32results about How to "Solving Lag Problems" patented technology

An image pickup apparatus having function for synthesizing image signals corresponding to a plurality of frames of different exposure amounts to generate wide dynamic range, synthesized image is provided with: at least two control means among a taking control means based on a normal taking mode, a taking control means based on a forced wide dynamic range taking mode, and a taking control means based on an automatic wide dynamic range taking mode for selectively generating wide dynamic range, synthesized image automatically on the basis of object information or information set for the image taking; and means for selectively setting one image taking mode out of the taking modes respectively corresponding to the two control means. It is thereby possible to take image by selecting one image taking mode from at least two taking modes among the normal taking mode, the forced wide dynamic range taking mode, and the automatic wide dynamic range taking mode.

The invention discloses a tunnel event detection method based on integrated learning time sequence prediction, which mainly solves the problem that similar methods fail to accurately predict values of a sensor and cannot effectively control tunnel ventilation. The method comprises the detection steps of: pre-processing acquired highway tunnel data to form a training data set; training a plurality of basic predictors according to the training data set, and forming a strong predictor by the weighted combination of the basic predictors; utilizing the strong predictor to calculate a predictive value of smoke concentration of a tunnel at t+1 time according to a value of a tunnel sensor at the current t time, and dynamically adjusting the basic predictors which take part in the integration according to prediction error; comparing the predictive value of the smoke concentration of the tunnel at t+1 time with a smoke concentration threshold, and judging whether the smoke concentration is an over-standard event; and for the over-standard event, calculating control parameters of a ventilation controller, and reducing the smoke concentration. The method has the advantages of strong prediction function and high control accuracy of the ventilation controller, and is used for operation monitoring, energy conservation and emission reduction of highway tunnels.

The invention relates to an anti-wind self-adaptive compensation method for a large antenna. The anti-wind self-adaptive compensation method for the large antenna is characterized by at least including the first step of establishing an accurate random wind load model at the position of the antenna, the second step of establishing a control-oriented model containing flexible information of the antenna, the third step of carrying out simulation on the established wind model and an antenna model, obtaining the flexible oscillation information of the antenna and calculating pointing errors, the fourth step of further predicting the pointing errors according to the Kalman filtering mode, wherein the prediction time equals the sum of an inertia time constant of a motor and an inertia time constant of a filter, the fifth step of designing a low-pass filter, solving a transfer function of the antenna model and a motor reducer model, and carrying out backstepping calculation on a predicted value to obtain a feedforward control signal, and the sixth step of feeding forward feedforward control voltage to a control system so as to enable the controlled model to produce a rigid rotation angle to offset the pointing errors caused by flexible oscillation. Through the method, pointing accuracy, electrical performance and work efficiency of the antenna are improved.

The invention relates to a catalytic cracking device optimization method based on a data model, and provides a reaction device model established according to big production history data. The method does not depend on a complicated process mechanism, and the yield and key properties of a product can be accurately predicted. The variable correlation algorithm provided by the invention can intelligently screen out variables which are strongly correlated with target variables from massive DCS bit numbers and Lims variables, so that the model complexity is reduced to the minimum, and meanwhile thereliability is guaranteed. In addition, the multi-neural network integrated learning prediction model constructed by the invention is high in operation speed, high in convergence and wide in adaptability, and the proposed optimization method can quickly calculate a result on the premise of consuming few computing resources and can meet the real-time online optimization requirement of the device. An intelligent algorithm is adopted to determine the time delay effect of different process parameters, so that the calculation is more accurate.

The invention discloses pantoprazole and the sodium salt enteric sustained-release micropill preparation thereof. The preparation comprises pantoprazole-containing micropills, an isolation layer, a sustained-release layer, another isolation layer and an enteric layer from inside to outside in sequence; the weight of the first isolation layer is 0.5% to 40% of that of the pantoprazole-containing micropills, the weight of the sustained-release layer is 5% to 100% of that of the pantoprazole-containing micropills, the weight of the second isolation layer is 0.5% to 40% of that of the pantoprazole-containing micropills, and the weight of the enteric layer is 20% to 200% of that of the pantoprazole-containing micropills. The pantoprazole enteric sustained-release micropills can stably release the drug.

The invention discloses an active noise control method and system based on a fuzzy neural network and an armored vehicle driver's helmet. The method includes collecting a reference noise signal in a reference area by a reference microphone, and outputting it to a fuzzy controller as a reference signal; the error microphone collects noise The residual noise signal in the control area is used as an error signal, and is output to a fuzzy controller; and the fuzzy controller is based on an adaptive FX-RBF network training algorithm of a fuzzy neural network, and analyzes the reference signal and the error signal , and output the anti-phase target sound signal to the speaker. The active noise control based on the fuzzy neural network of the present invention has obvious noise reduction effect on noise below 2000 Hz, and the noise reduction effect on low frequency noise below 1000 Hz is particularly significant.

The invention discloses an indoor active noise control method and system for an armored car cabin. The method comprises the steps that a reference microphone collects main noise source noise as reference signal input; an error microphone collects residual noise after noise control as error signal input; and a fuzzy controller receives a reference input signal and an error input signal, analyzes a reference signal and an error signal by adopting a fuzzy neural network-based adaptive FX-RBF network training algorithm, outputs inverted target sound signal y[n] to a loudspeaker and superposes the inverted target sound signal and main noise source noise x[n]. Through fuzzy neural network-based active noise control, the indoor active noise control method has an obvious noise reduction effect on the noise below 2,000Hz, wherein the noise reduction effect on low-frequency noise below 1,000Hz is particularly significant.

The invention discloses a resistance spot welding short-time amplitude modulation control method based on spatter statistical feedback. The method comprises the steps that a current time sequence diagram and a process signal time sequence diagram are established according to welding current and process signals collected in real time, spatter recognition is conducted on the process signal time sequence diagram, and the spatter occurrence moment of single-time welding is obtained; then, statistical analysis is conducted on the spatter occurrence time of one batch of welding spots, the spatter occurrence statistical time and the actually-measured frequency are obtained, and short-time amplitude modulation technological parameters are obtained through comparison with a preset spatter frequencyrange and and feature calculation; and iterative optimization is conducted on the current time sequence diagram according to the short-time amplitude modulation technological parameters till the preset spatter frequency range is reached or spatter does not occur, and welding current is set according to the optimized current time sequence diagram to achieve optimized welding. By combining currentdata and historical data, automatic regulation and control of spot welding spatter are achieved, the frequency of spot welding spatter occurrence can be effectively reduced under different plate combinations, different welding processes and different manufacturing working conditions, meanwhile, the welding spot quality is guaranteed, and the reliable and efficient resistance spot welding process is achieved.

The invention discloses an active noise control method and system based on a fuzzy neural network, and a tank helmet. The method includes collecting a reference noise signal in a reference area by a reference microphone as a reference signal, and outputting it to a fuzzy controller; the error microphone collects a noise control area The residual noise signal, as error signal, and output to fuzzy controller; And described fuzzy controller is based on the self-adaptive FX-RBF net training algorithm of fuzzy neural network, described reference signal and described error signal are analyzed, and Outputting the target sound signal in reverse phase to the loudspeaker, the active noise control based on the fuzzy neural network of the present invention has obvious noise reduction effect on noise below 2000 Hz, especially the noise reduction effect on low frequency noise below 1000 Hz.

The invention discloses a voltage frequency tracking method for distributed power grid-connected inverter protection. The method comprises steps: real-time sampling is carried out on voltage at the port of the inverter in the distributed power grid, and the sampling frequency is fixed to be 600Hz; the voltage sampling value is inputted to the inverter in real time, iterative computation is carried out, and a frequency estimation value is corrected; the frequency estimation value is sent to an inverter protection circuit in real time for participating in logic judgment, and after the system voltage frequency deviates from a normal range and certain delay is reached, the inverter protection action outlet isolates the distributed power and a main power grid system. The voltage frequency tracking method for distributed power grid-connected inverter protection has the beneficial effects that compared with the traditional Fourier principle-based frequency measurement technology, the time delay problem of frequency tracking can be solved from the aspect of the principle, tracking is stronger in real-time performance, convergence is quick and stable, and time limit requirements on inverter protection can be reliably met. As the fixed frequency is adopted for sampling, selection of the sampling frequency has little effect on calculation precision, and to a certain degree, calculation amount of the algorithm is saved.

The invention discloses an active noise control method and system based on a fuzzy neural network and a helicopter driver helmet. The method comprises the steps that: a reference microphone collects a reference noise signal of a reference area to be a reference signal and outputs the reference signal to a fuzzy controller; an error microphone collects a residual noise signal of a noise control area to be an error signal and outputs the error signal to the fuzzy controller; the fuzzy controller analyzes the reference signal and the error signal based on an adaptive FX-RBF network training algorithm of the fuzzy neural network and outputs an inverting target acoustic signal to a speaker. According to the method, the system and the helicopter driver helmet, the active noise control based on the fuzzy neural network has an obvious noise reduction effect for noise under 2000Hz and particularly has a significant effect for low frequency noise under 1000Hz.

The invention discloses an active control method and system for noise in a tank cab. The active control method comprises the steps that a reference microphone collects noise of a plurality of main noise sources and the noise serves as a reference signal to be input; an error microphone collects residual noise after the noise is controlled and the residual noise serves as an error signal to be input; and a fuzzy controller receives and analyzes the reference input signal and the error input signal based on the self-adaptive FX-RBF net training algorithm of a fuzzy neural network, and an inverted target sound signal y (n) is output to a loudspeaker to be overlapped with the noise x (n) of the main noise source. According to the active control method for the noise and system in the tank cab, active noise control based on the fuzzy neural network has an obvious noise-reducing effect on noise below 2000 Hz, wherein the noise-reducing effect on the low-frequency noise below 1000 Hz is especially obvious.

The invention relates to a semi-active optimal prediction control method for a structure. The method comprises the steps of: introducing a gray prediction system to predict the input of an LQG controller, then carrying out optimization design of the parameters of the LQG controller through adoption of a whale optimization algorithm, and finally achieving the semi-active prediction control of the structure through combination with an adaptive neural fuzzy inference system. According to the method, the problem that the weighting function of the active controller is difficult to determine can beeffectively solved, the defect that the traditional LQG control effect is insufficient is overcome, and the control signal precision of an MR damper can be improved. Moreover, the time lag in the system can be effectively compensated through a prediction mechanism, so that the control system can give full play to the performance of a magneto-rheological damper, and finally, the purpose of effectively inhibiting the response of a building structure is achieved.

The invention provides a poisonous gas alarm. The alarm includes a sensor sensing content of poisonous gas, a timing module for timing, a microprocessor having least square method predication function, and a vision alarm module and an auditory alarm module sending alarm to operators according to indexes such as poisonous gas content change values, out-of-limit time points and the like. The invention solves a problem of time lag of a prior alarm. The alarm response speed and risk vigilance of people can be improved substantially. Human health damage is also reduced greatly by the early alarm function. Besides, due to the application of the least square method, error alarm possibility is reduced since the alarm has the predication function, so that the predication accuracy of the alarm is improved greatly.

The invention relates to the field of subway tunnel construction, in particular to a soil supporting device with telescopic grouting filling type shield segments and a construction method. The soil supporting device comprises a soil supporting device body, a grouting pipe and a mortar supplementing pipe. The soil supporting device body comprises a telescopic bag and a fixed base. The telescopic bag is of a hollow bottom-open square structure defined by a top cover and a telescopic side wall. The fixed base is of a square structure. The front end face of the fixed base is provided with a plurality of connecting ports for being connected with the grouting pipe and the mortar supplementing pipe. The bottom of the telescopic bag is fixed to the rear end of the upper surface of the fixed base.An inner cavity of the telescopic bag communicates with the connecting ports. After the telescopic bag capable of doing telescopic movements is filled with mortar, the telescopic bag performs jackingto make close contact with a loose soil body and exert a certain magnitude of pressure stress, and the stripping of the soil body is effectively prevented; and the top cover has a certain waterproof effect, time is bought for the hardening of the mortar, and the retardation problem of the hardening of the mortar is effectively solved.

The invention provides an automobile active suspension H-infinity switching control method based on a T-S fuzzy model. The control method comprises the following steps of 1, establishing an active suspension T-S fuzzy model; 2, proposing an active suspension H-infinity control law based on the T-S fuzzy model; and 3, designing a control switcher. When the state is measurable, the switcher is controlled to be switched to the state feedback H-infinity controller; and when the state is unmeasurable, the controller is switched to an output feedback H-infinity controller. The method has the advantages that the T-S fuzzy model is used for describing the active suspension, local linear conversion is conducted on a nonlinear suspension system, meanwhile, the H-infinity controller is combined with the T-S fuzzy model, a state feedback and output feedback control switching mechanism is designed, and the comfort and safety of the active suspension under the complex road condition are effectively improved.

The invention relates to the technical field of a temperature control technology and water cooling equipment for a linear accelerator, in particular to a water cooling system for the linear accelerator. The water cooling system for the linear accelerator is characterized by comprising a main flow loop and a water mixing loop, wherein the main flow loop comprises a refrigerating machine; output cooling water of the refrigerating machine sequentially passes through a water inlet main pipe, a first flow meter and a water-mixing tee joint to be connected with a water separator; the water separator is connected with the linear accelerator through a first water pipe group; the linear accelerator is connected with a water collector through a second water pipe group; and the water collector sequentially passes through a first tee joint, a first water return main pipe, a second tee joint and a second water return main pipe to be connected with the refrigerating machine. By adopting a method for mixing one part of high-temperature return water flowing out from an outlet of a heat exchange runner of the accelerator into the cooling water, the problem of hysteresis in the dynamic adjustment process of the refrigerating machine can be solved and accurate temperature control on the linear accelerator is achieved.

The invention discloses a fixed time control method considering the parameter uncertainty and the interference of a mechanical arm. An end control method based on a robust control theory for a fixed time is adopted, the parameter uncertainty and the external interference of a mechanical arm system are considered, and tracking control over an expected space track of an end executer for the mechanical arm is realized. Great robust performance can be achieved for the parameter uncertainty and the external interference of the system, and the effects of the interference and the uncertainty on the system are suppressed, and a high-accuracy tracking control effect within the fixed time is obtained.

The invention discloses a high-real-time motion control system and method. The system comprises an ARM module, an FPGA module and a high-performance synchronous receiving module, wherein the FPGA module serves as a peripheral of the ARM module and is connected to an FMC high-speed bus; the high-performance synchronous receiving module is used for providing accurate second pulse synchronizing signals, the second pulse synchronizing signals are connected with an external interruption port of the ARM module, the ARM module receives the second pulse synchronizing signals, high-precision time is obtained through a software implementation method, and control system time is fed back to an upper computer. The upper computer compares the feedback time with the reference time of the upper computer until the synchronization is completed. The method effectively solves the time delay problem of an existing control system, and compared with the prior art, the complexity of the system is simplified, the cost of the control system is reduced, the tracking precision of control is improved, and the method is simple to operate and easy to implement.

The invention discloses a short-term power load prediction method, computer equipment and a storage medium. The prediction method comprises the following steps: S1, acquiring weather forecast data and power load data as sample data; s2, cleaning and normalizing the sample data, and dividing the sample data into a training set and a test set; s3, sorting the sample data from large to small by using a KNN algorithm to obtain a feature value K; s4, selecting all weather factors contained in the first K feature values as prediction input ends; s5, inputting all weather factor data and historical power load data contained in the first K feature values as training data of the prediction; s6, establishing a power load prediction model by adopting the BiLSTM and adjusting the hyper-parameters of the BiLSTM; s7, comparing and analyzing the prediction error; and S8, performing comparative analysis to obtain a characteristic value K of an optimal prediction result. According to the invention, the precision of short-term power load prediction is greatly improved.

The invention discloses a tunnel event detection method based on integrated learning time sequence prediction, which mainly solves the problem that similar methods fail to accurately predict values ofa sensor and cannot effectively control tunnel ventilation. The method comprises the detection steps of: pre-processing acquired highway tunnel data to form a training data set; training a pluralityof basic predictors according to the training data set, and forming a strong predictor by the weighted combination of the basic predictors; utilizing the strong predictor to calculate a predictive value of smoke concentration of a tunnel at t+1 time according to a value of a tunnel sensor at the current t time, and dynamically adjusting the basic predictors which take part in the integration according to prediction error; comparing the predictive value of the smoke concentration of the tunnel at t+1 time with a smoke concentration threshold, and judging whether the smoke concentration is an over-standard event; and for the over-standard event, calculating control parameters of a ventilation controller, and reducing the smoke concentration. The method has the advantages of strong predictionfunction and high control accuracy of the ventilation controller, and is used for operation monitoring, energy conservation and emission reduction of highway tunnels.

The invention discloses an improved self-adaptive generalized predictive control method for a data-driven time delay system, which comprises the following steps of: firstly, determining a controlled variable and a controlled variable, acquiring input and output data of a time-delay system to construct a historical database, processing abnormal data, and adopting a time-varying forgetting factor recursive least square method to establish an autoregressive integral moving average model with controlled quantity of the time delay system; designing a reference trajectory according to the expected value, solving a Diophantine equation, designing a GPC control law, and constructing an optimization performance index; and calculating and implementing the optimal control quantity, collecting input and output data, and updating a historical database. According to the invention, the time lag problem of the system can be effectively solved, the utilization rate of offline and online input and output measurement data is improved, the influence of input and output interference and abnormal data on control is effectively inhibited, and the stability and good robustness of the system are ensured.

The invention discloses a piezoelectric viscous damper. A bypass pipeline is changed into a flat pipe from a universal round pipe, the flat pipe plays a role of a damping hole in a round pipe damper, and the piezoelectric viscous damper is a damper related to the movement speed of a piston and is a semi-active damper with the fastest response speed in the world at present. A control system can control a piezoelectric actuator to generate mechanical strain so as to output displacement, and the required output displacement can be achieved without installing an amplifier device. The height of the flat pipe can be adjusted by adjusting the movement of a control panel according to the real-time acting force, so that the flow rate of viscous fluid is controlled by adjusting the size of the damping hole, and the impact energy of the force on a building structure is absorbed and consumed to the maximum extent; and according to the damper, the purpose of adjusting the damping hole can be achieved without adopting a displacement amplifying device, the problems of time lag and precision caused by the displacement amplifying device are avoided, and the cost and size of the damper are reduced.

The invention provides a recommending word display method and device. The recommending word display method comprises the steps that a recommending word is extracted from basic data for each dimensionality in at least two dimensionalities; at least two characteristics are extracted from the extracted recommending word, and the recommending word is processed according to the characteristics; the processed recommending word is displayed according to behavior data. The recommending word display method and device can get rid of the limitation of the knowledge background of a user and search words, really recommend relevant knowledge to the user, guide the user to search, and well meet the knowledge requirements that the user can not perform describing or the description is not accurate; meanwhile, the time delay problem can be effectively avoided through user behavior mining and historical big data analysis, and the recommended knowledge has higher timeliness.

The invention relates to a catalytic cracking device optimization method based on a data model, and provides a reaction device model established according to big production history data. The method does not depend on a complicated process mechanism, and the yield and key properties of a product can be accurately predicted. The variable correlation algorithm provided by the invention can intelligently screen out variables which are strongly correlated with target variables from massive DCS bit numbers and Lims variables, so that the model complexity is reduced to the minimum, and meanwhile thereliability is guaranteed. In addition, the multi-neural network integrated learning prediction model constructed by the invention is high in operation speed, high in convergence and wide in adaptability, and the proposed optimization method can quickly calculate a result on the premise of consuming few computing resources and can meet the real-time online optimization requirement of the device. An intelligent algorithm is adopted to determine the time delay effect of different process parameters, so that the calculation is more accurate.

The invention discloses a hotspot research discovery method based on social media academic conference information. The method comprises the following steps: acquiring original state data of a social media academic conference based on topic tags; cleaning and analyzing the original state data, and extracting key information; inputting the key information into a pre-established and trained conference information classifier to obtain a corresponding information category; respectively obtaining corresponding research hotspots from three aspects of keyword analysis, user posted text analysis and information spreading network analysis based on information categories; and endowing the obtained research hotspots with weights, and performing weighted summation on all the research hotspots to obtain a final research hotspot sequence. The conference information classifier realizes classification of information categories based on a social media academic conference information organization model. The method provided by the invention can solve the time delay problem of a traditional literature metrology method, and research hotspots in related fields can be found more timely.

The invention provides a poisonous gas alarm. The alarm includes a sensor sensing content of poisonous gas, a timing module for timing, a microprocessor having least square method predication function, and a vision alarm module and an auditory alarm module sending alarm to operators according to indexes such as poisonous gas content change values, out-of-limit time points and the like. The invention solves a problem of time lag of a prior alarm. The alarm response speed and risk vigilance of people can be improved substantially. Human health damage is also reduced greatly by the early alarm function. Besides, due to the application of the least square method, error alarm possibility is reduced since the alarm has the predication function, so that the predication accuracy of the alarm is improved greatly.

The invention provides a target air inlet density control method, device and equipment and a readable storage medium. The target air inlet density control method comprises the steps that the initial target air inlet density is obtained; based on a PI control method, the target air inlet density of the P item is obtained through calculation; the target air inlet density of the I item is determined; summation calculation is conducted on the target air inlet density of the P item and the target air inlet density of the I item, and the optimized target air inlet density is obtained; the optimized target air inlet density is limited, and the final target air inlet density is obtained; and the final target air inlet density is used for controlling the opening degree of a throttle valve. According to the method, the final target air inlet density is used for controlling the opening degree of the throttle valve, due to the fact that the throttle valve is high in control precision and free of response delay, the target air inlet density can be efficiently achieved through the throttle valve, and dynamic closed-loop control adjustment is formed between the target air inlet density and the actual air inlet density; and the time lag problem of the air inlet system of the engine is well solved.