53results about How to "Resolve imprecision" patented technology

The invention provides a neural network based sewage disposal process optimal control method and aims to solve the problem of excessive energy consumption during a sewage disposal process. The sewage disposal process is a highly non-linear, time-varying and complicated process, and on the premise that effluent qualities meet the standard, reduction of operation energy consumption is much challenging. The method mainly includes two neural networks, wherein one neural network is used for establishing a sewage disposal process prediction model so as to achieve prediction of performance indexes, and the other neural network is used for real-timely optimizing control variable set values according to system states and predicted performance indexes. Finally, optimal control of dissolved oxygen concentration and nitrate nitrogen concentration is achieved, so that requirements of effluent qualities can met, and meanwhile, operation costs of the sewage disposal process can be effectively reduced.

The invention discloses an electromotor dynamic / stable stage parameter identification method based on a stator resistance observer. For the parameter identification at a dynamic stage of an electromotor, a stator resistance T-S (Takagi-Sugeno) fuzzy observer is introduced to obtain an actually-measured value of stator resistance on the basis that the existing MRAS (Model Reference Adaptive System) method is still adopted, and dynamic correction is carried out on a rotating speed reference model and a resistance reference model according to the actually-measured value of the stator resistance; and for the parameter identification at a stable stage, the identification method for the rotating speed of a rotor is achieved by adopting the MRAS method and the stator resistance T-S fuzzy observer which are the same as those adopted in the parameter identification at the dynamic stage, and the identification method for the resistance of the rotor is achieved by adopting simple conversion. The electromotor dynamic / stable stage parameter identification method based on the stator resistance observer has the beneficial technical effects that the problem of inaccurate reference models is solved, the low-speed operating performance of the electromotor can be improved, the adoption of a current injection method is avoided, the current injection equipment is saved, the complexity of a system is lowered, the introduction of torque ripple is avoided, and the like.

The invention discloses a pixelate virtual tree illumination influenced area obtaining method, which comprises the following steps of: (1) extracting a tree structure type according to a geometric shape feature of a tree, and building a tree framework library; (2) automatically determining the tree structure type through geometric shape feature information of a three-dimensional model of virtual tree growth simulation; (3) forming an equivalent calculation three-dimensional model for producing a tree illumination influenced area; (4) generating a time sequence of illumination influenced area calculation, and calculating the altitude angle of the sun at each time point by using geographic location information of the virtual tree growth simulation; and (5) generating an illumination shadow of a virtual tree at each time point by using the equivalent calculation three-dimensional model, and carrying out superimposed calculation by using a pixel as a basic unit to obtain the pixelate virtual tree illumination influenced area. The pixelate virtual tree illumination influenced area obtaining method has the advantages of effectively reducing complexity, increasing speed, and being suitable for forest dynamic growth simulation with large scale and large dimension.

The embodiment of the invention discloses a makeup recommendation method and device, a storage medium and a mobile terminal. The method comprises the steps of acquiring a preview image of a user and matching a corresponding head model according to the preview image; extracting facial features in the head model through a pre-configured detection model, and generating makeup recommendation suggestions according to the facial features, the detection model being a deep learning model trained according to a three-dimensional face data sample associated with a set makeup effect; and displaying the makeup recommendation suggestions, and recommending makeup suitable for the facial features of the user through a makeup recommendation function in the mobile terminal. By the adoption of the technicalscheme, the problem that a makeup recommendation scheme in the related technology is inaccurate can be solved, a personalized makeup recommendation scheme is provided, and the functions of the mobileterminal are enriched.

The present invention discloses a Newton-Raphson and forward pushing coordinated flow algorithm for a transmission and distribution network comprising a distributed power supply. The algorithm comprises the following steps of: (1) inputting an original parameter of the transmission and distribution network; (2) initializing an amplitude and a phase angle of each node voltage; (3) setting the number of iterations of flow calculation as 1; (4) calculating a distribution network flow by using a forward pushing back substitution algorithm to obtain an impregnation power of an edge node of a distribution network; (5) calculating a transmission network flow by using a Newton-Raphson algorithm to obtain each node voltage of a transmission network; (6) determining whether a voltage difference of the edge node in each two adjacent iterations is smaller than a set convergence value, if the voltage difference is smaller than the set convergence value, then coordinating flow iteration convergence, and turning to the step (7), or turning to the step (4) to continue iteration and setting k=k+1; (7) outputting a calculating result. According to the Newton-Raphson and forward pushing coordinated flow algorithm for the transmission and distribution network comprising the distributed power supply, the Newton-Raphson algorithm and forward pushing back substitution algorithm are integrated into the coordinated flow computation, a coordinated flow solution of the transmission and distribution network comprising the distributed power supply can be accurately solved.

The invention relates to a medical detection appliance, in particular to a test electronic pen. The electronic pen comprises an outer shell (1), an upper shell (2), a lower shell (3), a test device (30) and test paper (20), wherein the front end of the outer shell (1) is connected with a liquid absorption plate (7), the liquid absorption plate (7) is covered by a protection sleeve (8), a circuit board (4) and a controller (14) are installed in the outer shell (1), the upper shell (2) and the lower shell (3), and the test electronic pen is characterized in that the test device (30) comprises an initial electrode (31) and a terminal electrode (32), and the test paper (20) comprises test paper (22) and reference test paper (21).

The invention relates to a laser pollution sensor, comprising a laser device, a beam splitter, two fiber bundles, fibers, quartz rods, two photoelectric converters and four optical connectors. The two ends of each fiber bundle are fixedly connected with the optical connector. The laser generated by the laser device enters the beam splitter. One splitting end of the beam splitter is connected with the quartz rod via one fiber bundle. The quartz rod is connected with one photoelectric converter via the other fiber bundle. The other splitting end of the beam splitter is connected with the other photoelectric converter via a fiber. The laser pollution sensor is applied to the automatic online monitoring of the pollution degree of the insulator surface of the high-voltage power transmission line, can effectively solve the imprecision problem of the conventional manual measurement and the like, and supply important data to the design, maintenance and security management of the high-voltage power transmission line.

The invention relates to a girder type structure crack damage identification method under a non-reference data condition, and belongs to the bridge structure health diagnosis technical field; the method comprises the following steps: acquiring a dynamic response signal Sij of a girder, carrying out Fourier transform so as to obtain a dynamic response signal frequency spectrum, and finding a corresponding frequency point of a first order vibration signal; determining the upper limit frequency fh and the lower limit frequency fl of the first order vibration signal according to a binary system wavelet transformation frequency range dividing method, and building a harmonic wave wavelet primary function frequency domain expression; carrying out harmonic wave wavelet packet transformation for the Sij, and extracting the first order vibration signal Sij<1>; respectively finding all peaks and valleys in the Sij<1>, and using a cubic spline fitting mode to calculate a peak envelope line lc anda valley envelope line lt; substituting ordinate data of lc and lt into a damage index calculating formula, and solving Dj; drawing Dj into the same graph; if the Dj in the measuring point k is the peak value, i.e., the measuring point k is the damage position, the bigger the gradient of the peak value-valley value connecting line, the bigger the damages.

The invention provides an acoustic positioning method and system based on an acoustic device. The method includes the following steps: according to an acoustic wave signal emission instruction, the acoustic device sends a plurality of first acoustic wave signals with different frequencies and the same signal strength; the first acoustic wave signals attenuate to second acoustic wave signals during transmission, and the second acoustic wave signals are acquired by a signal receiving device and are analyzed to obtain the signal strength of the second acoustic wave signals; and a processing device determines a relative distance between the signal receiving device and the acoustic device according to the signal strength of the first acoustic wave signals and the signal strength of the second acoustic wave signals, and determines a relative direction between the signal receiving device and the acoustic device according to the direction of the first signals. According to the invention, the problem that an audience position positioned by an acoustic system is not accurate can be solved, the acoustic system can regulate output parameters based on the change of the audience position, so that the best experience effect of the acoustic is achieved.

The invention discloses a method for controlling a terminal. The method comprises the following steps: acquiring AP information of the currently accessed AP and a state parameter in AP accessing whenan access point AP is accessed; acquiring a first switching strategy corresponding to the AP according to the AP information; acquiring mode control data corresponding to the AP information when the state parameter in AP accessing meets the first switching strategy, and executing the mode control data so as to control the terminal to switch to a second mode corresponding to the mode control data from the first mode in the AP accessing. The embodiment of the invention further discloses a device for controlling the terminal.

The invention discloses an upper limb rehabilitation robot control method based on adaptive online learning, and belongs to the field of upper limb rehabilitation robot control system design. The method comprises the operating steps of 1) using Lagrange's kinetic equation to establish a dynamic model of the single-arm system of the upper limb rehabilitation robot; 2) designing a control system model of the upper limb rehabilitation robot based on adaptive online learning; 3) establishing a control simulation model for adaptive online learning of the upper limb rehabilitation robot in MATLAB / Simulink; and 4) through simulation experiments, analyzing the angle and angular velocity tracking errors, and the error convergence of the upper limb rehabilitation robot under the adaptive online learning control method. The method of the invention has innovative and simulation basis, can solve the defects of large errors and insufficient robustness in the motion control of the existing upper limbrehabilitation robot, and has great guiding significance for the design of the upper limb rehabilitation robot control system.

The invention relates to the technical field of liquid crystal display screens, and in particular relates to liquid crystal display screen pressure detection equipment based on clamping type limiting. The equipment comprises a workbench, baffles are symmetrically installed on the surface of the workbench, a transverse plate is arranged on the surface of one baffle, and an air cylinder is connected to the surface of the transverse plate; a limiting piece at least comprises a lead screw rotationally connected to the two baffles, one side of the lead screw penetrates through the baffles, one end of the lead screw is connected with a motor, the surface of the lead screw is in threaded connection with a lead screw nut, a limiting plate is installed on one side of the lead screw nut, a push rod is installed on the bottom face of the limiting plate, one side of the push rod is movably connected with a supporting rod, a concave cavity is formed in the surface of the workbench, and the supporting rod slides in the concave cavity. The limiting plate approaches and extrudes the liquid crystal display screen, the supporting rod slides down along the concave cavity, the supporting rod is attached to the liquid crystal display screen all the time, the side face of the liquid crystal display screen is limited through the limiting plate, inclination and deviation are avoided, meanwhile, the liquid crystal display screen is kept horizontal, and pressure testing is facilitated.

The invention provides a method for extracting a parasitic capacitance. The method comprises the steps that integral circuit layout data is simulated; simulation data of integral circuit layout is acquired; and according to integral circuit technological parameters, the parasitic capacitance can be extracted from the integral circuit layout simulation data. The invention also provides a system for extracting the parasitic capacitance. According to the scheme provided by the invention, the problem that parasitic capacitance extraction is inaccurate due to geometric graph deflections of actual manufacture and integral circuit design during integral circuit manufacture can be solved effectively; and accuracy of the parasitic capacitance extraction can be increased.

The invention provides a parameter adjusting method and device of a PID controller and electronic equipment. The method comprises the steps of obtaining a current PID control parameter of the PID controller, a current control error of a target control object and a current error fluctuation rate; based on a preset PID evaluation index, determining weights corresponding to the control error and the error fluctuation ratio respectively; determining a target control quantity of the target control object based on the current control error, the current error fluctuation ratio and the weights corresponding to the control error and the error fluctuation ratio; and updating the current PID control parameter based on the target control quantity and the current control error. The problem that a traditional control method is not accurate is solved, intelligent electrification is achieved, the method is different from an existing PID parameter adjusting method, a membership function and a fuzzy rule base do not need to be selected, influences of human experience and other inaccuracy factors are reduced, the whole parameter adjusting process is simpler and more efficient, and timely processing of system overshoot is facilitated.

The invention discloses a liquid level detection system and method, computer equipment and a storage medium. The system comprises a controller and a plurality of liquid level sensors, and the controller is connected with each liquid level sensor; the liquid level sensors are arranged at different positions of the liquid level detection area, valves are arranged at different positions of the liquidlevel detection area, and the valves are used for adjusting the working mode of the liquid level detection area; the liquid level sensor is used for detecting the liquid level value of the position where the liquid level sensor is located and sending the liquid level value to the controller. The controller is used for determining the measurement weight of each liquid level sensor according to theworking mode of the liquid level detection area, and calculating a comprehensive liquid level value of the liquid level detection area according to each measurement weight and the liquid level valuesent by each liquid level sensor. By the adoption of the technical scheme, measurement errors of the liquid level sensors can be reduced, and an accurate liquid level value can be obtained.

The invention discloses an algorithm for processing large integers on the basis of a computer. The algorithm comprises the following steps: (S1) storing large integer data which is input by hands or from a file into the computer; (S2) preprocessing the large integer data input in the step (S1) by a preprocessing unit; (S3) calculating the preprocessed data acquired by preprocessing in the step (S2) by a calculation and processing unit in the computer; and (S4) carrying out outputting and processing on a result obtained from calculating the preprocessed data in the step (S3) by a result processing unit in the computer. The algorithm disclosed by the invention has the advantages that dynamically-distributed character strings are adopted as a large-integer storage structure, test for addition, reduction, multiplication, power, extraction of square root, Euclid division, division and divisibility of more than 4000 decimal integers can be carried out, so that the problem of inaccuracy of the existing computer in large-integer calculation is solved.

The invention discloses a multi-joint robot control method based on an adaptive neural network sliding mode controller, and belongs to the field of multi-joint robot control system design. The methodcomprises the following operation steps of 1) establishing a kinetic model of a multi-joint robot system by utilizing a Lagrange kinetic equation; 2) designing a control system model of the multi-joint robot based on adaptive neural network sliding mode control; 3) establishing a control simulation model of the multi-joint robot in MATLAB / Simulink; and 4) through a simulation experiment, analyzingan angle and angular velocity tracking error and an error convergence condition of the multi-joint robot under the action of the adaptive neural network sliding mode controller. The method has innovativeness and simulation basis, can solve the defects of large error, insufficient robustness and the like in the existing multi-joint robot motion control, and has important guiding significance for the design of a multi-joint robot control system.

The invention discloses a static heterogeneous reconfigurable array layout method based on multi-stage simulated annealing, and relates to the field of interconnection architectures of reconfigurablearrays. The method is characterized in that the graph distance is utilized to construct the position constraint of the heterogeneous resources, and a staged simulated annealing method is utilized to separate the mutual influence of the heterogeneous resources in the same simulated annealing. According to the method, a novel cost function based on the graph distance is constructed to solve the problem that a traditional cost function is inaccurate in heterogeneous array modeling, and then a staged simulated annealing layout scheme is provided according to the characteristics of different heterogeneous units. Compared with a result obtained by applying a traditional scheme to a heterogeneous array, the multi-stage simulated annealing method applying the novel cost function can averagely improve the throughput rate by 29.6% under the condition of the same interconnection resources.

The invention discloses a puncture path planning method, a medical image acquisition system, equipment and a medium, the method is executed by a puncture path guiding device, and the method comprises the following steps: acquiring an initial medical image of a target object based on a scanning device; determining a target puncture position of the puncture needle according to the initial medical image, determining a target first area corresponding to the target puncture position on the first substrate according to the target puncture position, and determining a target first hole in the target first area; based on the target puncture position, the target first hole and the second substrate, determining a target second hole corresponding to the target first hole in the second substrate; determining a target depth based on the target first hole and the target puncture position; and generating guide information of a target puncture path of the puncture needle based on the target first hole, the target second hole and the target depth. Through the technical scheme of the embodiment of the invention, the technical effect of accurately planning the puncture path is realized.

The invention discloses a parking lot intelligent vehicle finding method and a parking lot intelligent vehicle finding device. The parking lot intelligent vehicle finding method comprises the steps of: detecting to find out that a vehicle is parked in a parking space; calling a camera capable of shooting the parking space to shoot the parking space; judging whether a passenger getting off the vehicle appears in a multimedia file shot by means of the camera or not; acquiring feature information of the passenger from the multimedia file under the condition that the judgment result is yes; acquiring multimedia files shot by means of all cameras in the parking lot within a predetermined time period after the vehicle is parked in the parking space; acquiring a motion trajectory of the passengercorresponding to the feature information in the parking lot from the multimedia files shot by means of all the cameras; receiving a search request; and searching and displaying the motion trajectoryof the predetermined user in the parking lot according to the feature information of the predetermined user. According to the parking lot intelligent vehicle finding method and the parking lot intelligent vehicle finding device, the problem of imprecision in assisted vehicle finding by the user through recording the moving trajectory of the vehicle in the parking lot is solved, and the vehicle finding precision is improved.

The invention relates to a hyperspectral image denoising method based on multi-target low-rank sparsity and space spectrum total variation. In combination with a multi-objective optimization theory, ahyperspectral image denoising recovery problem is converted into a multi-objective optimization problem of a sparse noise image item, a low-rank clean image item and a data fidelity item in combination with a spatial-spectral total variation item. The sparse noise is modeled by using the 10 norm, and the low-rank term is modeled by using the nuclear norm. Then, the three target items are optimized at the same time through the strong optimizing capacity of the multi-target evolutionary algorithm, and the model is solved to achieve the optimal set of solutions. According to the method, the problems of inaccurate l1 norm sparsity modeling and difficulty in optimal regularization parameter selection in an existing hyperspectral image sparse denoising method can be solved, and in addition, a sub-fitness updating strategy is designed, so that the algorithm is carried out more effectively. According to the invention, the applicability and precision of hyperspectral image denoising can be effectively improved.

The invention relates to the field of light industrial machines, in particular to a filling valve applicable to a beer and CO2-contained beverage product linear filling machine. The filling valve is composed of an angle seat valve, a pipeline, a clamping plate, a positive electrode binding post, an insulation rubber sleeve, a filling head, a negative electrode binding post and the like. The angle seat valve and the filling head are connected together through the pipeline. The positive electrode binding post extends into a filling pipe of the filling head to make contact with filling liquid. The negative electrode binding post is installed on the insulation rubber sleeve. The positive electrode binding post and the negative electrode binding post are connected with a PLC. The filling head is sleeved by the insulation rubber sleeve which can move up and down along the filling pipe of the filling head so as to adjust the filling amount. The whole mechanism is connected with a part of a lifting platform of the linear filling machine through the clamping plate. By means of conductivity of the filling liquid, the negative electrode binding column makes contact with liquid inside the filling head, when the liquid level reaches a certain height, the liquid makes contact with the positive electrode binding post, a circuit forms an access, electric signals are transmitted to the PLC, the PLC controls an electromagnetic valve and then controls the angle seat valve to turn off the filling valve, and therefore quantitative filling is completed.

The invention discloses a smart identification device for injected coin. The device comprises a housing, an identifier and a controller. According to the invention, when a coin enters the side wall and triggers a switch, the switch stays in a triggered state. The controller generates an identification instruction and controls the identifier to identify the coin that stays inside the side wall to develop static identification. The identifier, according to the identified coin information, returns to the controller a first identification signal and a second identification signal. The controller, based on the first identification signal, controls the opening of a control valve. The controller, based on the second identification signal, controls the closing of the control valve. Since when the identifier identifies the coin, static identification is developed so that there is enough time for precise identification of the coin, which solves the problem that the coin cannot be precisely identified due to too short identification time.

The invention discloses an automatic monitor for an angle of a base station antenna and an automatic monitoring method. The monitor and the method utilize an MEMS sensor to accurately measure a lower The invention discloses an automatic monitor for an angle of a base station antenna and an automatic monitoring method. The monitor and the method utilize an MEMS sensor to accurately measure a lowerinclination angle of the base station antenna through the principle of a gravity accelerometer, utilize the principle of a reluctance sensitive device sensor to measure an azimuth angle of the base stinclination angle of the base station antenna through the principle of a gravity accelerometer, utilize the principle of a reluctance sensitive device sensor to measure an azimuth angle of the base station antenna and calculate and correct the azimuth angle so as to eliminate the deviation caused by magnetic disturbance measured by the position of the antenna and ensure the measuring accuracy, andation antenna and calculate and correct the azimuth angle so as to eliminate the deviation caused by magnetic disturbance measured by the position of the antenna and ensure the measuring accuracy, and utilize GPRS transmission and IP transmission modes to automatically transmit the antenna angle data obtained by measuring to a network management center. The monitor and the method solve the inaccurutilize GPRS transmission and IP transmission modes to automatically transmit the antenna angle data obtained by measuring to a network management center. The monitor and the method solve the inaccurate problems on installation and adjustment of the base station antenna in the prior modes, solve manual measurement error or instability, solve the measurement error influenced by the magnetic disturate problems on installation and adjustment of the base station antenna in the prior modes, solve manual measurement error or instability, solve the measurement error influenced by the magnetic disturbance at the position of the antenna, solve the problem that the antenna angle change is not discovered in time, and solve the problem of time and labor wastes for acquiring the antenna angle informatbance at the position of the antenna, solve the problem that the antenna angle change is not discovered in time, and solve the problem of time and labor wastes for acquiring the antenna angle information.ion.

The application discloses a method and device for intelligently finding a car in a parking lot. The method includes: detecting a vehicle parked in a parking space; invoking a camera capable of photographing the parking space to take pictures of the parking space; Get off the passenger from the vehicle; in the case that the judgment result is yes, obtain the characteristic information of the passenger from the multimedia file; obtain the multimedia files captured by all the cameras in the parking lot within a predetermined period of time after the vehicle is parked in the parking space ; Obtain the movement trajectory of the passenger in the parking lot corresponding to the characteristic information from the multimedia files captured by all cameras; receive the search request; search and display the movement trajectory of the predetermined user in the parking lot according to the characteristic information of the predetermined user. The application solves the inaccurate problem of recording the running track of the vehicle in the parking lot to assist the user in finding the car, and improves the accuracy of finding the car.