88results about How to "The implementation process is simple" patented technology

The invention discloses a medical image report structuring method and a related device, and relates to the technical field of natural language processing, and the medical image report structuring method comprises the steps: obtaining a target medical image report; based on a pre-obtained medical image report structured model, structured processing is conducted on the target medical image report, structured information of the target medical image report is obtained, and the medical image report structured model is obtained by conducting instruction fine adjustment on a pre-trained large language model through training data; the training data at least comprises a first training medical image report marked with structured information. The medical image report structuring method disclosed by the invention is an end-to-end structuring method based on a large language model, the key information contained in the medical image report can be accurately, efficiently and normatively extracted by the method, and the method is simple in implementation process and relatively low in manual annotation cost.

The invention discloses an anti-skid control method, device and system for a distributed wheel edge drive loader. The anti-skid control method comprises the steps that the current whole vehicle speed is estimated based on the obtained longitudinal acceleration of the loader, the measured vehicle speed of the loader and the rotating speed of each wheel edge motor; calculating the whole vehicle speed corresponding to the rotating speed of each wheel-side motor based on the rotating speed of each wheel-side motor; calculating the actual slip rate of each wheel based on the estimated current vehicle speed and the vehicle speed corresponding to the rotating speed of each wheel motor; estimating the current optimal slip rate based on the response speed of each wheel-side motor; when it is judged that a certain wheel slips on the basis of the actual slip rate and the current optimal slip rate, the anti-slip demand torque is calculated on the basis of the actual slip rate, the current optimal slip rate and the demand torque of the slipped wheel and an anti-slip algorithm, and the anti-slip demand torque is sent to a wheel edge motor corresponding to the slipped wheel. The maximum driving capacity of the loader can be achieved under the condition that it is guaranteed that the loader does not slip, and the running reliability of the loader on the road surface with the low adhesion coefficient is improved.

The invention provides a method for simulating a complex fault in a three-dimensional geologic model by utilizing spatial analytic geometry. The method comprises the following steps: acquiring fault occurrence information of a site to be researched; deriving a normal vector of the fault in a three-dimensional space according to the obtained fault occurrence information; determining representative origin position coordinates of each layer in the coordinate system of the to-be-simulated region; establishing a three-dimensional geological model according to the two-dimensional geological profile; and according to the normal vector of the fault in the three-dimensional space and the representative origin position coordinates of each layer in the coordinate system of the to-be-simulated region, performing accurate sorting and division on the target fault in the three-dimensional geologic model, thereby realizing consideration of complex fault information in the three-dimensional geologic model. The method can completely reflect on-site real fault distribution characteristics, and is simple and easy to implement and high in modeling efficiency.

The invention discloses an offset minimum sum LDPC (Low Density Parity Check) code decoding algorithm based on a model-driven neural network, and relates to an offset minimum sum LDPC code decoding algorithm. The invention aims to solve the problem that a traditional LDPC decoding algorithm is insufficient in convergence speed, calculation complexity, flexibility and the like. The method comprises the following steps: step 1, constructing a model-driven offset minimum sum algorithm network model, wherein the network model comprises an input layer, a check node layer, a variable node layer and an output layer; step 2, an input layer calculates a receiving channel LLR by using the receiving signal y to carry out initialization; step 3, the check node layer calculates the LLR value of the CN-to-VN; step 4, the variable node layer calculates the LLR value of the VN-to-CN; and step 5, calculating final decoding output by an output layer. The invention belongs to the technical field of decoding.

The invention discloses a bio-based carbon fiber and a preparation method thereof. The method comprises the following steps: dipping a Lyoorace fiber in an aqueous solution containing phosphoric acid and urea; and carrying out stabilization and carbonization treatment on the impregnated fiber to obtain the carbon fiber. By adopting the method, the continuous preparation of the bio-based carbon fiber can be realized, the method has the characteristics of simplicity in process implementation, high efficiency and easiness in industrial production, and the prepared bio-based carbon fiber filament is excellent in mechanical property.

The invention relates to a large language model alignment optimization method based on small data volume fine tuning and a storage medium. The invention relates to the large language model alignment optimization method based on small data volume fine tuning and the storage medium. The invention aims to solve the problems that the existing large model fine tuning needs a large amount of annotation data, needs to consume a large amount of computing resources, and needs higher fine tuning cost and development time. The method comprises the steps that 1, a data set for a specific target task is constructed and divided into two parts, one part is a fine adjustment data set, and the other part is an alignment data set; meanwhile, a problem data set for a specific target task is constructed; 2, obtaining a fine tuning model; 3, obtaining an evaluation model; 4, inputting the alignment data set problem into a fine tuning model, outputting a result by the fine tuning model, and calculating the similarity between the output result of the fine tuning model and a target result; inputting an output result of the fine-tuning model into the evaluation model, and outputting an evaluation result by the evaluation model; and 5, obtaining the target model after alignment optimization.

The invention belongs to the technical field of fiber production, and relates to a low-tension interlacing method for terylene, which comprises the following steps of: arranging a first interlacing nozzle and at least one tensioning yarn guide between a first roller and a second roller, and counteracting part of drafting force borne by the terylene by utilizing friction force between the tensioning yarn guide and the terylene; therefore, the tension of polyester between the first roller and the second roller is reduced when the first interlacing nozzle is used for interlacing. According to the low-tension interlacing method for the polyester, the tension of the polyester between the two rollers is obviously reduced under the condition that the speed of the rollers is not changed, so that the pre-interlacing effect is improved.

The invention discloses a DAB converter phase shift control method without DC bias, which introduces a new control variable D4 to control a driving signal, and is suitable for various phase shift control strategies such as single phase shift, expanded phase shift, double phase shift, three phase shift and the like. According to the method, only input voltage and output voltage values need to be sampled, and an additional hardware circuit does not need to be added. The initial value and the final value of the auxiliary inductive current in a single switching period are both 0, and the inductive current in each switching period is completely symmetrical, so that the problem of power shortage of the first half switching period in a common direct current bias elimination method is solved, and good dynamic response performance is ensured from control. The method is simple in implementation process, and meanwhile good dynamic response performance is guaranteed.

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 present invention discloses a wheel hub model retrieval method, comprising the following steps: receiving a first wheel hub picture to be retrieved; obtaining first feature information according to the first wheel hub picture; obtaining a first view of each wheel hub model according to all wheel hub models in a model database; obtaining second feature information of each wheel hub model according to the first view; sorting all wheel hub models according to the first feature information and the second feature information; and also discloses a picture retrieval storage medium and device. According to the first feature information of the first wheel hub picture, the second feature information of the first view is compared, and pictures in the wheel hub model database are listed for retrieval. The process is simple and improves the accuracy of wheel hub graphic similarity determination, effectively reduces the calculation amount of the picture database, and achieves the purpose of saving costs and improving retrieval efficiency.

The invention discloses a novel gastrointestinal tract environment image enhancement and optimization method. The method comprises the following steps: receiving original image data of a gastrointestinal tract environment; performing real-time shooting by using special medical imaging equipment or obtaining the original image data in other legal modes; analyzing color and contrast information of the original image data on the basis of a preset algorithm: implementing by using multiple algorithms; adjusting image parameters according to the color and contrast information; generating an enhanced gastrointestinal tract environment optimization image; analyzing color and contrast information of the original image data based on a preset algorithm; dynamically adjusting exposure time of an image based on light sensor data of real-time image acquisition equipment so as to stabilize a light environment in gastrointestinal endoscopy. According to the scheme disclosed by the embodiment of the invention, the problem of non-uniform image brightness caused by unstable light environment in gastrointestinal endoscopy can be solved.

The present application discloses a lithium extraction method for an alkaline solution, in which a lithium adsorption material is used in an alkaline environment. Lithium ions in the alkaline solution are adsorbed, the lithium adsorption material is replaced with an alkaline high-lithium low-impurity solution, and then an acid solution is used for desorption, so that a high-lithium salt solution having higher lithium content can be obtained. Lithium concentration can reach 5 g / L or more, and the high-lithium salt solution can enter a bipolar membrane system for electrolysis, thereby preparing an alkaline high-lithium low-impurity solution and an acid solution for replacement and desorption of the lithium adsorption material. In the method provided by the present application, lithium in an alkaline solution is adsorbed by resin, and the lithium is preliminarily separated from sodium and potassium. Then, the sodium and the potassium are gradually removed from the resin according to the difference of the retention characteristics of the lithium, the sodium and the potassium on the resin, so that the function of lithium purification is achieved, and lithium concentration is ensured.

The invention discloses a method for realizing zero-interface secure single-point sign-on based on digital certificates, comprising: a user accesses a page of an application system through a browser; the application system detects whether there is user information in a session; the user browser is redirected to the address indicated by a login link; the system is uniformly logged in and the parameters of the login link are verified to be legal; if there is no user information in the session, the login page is returned to the user browser; the user browser is redirected to the address indicated by a login success link; the application system uses the application system private key to decrypt and verify the parameters in the login link to obtain login user information; the application system uses the uniform login system public key to verify whether the login request in the login link is valid; the page address that the user needs to access is returned to the user browser; the user browser is redirected to the page address that the user needs to access, and the single-point sign-on is completed. The invention improves the security of user identity authentication and data transmission.

The present invention relates to a three-level converter SVPWM method and system based on multi-vector synthesis, the method comprising: dividing a three-level space voltage vector diagram into a plurality of sectors; controlling the duty ratio of a small vector and a middle vector in each sector to be equal, obtaining the constraint conditions of the small vector and the middle vector, combining the constraint conditions of the small vector and the middle vector, calculating the duty ratio of the output voltage vector in each sector; according to the duty ratio of the output voltage vector, combining the preset output voltage vector arrangement sequence, obtaining the output state of the three-phase bridge arm voltage in the three-level converter and the driving signal of each phase switch tube. The present invention uses multi-vectors to synthesize space vectors, and only needs to simply divide the space vector diagram into six sectors, and calculate the time for the synthetic vectors in the six sectors to act respectively, which significantly reduces the amount of calculation, and the implementation process is simple. By controlling the time for the small vector and the middle vector to act in each sector, the balance problem of the three-level midpoint voltage is solved.

The invention provides an intermodulation detection method, electronic equipment and a computer readable medium. The intermodulation detection method comprises the following steps: receiving an intermodulation detection command sent by a network management server under the condition that a key performance index parameter of a base station meets a preset condition; under the condition of not interrupting the service, intermodulation detection is carried out on the channel needing intermodulation detection to obtain an intermodulation detection result; wherein the intermodulation detection result comprises intermodulation power of all channels needing intermodulation detection; and sending the intermodulation detection result to the network management server.

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 invention relates to the technical field of turbojet engines, and in particular to an impeller optimization method for a micro-sized turbojet engine based on a BP neural network. The method comprises the following steps: S1, constructing a sample library, wherein the sample library comprises a training sample library and a test sample library; S2, using the training sample library to train an initial BP network model, and using the test sample library to verify the trained initial BP network model, wherein an initial weight and an initial threshold are provided in the initial BP network model; S3, using a PSO-GA phase coupling algorithm to optimize the weight and threshold in the initial BP network model; S4, updating the BP network model to obtain a final BP network model; S5, inputting an optimization target into the final BP network model, outputting variables to be optimized, and completing the structural design of the impeller according to the variables to be optimized, thereby realizing efficient, accurate and low-complexity impeller optimization design.

The invention discloses a low-orbit satellite precise orbit determination method and system, a terminal and a medium under a GNSS denial condition, and the method comprises the steps: obtaining GNSS data received by a satellite-borne receiver of an anchoring satellite, and employing a dynamics orbit determination method to solve an anchoring satellite precise orbit and dynamics parameters thereof; wherein the anchored satellites are low-orbit satellites which establish links with GNSS denial satellites in the same constellation and are not subjected to same-frequency interference; acquiring inter-satellite link ranging data of the anchoring satellite and the GNSS denial satellite; and based on the anchoring star precision orbit and the inter-satellite link distance measurement data of the anchoring star and the GNSS denial star, a dynamic orbit determination method is adopted to solve the GNSS denial star orbit. And resolving a GNSS denial satellite orbit based on the precise orbit of the anchored satellite and inter-satellite ranging data, so as to restrain the orbit error divergence problem caused by orbit forecast. And low-orbit satellite precise orbit determination which does not directly depend on GNSS satellites and does not depend on a ground monitoring station under the GNSS denial condition is realized.

The invention provides a communication method for IPV4 and IPV6 dual-protocol stack seamless switching of Internet of Things equipment, which comprises the following steps of: creating a user side virtual handle visible to a user side in a session layer between a transmission layer and an application layer; when a user side creates a session link with a target terminal with a target domain name on a session layer, if the current state is a default state or is switched to an IPV4 protocol stack state, executing an IPV4 protocol to carry out a communication process; and if the current state is switched to the IPV6 protocol stack state, executing an IPV6 protocol to carry out a communication process. According to the communication method for seamless switching of the IPV4 and IPV6 dual protocol stacks of the Internet of Things equipment, seamless switching of the IPV4 and IPV6 dual protocols is achieved, the use experience of a user side is improved, and the communication method has the advantages that the implementation process is simple, and configuration of the user side is simplified.

The invention discloses an acoustic superstructure design method and device based on an automatic encoder, and the method comprises the steps: sequentially and independently training a decoder and an encoder, artificially constraining an output result of a bottleneck layer, and overcoming a common non-uniqueness problem of a conventional reverse design. And meanwhile, each detail in the neural network is not deeply studied excessively, the training process of the automatic encoder is changed from a complete'black box 'problem to a'grey box' problem, a decision-making process is further deeply understood, and the interpretability of the network is improved to a certain extent. The ultra-thin broadband sound absorption superstructure is successfully designed according to the method, numerical simulation verification and experimental verification are carried out on the ultra-thin broadband sound absorption superstructure, and the simulation result and the experimental result are well matched with the target curve. When the method is widely applied to on-demand reverse design of other types of acoustic superstructures, the design efficiency of the acoustic superstructures can be greatly improved, and the practical application of the acoustic superstructures is further promoted.

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 invention discloses a software implementation method and device for lottery drawing activity deployment, and relates to the technical field of finance, and the method comprises the steps: receiving customer transaction data; the customer transaction data comprises identification information of a first page where a transaction function executed by a customer is located; determining identification information of a second page of all sub-functions embedded in the first page according to the identification information of the first page; screening multiple pieces of lottery drawing activity information from a pre-stored lottery drawing activity maintenance table by utilizing the identification information of the first page and the identification information of the second page; the lottery drawing activity maintenance table records lottery drawing activity information according to the identification information of the first page and the identification information of the second page, and the lottery drawing activity information comprises parameters used for presenting a lottery drawing activity display page; and deploying the multiple pieces of lottery drawing activity information to the first page and the second page according to the identification information of the first page and the identification information of the second page. Development cost can be reduced, and lottery drawing activity deployment efficiency can be improved.

The present invention relates to a method for quantitatively detecting nitrate of water quality based on Raman spectroscopy, and belongs to the technical field of Raman spectroscopy detection. The present invention has the following advantages: (1) Raman spectroscopy is used as the basis for determination, so as to avoid damage to the water body to be measured, and the method is non-destructive; (2) data acquisition and implementation processes are relatively simple, and the nitrate content of the water body to be measured can be calculated immediately after the Raman spectrum of the water body to be measured is collected, and the method has real-time and sensitivity; (3) the training processing layer and the application processing layer both consider the possible offset phenomenon of the Raman spectrum, and the peak data can also be obtained for the Raman spectrum with a certain offset, and the method has anti-interference; (4) the training layer and the application layer are both provided with a verification part to ensure accuracy; (5) each component has clear logic, clear division of labor, and can be deployed in a distributed manner, and the method has flexibility; (6) the method has the characteristics of non-destructive, rapid, high sensitivity and anti-water interference, and the method can be widely used in environmental protection construction, urban planning, emergency monitoring, agricultural supervision and industrial layout planning, etc.

The invention discloses a water drinking behavior determination method and device, electronic equipment and a storage medium, and relates to the computer technology. The method comprises the following steps: determining an initial frame image and a target frame image according to a video stream; determining an initial liquid volume and an initial liquid level area from the initial frame image, and determining a current liquid level area from the target frame image; and determining a liquid volume change rate according to the initial liquid volume, the initial liquid level area and the current liquid level area, and determining whether the user drinks water according to the liquid volume change rate. According to the method, whether the user drinks water or not can be determined only by analyzing the liquid level change conditions in the initial frame image and the target frame image, the body features or facial features of the user do not need to be recognized, and the beneficial effect of protecting the privacy of the user is achieved while the scheme implementation process is simplified; and the embodiment does not need to perform water drinking behavior analysis on each frame of image in the video stream in real time, so that the beneficial effects of saving the system overhead and reducing the scheme implementation cost are achieved.

The invention relates to a high-performance glass microsphere thermal insulation mortar and its preparation method and use, wherein the thermal insulation mortar includes microcapsule phase change paraffin as a thermal insulation component, a highly concentrated silicone hydrophobic agent as a waterproof component, a zinc borate flame retardant as a flame retardant component, and basalt fiber as an anti-cracking component. The high-performance glass microsphere thermal insulation mortar obtained by the present invention has a linear shrinkage of less than or equal to 0.297%, a thermal conductivity of less than or equal to 0.0458W / (mK), and a continuous burning time of less than or equal to 23.85s. The thermal insulation mortar of the present invention can meet the requirements of flame retardancy, thermal insulation, hydrophobicity, and crack resistance of thermal insulation materials for the exterior walls of residential buildings under severe cold, cold, and hot summer and cold winter climate conditions.

The embodiments of this specification provide a method, device, equipment and medium for evaluating solar radiation stability, wherein the method includes: obtaining historical solar radiation observation data of a target area, calculating the average state of the evolution of solar radiation intensity with a preset time interval in a preset time period as a reference sequence; collecting the solar radiation intensity of a target day as a first sequence according to the preset time interval; calculating the probability distribution functions of the reference sequence and the first sequence respectively, and calculating the Wasserstein distance of the two probability distribution functions, and determining the solar radiation stability of the corresponding day according to the Wasserstein distance, so as to solve the technical problem that the existing quantitative evaluation indicators mainly consider high-frequency fluctuations within a cycle, while low-frequency fluctuation information within a cycle is rarely considered, resulting in inaccurate evaluation results.

The invention provides a freezing diagnosis method and system for a drain valve of a fuel cell, and the method employs the change amplitude of a proportional valve control duty ratio in different states of the drain valve as a judgment basis, i.e., when the drain valve operates normally, the duty ratio of the proportional valve is obviously improved, and when the drain valve is frozen, the duty ratio of the proportional valve is not obviously reduced. And the control duty ratio of the proportional valve is basically unchanged. According to the method, the freezing state of the drain valve can be pre-judged before cold start, cold start failure caused by forced start of the fuel cell is prevented, damage to an electric pile is reduced, reference can be provided for heating unfreezing time, invalid cold start time is prevented from being increased due to too long heating, and energy waste is reduced.

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.