27 results about "Field training" patented technology

The invention discloses a power industry large model continuous pre-training method and system based on dynamic self-constraint, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining industry pre-training corpora and instruction training corpora, dynamically adjusting the mixing ratio of the industry pre-training corpora and the instruction training corpora through a curriculum-type strategy, and obtaining an industry pre-training corpora and an instruction training corpora; obtaining a dynamic mixed data set; configuring a reference model based on the dynamic mixed data set, and training a target power industry large model by adopting a differential loss function and the reference model for different types of data in the dynamic mixed data set; according to the differential loss function, self-adaptive KL divergence is calculated according to inter-partition optimization logic, and the self-adaptive KL divergence is adopted to construct a loss function; and obtaining the probability of the reference model through an online reasoning framework, and enabling the training to be continuously carried out based on the probability of the reference model. According to the method, the knowledge conflict problem in professional field training is effectively solved, and the generality of the model is kept while the professional property of the power field is improved.

The utility model relates to the technical field of outdoor knapsack processing and manufacturing, in particular to a field training knapsack which is reasonable in structure, stable in carrying, simple and convenient to operate, labor-saving in carrying and anti-fatigue, is provided with a knapsack main body, and double shoulder straps are arranged on the back surface of the knapsack main body, and is characterized in that an interlayer with an inserting channel is arranged on the back surface of the knapsack main body; and an insertion plate is arranged in the interlayer with the insertion channel.

The present invention provides a field training device suitable for javelin throwing practice, comprising: a gantry, which is symmetrically arranged with respect to each other and fixedly connected by a plurality of connecting frames. The gantry comprises two symmetrical main columns, which are fixedly connected by two groups of gantry trusses, and the surfaces of the main columns are movably sleeved with column sliding sleeves. By arranging a traction shoulder-turning mechanism and a throwing mechanism at both ends of the gantry, the present invention can respectively carry out torso training and throwing training, can accommodate multiple people for training at the same time, increase the training types of the training equipment, and carry out multiple types of training without interfering with each other.

This invention relates to the field of reliability assessment technology for electric vehicle charging equipment, specifically disclosing a method for predicting the lifespan of charging connectors by integrating accelerated life testing and field data. The method includes: constructing a laboratory lifespan dataset and a field dataset; preprocessing the two types of data and converting the laboratory degradation time into equivalent field usage time using an acceleration factor model; pre-training a priori lifespan inference model using the preprocessed laboratory data, synthetic data, and the mapped equivalent field time; inputting historical data of the target device into the priori lifespan inference model and directly outputting the lifespan prediction result; and using the lifespan prediction result for operation and maintenance decisions. This invention uses laboratory accelerated life testing as the main data source, combines a small amount of field data, and can infer lifespan status without field training. It can model the lifespan degradation of key components of charging guns under complex operating conditions, predict remaining lifespan, and assess failure risk.

The application discloses a kind of based on big language model's molecular sieve vertical field intelligent system construction method, its characteristics are that the method includes: structure data standardization and literature information extraction, inject field knowledge and tool call specification by supervision fine tuning, carry out reinforcement learning training using GRPO group relative strategy optimization algorithm and multidimensional reward function, construct GNN graph neural network agent model, realize millisecond level property prediction and uncertainty quantification, and system deployment based on asynchronous scheduling and multistage cache etc. steps.Compared with the prior art, the application has the functions of knowledge question and answer, property prediction, reverse design and synthesis planning in the field of molecular sieve, the monomer big language model internalizes professional knowledge through deep field training, uses GNN agent model as the reward signal source of reinforcement learning to greatly improve the training efficiency, balances response speed and calculation accuracy using native Function Call capability, the method is innovative, the system is complete, and has a broad prospect of material research and application.

1. The name of the design product: graphical user interface of insurance industry field training system for mobile phone. 2. The use of the design product: for running programs. 3. The design points of the design product: the graphical user interface content displayed on the mobile phone. 4. The picture or photo that best indicates the design points: design 1 front view. 5. The mobile phone is a conventional design, and the other views of design 1 and design 2 are omitted. 6. Design 1 is designated as the basic design. 7. The use of the graphical user interface: for insurance industry field training system, and the design 1 front view is the medal display page. Click "view all 32" on the right side of the design 1 front view to jump to design 1 interface change state diagram 1; click the "621 master" button in the lower right corner of design 1 interface change state diagram 1 to jump to design 1 interface change state diagram 2; and the design 2 front view is the work standard interface. Click the "work standard" button above the design 2 front view to jump to design 2 interface change state diagram 1; click the "visit one customer a day" button in the middle of design 2 interface change state diagram 1 to jump to design 2 interface change state diagram 2.

The invention provides a cross-domain ship large model training method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a field training data set of a plurality of fields involved in a ship design whole process; training each subdivided professional adaptation layer of each field in parallel by adopting the field training data set of each field to obtain each trained professional adaptation layer of each field; generating a full-domain adapter based on each trained professional adaptation layer of each domain, and loading the full-domain adapter to the basic large language model to form an initial ship large model; and obtaining a cross-domain training data set and a professional training data set, and performing fine adjustment on the initial ship large model by using the cross-domain training data set and the professional training data set until a preset condition is met, thereby obtaining a final ship large model. By the adoption of the method, accurate and reliable response can be made to cross-domain tasks in the whole ship design process.

The invention discloses an army officer and soldier mental health classification method in the technical field of mental health, and the method comprises the following specific steps: 1, obtaining psychological values A of a plurality of samples and a plurality of sample feature value matrixes I, the psychological values A being 0 or 1; step 2, storing a plurality of psychological values A and a plurality of sample characteristic value matrixes I in a table form; step 3, carrying out standardization processing on the sample characteristic value matrix I after the label listing, extracting principal component characteristics based on principal component analysis, and then selecting characteristic vectors corresponding to the first k maximum characteristic values to form a transformation matrix; and step 4, performing model training for the dichotomy task, finally predicting principal component features of the test set, and outputting a probability value or a category label of the mental health state. According to the method, through combination of principal component analysis and machine learning, high-dimensional feature dimension reduction and automatic risk prediction are realized, and the method is suitable for mental health classification in army officer and soldier training, resident training and task deployment scenes.

The utility model relates to an oxygen generation vehicle suitable for being used in high-altitude regions. The oxygen generation vehicle comprises an oxygen uptake chamber, a filling chamber, an equipment chamber and a power generation chamber. The oxygen generation vehicle is provided with a generator, an oxygen generator, a filling machine, a dispersed oxygen outlet, a nasal suction pipe, an oxygen uptake and filling oxygen bottle and an oxygen bag, can achieve the functions of oxygen generation, filling, oxygen storage, oxygen uptake and the like anytime and anywhere, is provided with a vehicle-mounted forklift, can carry materials and remove obstacles and rescue, and is further suitable for field training and fire rescue in high-altitude regions. Wherein the suction dryer in the equipment room effectively removes air moisture, and is not easy to freeze under the cooperation of the air compressor and the heater, so that the environment adaptability is stronger; the control panel independently mounted in the filling chamber has good electrical safety performance, and all equipment control switches can be concentrated in the control panel and are easy to operate; air quality adjustment and temperature adjustment are carried out in the oxygen uptake chamber through the fan heater and the air conditioner, and the comfort degree of oxygen uptake personnel is improved.

The invention discloses a portable handheld Beidou trainer which comprises a handheld Beidou host and a host supporting frame and further comprises an arm-worn quick-release structure, the arm-worn quick-release structure comprises an elastic fixing band, and the two ends of the elastic fixing band can sequentially penetrate through a wearing opening from the inner side of a groove and extend to the outside of a shell. The extending end part of the hook-and-loop fastener is provided with a hook-and-loop fastener male surface and a hook-and-loop fastener female surface which are mutually matched. According to the invention, flexible switching of three use modes is realized through an innovatively designed arm-worn quick-release structure, through cooperation of a wearing groove, a wearing port and an elastic fixing band, the equipment can be held by hand for operation, and can be quickly fixed on an arm to realize mobile observation without two hands, and static fixed teaching can be carried out through a support frame, so that the teaching efficiency is improved. According to the multi-mode design, seamless connection of classroom learning, field training and equipment protection is achieved in Beidou teaching equipment for the first time, students can still continuously observe the state of the equipment in tactical actions such as advancing and climbing, and the actual combat level and the teaching effect of training are remarkably improved.

The utility model relates to the technical field of equipment box adjustment, and particularly discloses a portable foldable field training equipment box support adjusting device which comprises an equipment box body, one side of the top of the equipment box body is connected with a sealing cover through a hinge, and two electric telescopic rods are fixedly connected into the equipment box body. The top ends of the two sides of the inner side wall of the equipment box body are fixedly connected with bearing plates through fixing shafts, the surfaces of first connecting shafts are rotationally connected with stabilizing plates, the surfaces of the fixing shafts are rotationally connected with lifting plates, and one side of the top of a first placing plate is connected with a second placing plate through a hinge. The lifting plate, the electric telescopic rod, the first connecting shaft, the ejector rod, the first placing plate, the first bevel gear and the bearing block are matched for use, so that when the equipment box body is used, training articles can be placed on the first placing plate and the second placing plate in the equipment box body, and meanwhile, the heights of the two placing plates can be adjusted; the requirements of different trainees for heights are met, and the flexibility is high.

The utility model discloses a multilayer structure heatable sleeping bag which comprises a sleeping bag body, the interior of the sleeping bag body is of a multilayer structure, a zipper used for opening and closing the sleeping bag is arranged on the right side of the sleeping bag body, a heating assembly is arranged on the inner layer of a bottom piece of the sleeping bag body, and the heating assembly is connected with a power interface through a temperature adjusting device. A linear neck guard is arranged on a bottom piece of the sleeping bag body, a U-shaped neck guard is arranged on an upper piece of the sleeping bag body, and an adjusting rope is arranged at a neckline of the sleeping bag body. The heatable sleeping bag with the multi-layer structure has the functions of wind prevention, heating, warm keeping and the like, and the warm keeping requirement of camping in a cold environment is met. As a military product, the sleeping bag is mainly suitable for troops to use during field training and actual combat camping; and the quilt can also be used as a civil product, such as a sleeping tool during travel and camping, and has windproof and warm-keeping functions.

The utility model discloses a target box with reset folding, comprising main frame, by a plurality of metal pipe detachable type sleeve joint is formed, and the frame structure of the outer contour is four square. The target box passes through the design of detachable metal pipe main frame and the vice frame of ribbon or rope or magic tape fixed, has improved portability and assembly efficiency obviously. Main frame is by a plurality of metal pipe detachable type sleeve joint is formed, and the frame structure of the outer contour is four square, and the weight is light and dismounts simple and easy, is suitable for field training or quick deployment scene. The vice frame is connected with the main frame diagonally through the ribbon or rope or magic tape, and the installation is flexible, and the space is small after dismounting. The cloth cover is wrapped frame and is fixed through the ribbon or rope or magic tape, and the compactness of overall structure is further enhanced, and transportation and storage are convenient.

The invention belongs to the technical field of motor vehicle driving training, and particularly relates to a high-efficiency in-site district training method for acquiring out-of-site road driving skills, which comprises the following steps of: acquiring multi-source signals, positioning a vehicle by a site base station and a vehicle-mounted radar, acquiring a state by an OBD (On-Board Diagnostic), updating skill data of trainees in real time, and importing configurable site parameters; dynamically planning an intra-field training area by fusing subject complexity, student mastering degree and field adaptation degree based on a task pushing algorithm; and vehicle-mounted voice teaching is triggered through matching of the vehicle position and the preset skill point location, and the skill model is iteratively updated in real time based on the training data. According to the invention, efficient simulation of an off-site road scene by an in-site environment is realized, traditional training limitations are broken through from the dimensions of safety controllability, skill pertinence, site utilization rate and the like, driving training is promoted to be upgraded to a data-driven intelligent mode, and a systematic solution is provided for cost reduction and efficiency improvement of driving schools and skill improvement of students.

A gear workpiece is placed on a plastic base, a groove block on the plastic base is right opposite to a gear key groove, circumferential positioning of the gear workpiece is achieved, press-fitting work is completed in a press-fitting acting face, a square groove is right clamped on a flat key of a gear shaft through pulling of a handle, and the gear and shaft key groove positioning press-fitting tool is achieved. Meanwhile, the stroke can be adjusted and controlled, use is convenient and easy, the mistake proofing function is achieved during field use, parts cannot be crushed, and the operation method can be mastered only by conducting field training on staff. The tool has the advantages of being reasonable in design, simple, practical, reliable in positioning, convenient and fast to detect, low in manufacturing cost, convenient to use and manage, and particularly suitable for sample development and small-batch production and use.

The application discloses a large language model field adaptation fine-tuning method based on knowledge dimension and task dimension collaborative courses, and belongs to the technical field of artificial intelligence and natural language processing. The application comprises the following steps: firstly, a general field training set and a target field training set are constructed, and the general field training set is filtered to remove the field, so as to reduce the field overlap between the two types of corpus; secondly, the general field samples and the target field samples are mapped to a unified semantic space, a plurality of knowledge clusters are obtained through knowledge clustering, and a knowledge specificity score is calculated according to the distance between each knowledge cluster and the general field center, so as to construct a knowledge dimension course from general to professional; then, the target field training samples are annotated for task complexity, and are merged to form three types of task complexity levels, i.e. low, medium and high, so as to construct a task dimension course from shallow to deep; finally, a two-dimensional course sequence is generated by combining the knowledge dimension sorting and the task dimension sorting, an exponential mixing strategy is adopted to dynamically improve the proportion of the target field samples in the training batch, the basic large language model is fine-tuned for field adaptation, and a target field adaptation model is obtained. The application can improve the performance of the target field while effectively maintaining the original general knowledge and reasoning ability of the model, reduce the catastrophic forgetting, improve the training stability and the cross-field application ability.

A method for adjusting correction values for metering fuel using at least one fuel injector, from a high-pressure accumulator into a combustion chamber of an internal combustion engine, using training and correction datasets. The method includes: in the correction dataset, adjusting the correction values based on training values of the training dataset, the correction value in each active neighboring field being adjusted based on a mean neighboring field training value, the correction value in each field of each field region that includes an active field being adjusted based on a mean field training value, wherein the correction value in each inactive field is adjusted based on a transfer training value, the transfer training value being determined according to at least one correlation rule, based on the training values of the active neighboring fields; and providing the correction map having the adjusted correction values for further use.

The invention discloses a model training method and device, electronic equipment and a computer program product. The method comprises: detecting a learnable coefficient of a to-be-trained model relative to a field data set, the field data set comprising multiple groups of field training data obtained based on corpus data of the same field, a first language model being a to-be-trained model before training by using a preset training set, and a second language model being a to-be-trained model before training by using a preset training set; the second language model is a to-be-trained model trained by using a preset training set, and the preset training set at least comprises multiple groups of universal training data obtained based on universal corpus data; according to the learnable coefficient, the mixing proportion of the domain training data in the training data set of the model to be trained is determined, and the learnable coefficient is in positive correlation with the mixing proportion; and performing additional training on the model to be subjected to additional training by adopting the additional training data set. The technical problem that in the prior art, under the condition that model training is carried out on field data, the model universality and the field adaptive capacity are difficult to balance is solved.

The utility model discloses an industrial robot multifunctional teaching practical training device, which comprises an operation table, a warning device, a pneumatic assembly, an industrial robot and a control unit, the operation table is provided with a sorting assembly training system and a space motion training system, and the space motion training system comprises a track practice unit, a support plate frame and a clamp quick-change unit. The training device comprises a training unit, a track training unit, a support plate frame and training plates, and the training plates are divided into a plane track training plate, a curved track training plate and a handwriting training plate, so that the basic understanding and practical capabilities of students on robot programming, touch screen configuration, positioning control and pneumatic principles are trained; students can master the industrial robot automatic production line technological process, training in the professional fields of mechanical and electrical integration and intelligent control is taken into consideration, and the teaching requirements of vocational education individuation are met.

The utility model belongs to the technical field of portable beds, and particularly relates to a camp bed for field training, which comprises two connecting plates, a first cloth rolling assembly and a second cloth rolling assembly, the two connecting plates are located on the left side and the right side of the two cloth rolling assemblies, each connecting plate is provided with a first hook-shaped sliding groove and a second hook-shaped sliding groove, the first cloth rolling assembly is slidably connected into the first hook-shaped sliding groove, and the second cloth rolling assembly is slidably connected into the second hook-shaped sliding groove. Through the arrangement, different working states of the camp bed can be conveniently switched, the actual size of the folded camp bed can be reduced, and the camp bed is convenient to carry by personnel.

The application provides a cross-field ship large model training method and device, computer equipment, a storage medium and a computer program product. The method comprises: acquiring field training data sets of multiple fields involved in a ship design whole process; training each subdivided professional adaptation layer of each field in parallel by using the field training data set of each field, obtaining each trained professional adaptation layer of each field; generating a whole field adapter based on each trained professional adaptation layer of each field, loading the whole field adapter to a basic large language model to form an initial ship large model; acquiring a cross-field training data set and a professional training data set, and fine-tuning the initial ship large model by using the cross-field training data set and the professional training data set until a preset condition is reached, obtaining a final ship large model. The method can accurately and reliably respond to cross-field tasks in the ship design whole process.

The present application belongs to the technical field of medical apparatus and instruments, in particular to a field training protection and medical transfer equipment and a use method thereof. In view of the problem that the existing equipment cannot effectively protect arms and legs and is inconvenient for the transfer of training personnel, the present application proposes the following scheme, which comprises two first protection plates, a second protection plate, a first rotating shaft fixed to the side of each of the two second protection plates away from each other, a second rotating shaft rotatingly penetrated into each of the two second protection plates, and the first protection plate rotatingly sleeved on the outer wall of the corresponding first rotating shaft. In the present application, the calf and thigh can be protected during field training, reducing the risk of leg injury when the user rolls over. When the wounded need to be transferred, the first protection plate, the second protection plate and the first fixed shaft form a whole to form a simple stretcher, which can easily complete the transfer of the wounded and save the physical exertion of medical personnel.

The present invention discloses a method for detecting warheads in a falling area based on the joint spatiotemporal characteristics of strongly correlated targets. First, camera parameter calibration and three-dimensional spatial coordinate reconstruction are performed. The internal parameters of the camera are calibrated based on a black and white checkerboard pattern, and external parameter calibration is completed in combination with the pixel-to-world coordinate conversion process. The least squares method is used to calculate the intersection of multiple optical axes to achieve three-dimensional reconstruction of spatial coordinate points under multiple fields of view. Then, a preliminary detection of the warhead motion state is performed under low perception accuracy conditions based on optical flow tracking. Finally, the warhead motion state is corrected based on the spatiotemporal characteristics of the strongly correlated targets and the projectile-target intersection parameters, and the preliminary detection results of the warhead are corrected to complete the accurate detection of the warhead. The present invention realizes the accurate detection of warheads in the complex dynamic environment conditions of field training, reduces costs and operational difficulty, and realizes large-scale rapid deployment and application in training sites.

The utility model provides a portable field operation digital integrated audio-video system which comprises a mainframe box and a processing control system integrated in the mainframe box. The processing control system comprises a comprehensive processing assembly, an Android assembly, a network service assembly, a satellite television receiving assembly, a frequency modulation broadcast receiving assembly, a digital sound console and a touch display screen. The sound system is partially integrated in the sound box; the sound system comprises a sound assembly, and a microphone and a headset assembly which are used in cooperation with the sound assembly. The sound assembly is integrated in the sound box; the projection assembly, the satellite receiver, the wireless AP and the router are placed in the accessory box. Modularization and high integration are adopted, integration of multiple functional devices is achieved, portability is high, operation is easy, and the device is suitable for entertainment and teaching activities under the army field resident training condition.

The utility model discloses a pair of compress bag underpants for field exercise and training environments, which comprises T-shaped pieces, the bottoms of the two T-shaped pieces are connected through a front connecting part and a rear connecting part, the two sides of the two T-shaped pieces are connected through a side connecting part, the two sides of the T-shaped pieces are in transition through a concave arc-shaped transition part, and a left compress bag cavity and a right compress bag cavity are formed in the middle sides of the T-shaped pieces. The left compress bag cavity and the right compress bag cavity are in bilateral symmetry. Aiming at common varicocele, in a special environment of field training, night external application can be conveniently carried out, and if emergency gathering occurs, through bonding or easy-to-tear connection of the side connecting part, the external application can be rapidly taken out, and training is not influenced. The left compress bag and the right compress bag are loaded through the left compress bag cavity and the right compress bag cavity, and different compress bags can be selected according to different symptoms. According to the utility model, the biarc structure of the left compress bag cavity and the right compress bag cavity is adopted, so that the position deviation of the compress bag arranged in the biarc structure can be effectively avoided.

The present invention discloses a two-way, fully openable semicircular tent and its opening method, which belong to the technical field of outdoor tents. It addresses the shortcomings of conventional tents in the prior art, such as cumbersome installation and disassembly, poor stability, and unsuitability for applications such as field training, disaster relief, and international rescue. Its main structure includes a tarpaulin, reinforcing cables, and two sets of opening and closing mechanisms. Two or more structural beams are installed between the two sets of opening and closing mechanisms. The tarpaulin is installed on the structural beams using straps, and the reinforcing cables are connected to the structural beams via shackles on the structural beams. The present invention is primarily used in scenarios requiring rapid assembly and disassembly, such as outdoor activities, emergency rescue, and military use.

The application relates to the field of helicopter navigation aid, in particular to a portable navigation aid lamp for helicopter field training, which comprises a lampshade, a navigation aid lamp is arranged in the lampshade, and a lamp cap cover is arranged between the navigation aid lamp and the lampshade; a front cover is arranged, the front cover is downwardly open, a top center hole of the front cover is embedded with a hemispherical lampshade outer edge; an outer shell is a metal piece and is in a barrel shape; the outer shell is separately composed of a battery box shell or is composed of the battery box shell and an extended battery box shell; the outer shell and the front cover are connected through a screw thread; a groove is arranged at the top of the outer shell, a top O-shaped ring is arranged in the groove, and the top of the outer shell is connected with the bottom of the lampshade; a rear cover is in a disc shape, the top of the rear cover is connected with the outer shell through a screw thread, and a bottom O-shaped ring is arranged at the connecting position of the two; the application does not need to additionally connect a power line, a power supply loop is formed through metal conduction, the limitation and deficiency that the navigation aid lamp cannot be placed and fixed in the field are overcome, and the problems of battery endurance and battery replacement are solved.