11results about How to "Improve reliable operation" patented technology

The invention discloses a prediction control method based on a world model and online fine tuning, a medium and equipment, and belongs to the technical field of control algorithms. The method comprises the following steps: acquiring multi-modal environment observation data, wherein the environment observation data comprises point cloud data, an RGB image and IMU attitude angle information; inputting the environment observation data into a trained variational auto-encoder network, and predicting a potential state; for the predicted potential state and candidate action sequence, predicting a potential state sequence and a decoded occupation grid map of subsequent steps by using a trained cyclic state space model; and for a current action and a plurality of candidate action sequences, taking minimization of a total cost function of each track as an optimization target, and searching an optimal action sequence in an action space. According to the invention, the prediction efficiency and accuracy are improved.

The utility model relates to a new energy commercial vehicle resolver signal detection circuit comprising a filter circuit comprising a first filter unit and a second filter unit, and the output end of the second filter unit outputs the positive differential output and the negative differential output of a resolver signal to a decoding chip; the differential circuit comprises a buffer amplification unit and a third filtering unit, the input end of the buffer amplification unit is connected with the second output end of the first filtering unit, the output end of the buffer amplification unit is connected with the input end of the third filtering unit, and the output end of the third filtering unit outputs a decoding signal; and the micro-control unit is connected with the output end of the third filtering unit, and the micro-control unit is used for judging the fault type of the rotary transformer in combination with the information of the decoding chip. According to the utility model, through combination of multi-stage filtering, differential signal processing and the micro-control unit, the stability of a rotary transformer signal can be effectively improved, and faults can be accurately diagnosed.

This invention discloses a machining process for precision screws, belonging to the field of mechanical manufacturing technology. It includes the following steps: S1, rough machining and forming: treating the surface of the bar stock and machining functional slot structures, followed by pre-deburring; S2, structural strengthening: heat-treating the workpiece for strengthening; S3, precision finishing: correcting the form and position tolerances of the heat-treated workpiece and precision grinding each surface; S4, quality assurance delivery: completing inspection, cleaning, and packaging. This process flow is clear, employing a logical arrangement of "forming first, then strengthening, and then fine-tuning," resulting in a workpiece with not only a high-hardness surface but also improved coaxiality of its stepped axis and thread trajectory accuracy. This significantly enhances the screw's operational stability at high speeds. Furthermore, through staged quality control and precision correction, it improves the first-pass yield of the workpiece and reduces the risk of scrap due to heat treatment deformation, possessing extremely high industrial application value.

The invention discloses a fault diagnosis method and a fault diagnosis device for a redundant rotary transformer. A decoding method comprises the following steps: acquiring an excitation signal and a voltage induction signal of the redundant rotary transformer; real-time fault diagnosis is carried out on the voltage induction signals, and the current working mode of the redundant rotary transformer is judged; the working modes comprise a healthy mode, an excitation winding disconnection mode and an output winding disconnection mode; according to the working mode of the redundant rotary transformer, constructing a corresponding demodulation matrix so as to demodulate the voltage induction signal into a sine component and a cosine component; and inputting the sine component and the cosine component into a phase-locked loop to solve the position angle and the speed of the rotor. According to the fault diagnosis method and the fault diagnosis device for the redundant rotary transformer, millisecond-level rapid detection and fault-tolerant decoding of various broken line faults can be realized without adding extra hardware, and high-reliability operation of a motor driving system is ensured.

The utility model provides an integrated data acquisition and control module. The integrated data acquisition and control module comprises a circuit board and a shell for accommodating the circuit board, the circuit board is divided into a central area and a peripheral area surrounding the central area, the core control unit is arranged in the central area, and the power supply unit, the communication unit and the signal input and output unit are arranged in the peripheral area; the circuit board is a multi-layer circuit board and comprises a digital signal processing layer, an analog signal processing layer, a power supply layer and a grounding layer, wherein the power supply layer and the grounding layer are arranged between the digital signal processing layer and the analog signal processing layer. According to the utility model, through combination of a specific plane layout and a three-dimensional layered structure, digital signals and analog signals are physically isolated and electrically shielded, signal crosstalk under a high-density layout is fundamentally inhibited, the anti-interference performance and the operation stability of the module are improved, and the module is suitable for severe industrial environments such as rail transit and the like.

The invention discloses an artillery production workshop intelligent management and control method and system based on digitization, and relates to the technical field of workshop management, and the method comprises the steps: collecting multi-source operation state data and key quality detection parameter data of artillery production workshop processing equipment and stations, and carrying out the time alignment processing; a workshop state data set with a unified time sequence is formed, an equipment health index model based on digital twin residual and multi-dimensional feature adaptive fusion is constructed based on the workshop state data set, an equipment health vector is obtained, and an equipment residual life vector is obtained based on residual life prediction of recursive filtering and future working condition virtual deduction; and calculating the beat deviation according to the actual processing time and the preset standard beat time, and calculating the equipment load rate in combination with the task arrival rate and the equipment productivity parameters. According to the method, through multi-dimensional state perception and dynamic collaborative optimization, the precision and robustness of artillery workshop management and control are remarkably improved.

The application discloses a magnetic suspension micro gas turbine power station and a control method thereof, and belongs to the technical field of micro gas turbine power generation. The magnetic suspension micro gas turbine power station comprises a micro gas turbine, a magnetic suspension motor and an electronic control system. The micro gas turbine is one of a type without a regenerator, a type with a built-in regenerator or a type with an external regenerator. The air inlet of the centrifugal compressor of the micro gas turbine is arranged in the magnetic suspension motor, and the motor stator is fixed in the air inlet. The centrifugal compressor and the radial turbine are back-to-back arranged on the rotating shaft, the radial turbine is fixed on the cantilever end of the rotating shaft and located outside the two radial magnetic bearings to form cantilever support, and the centrifugal compressor, the radial turbine, the motor rotor and the magnetic bearing rotor adopt a coaxial mounting structure. The application also adopts a double cooling gas path and a three-ring nested control strategy, and realizes high integration, oil-free, efficient cooling and precise dynamic control of the power station.

This invention provides a fault-tolerant control system and method for a magnetic levitation motor based on virtual sensing technology, relating to the field of magnetic levitation motor technology. The system includes a magnetic levitation motor body, a turbine expander, a measurement and monitoring unit, a power amplification and power supply unit, and a digital controller. The digital controller receives multi-source signals synchronously acquired by the measurement and monitoring unit to obtain the physical measured displacement of the rotor. It performs a mechanism-data dual-drive composite aerodynamic disturbance estimation for the magnetic levitation motor to obtain the aerodynamic compensation force and calculate the virtual displacement of the magnetic levitation rotor. It performs orthogonal reconstruction of the physical measured displacement and the virtual displacement to obtain the reconstructed displacement feedback signal for the digital controller to control the rotor. The digital controller calculates the basic feedback control quantity based on the closed-loop error according to the reference levitation control law. The digital controller reassembles the feedback source, aerodynamic compensation force, and control parameters of the basic feedback control quantity online to form the final control output.

The present application provides a kind of integrated residual detection and ignition point prediction mineral oil transformer low carbonization intelligent reconstruction method and system, method includes: construction low carbonization refill process system, replace the mineral oil in transformer with natural ester insulating oil;The residual proportion of mineral oil in mixed insulating liquid sample is quantitatively detected, and the residual detection result is obtained;In the depth filling stage of low carbonization refill process, a predetermined proportion of nano additive is added to the natural ester insulating oil, and a performance optimization scheme is constructed;The residual detection result and the nano additive concentration in the performance optimization scheme are input into the pre-trained model, the predicted ignition point of the current mixed insulating liquid is calculated and output in real time, and safety warning is carried out based on the predicted ignition point.The technical scheme of the present application not only directly solves the single bottleneck of performance optimization, but also indirectly strengthens the effect of residual control and safety monitoring, to ensure the high reliable operation of the transformer after reconstruction under the premise of low carbon emission.

The invention discloses a satellite-borne optical amplifier power closed-loop control system based on an FPGA, and belongs to the technical field of satellite-borne optical communication, and the satellite-borne optical amplifier power closed-loop control system adopts an anti-radiation FLASH type FPGA as a core controller, and a configuration storage unit is based on a floating gate transistor physical structure so as to eliminate the configuration upset risk caused by single-particle bombardment; the FPGA internal control logic adopts a pure hardware pipeline architecture, comprises combinational logic and sequential logic, and has no software instruction scheduling link; the feed-forward unit realizes quick response through a pre-stored'input optical power-pumping current 'mapping table, and the feedback unit realizes fine steady-state correction by adopting parallel hardware PID (Proportion Integration Differentiation); and the feed-forward quantity and the feedback quantity are subjected to weighted fusion through a hardware multiplier-adder to form a closed-loop control system. By adopting a closed-loop control architecture combining feedforward and feedback, the feedforward unit can output a pre-regulating variable within 50ns at the moment of sudden change of input optical power, initial overshoot of pump relaxation oscillation is quickly suppressed, and the response time of the system is greatly shortened.

The invention discloses a giant magnetostrictive ultrasonic knife handle, a temperature control device and a method for achieving temperature control. Through the ultrasonic knife handle comprising the temperature control device in the embodiment of the invention, on the basis of not influencing the rotating machining function of a main shaft, the temperature of the main shaft is controlled; non-contact, dynamic response and closed-loop control temperature management and amplitude stable control are carried out on the giant magnetostrictive ultrasonic knife handle, good engineering realizability and system compatibility are achieved, and stable control over the ultrasonic amplitude and high-reliability operation of the system are simply achieved.