly detecting the rolling oil acid value in emulsified liquid for cold rolling production, which comprises the following steps: S1, taking the cold rolling emulsified liquid to be detected, filtering, and then sequentially carrying out organic solvent mixing and extraction, high-speed centrifugal separation and negative pressure rotary evaporation operation on the cold rolling emulsified liquid to be detected, so as to obtain a rolling oil sample to be detected; s2, preparing rolling oil standard samples with different acid values; and S3, respectively detecting the rolling oil standard samples with different acid values and the rolling oil sample to be detected by utilizing an infrared spectroscopy method to obtain infrared spectrums of the corresponding samples so as to obtain a standard curve of the acid value of the rolling oil, and comparing to obtain the acid value of the rolling oil in the cold rolling emulsion to be detected. According to the method, the acid value of the rolling oil in the on-site emulsion can be quickly read and obtained, the result accuracy is high, the requirements for quick measurement and quick adjustment of on-site production of a modern cold rolling unit can be met, and high practicability is achieved.

The application discloses a model-driven neural network-based offset min-sum LDPC code decoding algorithm and relates to an offset min-sum LDPC code decoding algorithm. The application is designed to solve the problems of the traditional LDPC decoding algorithm in convergence speed, calculation complexity, flexibility and the like. The application comprises the following steps: 1, a model-driven offset min-sum algorithm network model is constructed, wherein the network model comprises an input layer, a check node layer, a variable node layer and an output layer; 2, the input layer is used to calculate the received channel LLR by using a received signal y to perform initialization; 3, the check node layer is used to calculate the CN-to-VN LLR value; 4, the variable node layer is used to calculate the VN-to-CN LLR value; and 5, the output layer is used to calculate a final decoding output. The application belongs to the technical field of decoding.

The present application relates to the technical field of unmanned aerial vehicle, and especially relates to a detection method and device for external parameters of unmanned aerial vehicle. The method comprises the following steps: determining a first description value of a bearing surface according to attitude data of the unmanned aerial vehicle when the unmanned aerial vehicle is static on the bearing surface; determining a second description value of the bearing surface according to camera data of the bearing surface shot by the unmanned aerial vehicle in a flight state; and determining an external parameter error of the unmanned aerial vehicle according to the first description value and the second description value. The embodiment of the present application can realize online detection of the external parameters of the unmanned aerial vehicle more simply, and the external parameter error is detected depending on fixed environmental information such as the bearing surface, so that the detection result has higher robustness.

The application discloses a kind of generation methods of customized software system.Compiles complete embedded software system, and each software module generates static library file in output directory.For current customer, customer-level open source software module and customer-level closed source software module are divided.New target path is established.For customer-level open source software module, customer-level closed source software module and system-level closed source software module, different contents are copied into target path.In the subdirectory of each customer-level closed source software module under target path, the original GN construction configuration file config.gni is modified.All files and subdirectories under target path constitute the customized embedded software system for current customer.The application distinguishes the software module open source or closed source for customer by whether there is source file or the original or modified GN construction configuration file, and is simple and convenient to realize and manage.

The invention provides a vehicle body pose measurement method, an error measurement method, equipment, a medium and a vehicle, and belongs to the technical field of vehicle body positioning. The measurement method comprises the following steps: establishing a regional coordinate system, and arranging at least two reference points on each of two sides of a preset positioning region; after the target vehicle enters the preset positioning area, collecting linear distances from a projection point of each wheel on the ground to two reference points on the same side through a sensor; obtaining projection point coordinates of each wheel in the regional coordinate system based on the linear distance; and based on the projection point coordinates, obtaining a measurement result of the vehicle body pose of the target vehicle in the regional coordinate system. According to the method, the reference point is set, the distance between the wheel projection point and the reference point is collected through the sensor, the vehicle body pose parameters are obtained in combination with geometric calculation, high-cost RTK equipment or laser radar equipment is not needed, and the method has the advantages that the hardware cost is reduced, the implementation process is simplified, and the deployment flexibility is improved.

The invention discloses a capacitor voltage self-balancing method of a three-level dual-active bridge converter. Different from the conventional balance control method, the invention provides an improved switch sequence which realizes the self-balance of the capacitor voltage from the modulation level. In the improved switch sequence, when the polarity of the output voltage of the NPC full bridge is not changed, only a single capacitor is used for energy exchange. The unbalance of the capacitor voltage enables the current of the transformer to be superposed with an asymmetric current component, thereby facilitating the balance of the capacitor voltage. The improved switch sequence does not change the original output voltage characteristics of the NPC full bridge. And meanwhile, additional capacitor voltage balance control is not needed, so that the complexity of a converter control system is greatly reduced, and the robustness of the converter is improved.

The application discloses a heavy plate large heat input welding method capable of stress relief annealing, which comprises the following steps: processing a groove; symmetrically processing an X-shaped groove on a welding section of a thick steel plate to be welded, the X-shaped groove comprising a central flat part and two inclined parts symmetrically arranged on both sides of the central flat part in the thickness direction; and welding; controlling the assembly gap of the tailor-welded thick steel plate to be zero and adopting double-wire submerged arc welding to perform single-pass welding on the upper and lower surfaces of the thick steel plate respectively. The welding method can effectively improve the stress relief heat treatment adaptability of the weld by adopting the double-wire submerged arc welding method, reasonably setting the power polarity and the angles of the welding guns, matching the specific groove form, the process parameter window and the welding material, thereby improving the large heat input efficient welding and avoiding the large decrease of the low-temperature impact toughness of the weld after the stress relief heat treatment.

The application discloses a method for inhibiting wing roll motion of a flying wing layout aircraft. The method comprises the following steps: determining an angle of attack range of the wing roll motion of the flying wing layout through a free roll motion wind tunnel test; obtaining a position of a reattachment region of a leading edge vortex on an upper surface of the wing through a space flow field display wind tunnel test when the angle of attack of the wing roll motion passes through the space; and determining a gap jet with a proper position and size according to the position of the reattachment region, so that the gap jet generates a suction effect in the roll motion, controls a dynamic breakage and recovery mechanism of the leading edge vortex, and inhibits the wing roll motion of the flying wing layout. The method has the advantages of simple implementation process, obvious inhibition effect and strong engineering practicability.

The application discloses a multi-module degradation unit cooperative correction large-signal reliability modeling method, comprising the following steps: establishing a compact model topology for a radio frequency active device, and extracting initial parameters of each element in the compact model topology; based on the initial parameters, analyzing the degradation behavior of key characteristics of the device, and simulating by using the compact model topology to determine sensitive parameters; constructing a correction network, and realizing fast prediction of different sensitivity degrees and drift trends of the sensitive parameters by training the correction network, so as to map a multi-dimensional input vector into a sensitive parameter correction amount; constructing multiple degradation correction modules, each degradation correction module comprising multiple parallel distributed correction networks, so as to cooperatively correct different categories of sensitive parameters, and embedding the output sensitive parameter correction amount into the compact model topology to form a large-signal reliability model. The method can quickly and accurately realize degradation simulation and prediction of the large-signal characteristics of the device, and the realization process is simple and has strong applicability.

The application provides a link re-establishment method, device, terminal and network equipment, wherein the link re-establishment method comprises the following steps: in the case that the PC5 link of the first terminal occurs RLF, according to the RRC state of the Uu interface of the first terminal, the first information is sent to the network equipment by using the first mode; the first information comprises PC5 RLF information; the candidate second terminal information sent by the network equipment is received; PC5 link re-establishment is carried out according to the candidate second terminal information; wherein the first mode comprises any one of the following: through the Uu interface; through the information of random access; through the pre-configured resource. The present scheme can avoid the situation that the link can only be re-established by sending information to the network equipment through the second terminal, simplifies the implementation process, shortens the link re-establishment delay, and well solves the problem that the link re-establishment scheme for PC5 link failure in the prior art has a large delay.

The application discloses a heavy plate large heat input welding method capable of stress relief annealing, which comprises the following steps: processing a groove; symmetric X-shaped grooves are processed on the welding section of the thick steel plate to be welded, the X-shaped grooves comprise a central flat part and two inclined parts which are symmetrically arranged on the two sides of the central flat part in the thickness direction; welding; the assembly gap of the tailor-welded thick steel plate is controlled to be zero, and three-wire submerged arc welding is adopted to perform single-pass welding on the upper and lower surfaces of the thick steel plate respectively. The welding method adopts three-wire submerged arc welding, the welding gun angle and the power polarity are reasonably set, the welding material and the process parameter window are reasonably matched, the optimized groove form and the process parameter window are matched, and thus the low-temperature impact toughness of the weld after the stress relief annealing heat treatment is avoided from greatly decreasing.

The application discloses a kind of enterprise-level solid state disk fault early warning methods of multi-instance confrontation learning, based on the most relevant monitoring index, using multi-instance learning method long time sequence SMART data is converted into the instance of multiple solid state disk state degradation data, bagging analysis is carried out to multiple instances, realize solid state disk long time sequence fault information fusion;Using the alignment known solid state disk fault feature distribution and unknown solid state disk fault feature distribution of confrontation domain adaptation network based on multi-instance learning, the known solid state disk fault feature knowledge is migrated to unknown solid state disk data.Through self-adaptive fault early warning model, the health status of solid state disk at the current time is determined.The application can learn multiple solid state disk multiple fault feature knowledge, realize online fault early warning based on solid state disk state monitoring, can guarantee in zero false alarm, obtain high level of fault detection rate.

This application provides a method for determining the positions of key points on a display device and the human body. This method eliminates traditional dedicated hardware such as sensors / radar during user posture data acquisition and processing, collecting data solely through a camera and directly acquiring the video stream using WebRTC technology natively supported by the browser, without requiring additional equipment. The human posture recognition model is integrated into the user device using TensorFlow.js, enabling on-device inference and avoiding data uploads to the cloud, thus eliminating transmission latency at its source. Model inference is completed on the browser side, accelerated by a web graphics library to improve processing efficiency and ensure real-time performance. The approximate positions of key points are located using the maximum probability of a heatmap, and then corrected using offsets to improve the positioning accuracy of human key points, further optimizing transmission efficiency and solving the problem of data transmission latency and its impact on transmission efficiency in current user posture data acquisition and processing.