37 results about "Frequency level" patented technology

The invention relates to the technical field of machine learning, and discloses a load prediction method based on dual feature processing and error correction. The method comprises the following steps: acquiring multi-source time sequence data; decomposing the historical load data into a plurality of modal components by adopting a variational modal decomposition algorithm; classifying each modal component into different frequency levels according to the size of the sample entropy; performing phase-space reconstruction according to the modal component of each frequency level and the corresponding external influence factor data, and generating a multivariable phase-space data set of each frequency level; respectively inputting the multivariable phase space data set of each frequency level into the corresponding load prediction sub-model, generating prediction output results, and superposing the prediction output results; constructing a residual sequence based on the historical load data and the initial load prediction result; inputting the residual error sequence into a residual error prediction model to obtain a load residual error prediction value; and compensating the initial load prediction result through the load residual prediction value. According to the scheme, the load prediction accuracy can be improved.

The invention relates to the field of memory management, and provides a DDR memory frequency switching test method and device. The method comprises the following steps: acquiring a plurality of preset frequency gears of the DDR memory; constructing a multi-stage test flow of at least two different switching modes; for each test stage, generating target data before executing the frequency switching action, acquiring actual data from the DDR memory in the process of executing the frequency switching action, performing consistency comparison on the target data and the actual data, and detecting whether the frequency switching process causes data hopping or not; performing reliability judgment on a corresponding switching mode of the DDR memory based on a comparison result of each test stage, establishing a fault association relationship among the plurality of test stages, and when any test stage detects data hopping, determining a potential frequency switching fault currently existing in the DDR memory; and carrying out collaborative verification on the actual data in different switching modes in the multi-stage test process to form a frequency switching test report. The frequency switching test reliability and test efficiency can be improved.

The present invention provides an anti-interference method and an anti-interference system for a mainboard, which relate to the technical field of mainboards. The method comprises the following steps: obtaining a mixed signal in an environment in which the mainboard is located, and performing multi-scale decomposition on the mixed signal to obtain signal components at different frequency scales, wherein the mixed signal includes an interference signal and a useful signal; performing signal classification processing based on a spectrum diagram of the signal component at each frequency scale to obtain a spectrum diagram of the interference signal at each frequency scale; matching a corresponding mask strategy from a preset strategy library and performing mask processing according to the frequency level of the frequency scale to which the spectrum diagram of the interference signal belongs; performing inverse transformation on the spectrum diagram of the interference signal after mask processing to obtain a denoised interference signal; and performing multi-scale weighted reconstruction on the useful signal and the denoised interference signal to obtain a denoised mixed signal. The present invention achieves effective suppression of complex interference signals in the environment in which the mainboard is located, thereby improving the stability and anti-interference capability of the mainboard.

According to an aspect of the invention there is provided a control system for a renewable energy generator of a renewable energy power plant comprising a plurality of renewable energy generators. The control system comprises one or more controllers configured to execute machine readable instructions to: determine a first power reference value for the renewable energy generator using a droop control technique, the first power reference value being determined based on a frequency signal indicative of a frequency level of the renewable energy generator; receive a second power reference value from a power plant controller associated with the plurality of renewable energy generators; and control a power level of the renewable energy generator based, at least in part, on the first and second power reference values.

A neural network computing system includes a processor including heterogeneous computing devices, a memory, a memory controller that controls the memory, and a system bus that communicates between the processor and the memory controller. The processor generates a frequency level combination table representing frequency level combinations of hardware devices that include the heterogeneous computing devices, the memory controller, and the system bus, each of the frequency level combinations maximizing a decreasing amount of execution time relative to an increasing amount of energy consumption when a performance level of the neural network model is increased by one level, determines, a target performance level, selects a frequency level combination among the frequency level combinations according to the target performance level, and controls, based on a selected frequency level combination, the hardware devices while the neural network model is executed.

The invention discloses a tACS artifact removal method and device combining multi-resolution analysis and principal component analysis, equipment and a medium. The method comprises the steps that electroencephalogram signals are obtained; dividing the electroencephalogram signal by adopting a multi-resolution analysis method based on wavelet transformation to obtain wavelet coefficients of different scales; performing principal component analysis on the wavelet coefficient by adopting a PCA algorithm, and removing components according to an analysis result; and processing the residual PCA components by adopting a power spectrum interpolation method, and reconstructing signals to obtain electroencephalogram signals without tACS artifacts. According to the method, multi-level frequency decomposition is carried out on the signals through wavelet transform, the frequency range where tACS artifacts are located is accurately positioned and separated, then PCA analysis is carried out on data of each frequency level, components containing the strongest artifacts are further distinguished and removed, other important physiological information is reserved, and the flexibility and accuracy of artifact processing are improved. The method can be widely applied to the field of electroencephalogram signal processing.

The invention discloses a method and device for determining the performance grade of an application program interface and electronic equipment, and relates to the technical field of software engineering.The determining method comprises the steps that a plurality of external requests for calling a target application program interface within a preset time period are received, and collecting first response information of each external request in the target application program interface and second response information of an instance corresponding to each external request in the target application program interface, determining an external request score based on all the first response information and all the second response information, and determining an external request score based on the external request score and a preset score level. Determining a score level of the target application program interface, determining a frequency level of the target application program interface in a preset time period based on a preset frequency level, and determining a performance level of the target application program interface based on the score level and the frequency level. According to the method and the device, the technical problem of relatively low accuracy of performance evaluation on the application program interface in related technologies is solved.

One or more embodiments of the invention provide a method and a device for identifying an anti-shake state of a front-end interaction element. According to the method, based on a user session sequence, jump elements which do not adopt an anti-shake mechanism and the use frequency level of each jump element are identified. Then, a generated data table including jump elements that do not employ an anti-shake mechanism and use frequency levels of the jump elements is generated. Afterwards, in response to a received query request of a platform user for the anti-shake state of a front-end interaction element in a corresponding service product, a corresponding anti-shake state report is displayed to the platform user, so that the platform user can execute the anti-shake state report according to the use frequency level corresponding to each jump element which does not adopt the anti-shake mechanism in the anti-shake state report. And determining an anti-shake strategy for each jump element which does not adopt the anti-shake mechanism.

According to one aspect of the present invention, there is provided a method of controlling a renewable energy power plant to provide frequency regulation for a power network to which the power plant is connected. The power plant includes a plurality of power units operable to provide up-regulation and / or down-regulation of the power network. The method comprises: receiving a measured frequency level of the power network, the measured frequency level indicating a frequency deviation within a frequency emergency dead zone of the power network; identifying a sub-band of the frequency deviation from a plurality of sub-bands within the frequency emergency dead zone based on the measured frequency level; identifying one or more power units authorized for at least one frequency regulation service within the identified sub-band; determining a deliverable power offset from the baseline power level for each authorized power unit, the deliverable power offset adjusting the service based in part on the at least one authorized frequency; and determining a power setpoint and assigning the power setpoint to one or more authorized power units to meet a plant power offset request for the measured frequency level. The setpoint is determined by selecting one or more of the authorized power units to provide a deliverable power offset determined for that power unit, wherein a cumulative power offset of the one or more selected power units is less than or equal to the plant power offset request; and if there is a power shortage between the plant power offset request and the accumulated power offset, determining a setpoint for a remaining authorized power unit to meet the power shortage.

The invention discloses a self-adaptive Beidou short message emergency remote control and telemetering method. The method comprises the following steps: identifying the communication length and the frequency level of Beidou short message equipment on a host machine; determining a telemetering transmission scheme, and filling downlink data frames; determining a remote control transmission scheme, and receiving and filling uplink remote control data frames; establishing a telemetering sending task, periodically and cyclically inquiring whether the sending countdown of the Beidou short message is 0 or not by the telemetering sending task, if so, sending a downlink data frame, and if not, quitting and waiting for next inquiry; and establishing a remote control receiving task, regularly querying whether the asynchronous communication interface connected with the short message equipment receives data or not by the remote control receiving task, storing the received data into a buffer area, and extracting a complete uplink remote control data frame according to a frame header, a frame tail and frame verification. The method can adapt to different levels of Beidou short message communication frequencies and length levels, is high in program adaptability, and can automatically match the maximum communication transmission capability when the level of the user card changes.

The present invention relates to a multimodal sentiment analysis method for online education. The method comprises obtaining a text comprehensive modal representation, a speech comprehensive modal representation, and a visual comprehensive modal representation in data to be analyzed, and performing multimodal fusion. The method performs analysis in a multimodal manner and utilizes the complementarity between different modalities to improve the comprehensiveness and accuracy of sentiment recognition. The method further comprises determining an optimal wavelet basis function and an optimal number of decomposition layers based on the fusion result, performing a discrete wavelet transform on the fusion result to obtain high-frequency features and low-frequency features, and finely separating local frequency characteristics. The method further comprises determining a multi-frequency graph attention weight based on the high-frequency features and the low-frequency features, performing feature fusion on the high-frequency features and the low-frequency features based on the multi-frequency graph attention weight to obtain a multimodal sentiment analysis result, capturing the synergy between modalities at different frequency levels, effectively fusing heterogeneous data, more comprehensively understanding complex sentiment expressions, and improving the accuracy of multimodal sentiment recognition.

The invention discloses a motor bearing fault recognition method based on vibration signal analysis. According to the method, recognition of a motor bearing of an orbital wing type wind power generation device is achieved through a hybrid method of multi-modal feature extraction and a convolutional neural network. The method comprises the following steps: firstly, decomposing a complex vibration signal by adopting an empirical Fourier decomposition (EFD) method to obtain multi-scale signal components of a plurality of frequency bands, thereby effectively revealing information of an internal structure and different frequency levels in the signal; then, multi-modal feature extraction is carried out on each component in a time domain, a frequency domain and a time-frequency domain, a time-frequency signal obtained through S transformation can further capture transient features of a bearing fault, background noise and working condition disturbance are effectively inhibited, and key state change information is extracted from a non-stationary signal; and finally, the structure of the convolutional neural network is improved, multi-scale and multi-modal feature information is fused, an efficient and robust classification model is constructed, and rapid and accurate recognition of different types of bearing faults is realized.

The present invention discloses a control method, system, device and storage medium for an intelligent music fountain, which relates to the field of music automation control. The main solutions are as follows: collecting music signal data; respectively calculating the volume, average amplitude frequency, timbre and obtaining a comprehensive index; calculating the frequency transfer value and comparing it with a threshold value to determine the fountain frequency level and shape set; calculating the sum of the amplitude differences between adjacent sampling points and the rhythm intensity index; calculating the music comprehensive characteristic index, comparing it with the threshold value to select the water spraying mode; calculating the comprehensive value of the water spraying mode command according to the music comprehensive characteristic index and the music signal data, and then calculating the nozzle adjustment angle, calculating the light brightness and color value in combination with the rhythm intensity index, and comparing the light color value with the threshold value to select the light color set; this method can automatically adjust the fountain and light effects according to the music characteristics, bringing a more ornamental and artistic audio-visual experience.

According to one aspect of the invention, a control system for a renewable energy generator in a renewable energy power plant is provided, the renewable energy power plant including a plurality of renewable energy generators. The control system includes one or more controllers configured to execute machine-readable instructions to: determine a first power reference value for the renewable energy generators using droop control techniques, the first power reference value being determined based on a frequency signal indicating a frequency level of the renewable energy generators; receive a second power reference value from a power plant controller associated with the plurality of renewable energy generators; and control the power level of the renewable energy generators based at least in part on the first power reference value and the second power reference value.

The present application belongs to the technical field of speech signal processing, and particularly relates to a speech enhancement method and system based on deformable embedding and time-frequency channel jointing. The method adopts multi-scale dense connection processing to capture the context relationship of speech and noise at different time-frequency levels. Adaptive feature embedding is realized in the multi-scale space through deformable convolution, thereby improving the robustness to speech structure changes and the generalization ability to different noise conditions. Through time-frequency channel jointing processing, not only the calculation advantage of lightweight convolution is retained, but also the structure perception ability is enhanced, and the speech dominant area can be focused and irrelevant interference can be suppressed in a complex noise environment.

Disclosed herein are systems and methods for automatic correction of real ear measurements (REMs). A sound signal is produced through a receiver of a hearing device, and a sound pressure signal is sensed using a microphone placed inside the ear canal. The sound pressure signal is transformed to obtain a frequency response signal, a local minimum of the frequency response signal is detected above a programmable frequency level, and a spectral flatness of the frequency response signal is calculated in a selected frequency band surrounding the local minimum. If the spectral flatness is greater than a selected threshold value, acoustic correction is applied to the frequency response in the selected frequency band using an estimated transfer function to obtain a corrected sound pressure frequency response. The corrected sound pressure frequency response is used to modify, or make a recommendation to modify, a physical or operational characteristic of the hearing device.

The present invention belongs to the technical field of image processing, and particularly relates to a video optimization acquisition method for an intelligent factory area based on machine vision, including: performing Gaussian mixture background modeling on the production line area in the surveillance video, and dividing pixel points into several first categories according to the distances between the Gaussian mixture models of different pixel points; constructing a sub-Gaussian serial number sequence for each pixel point, and dividing the pixel points in the first category into several second categories according to the differences between the sub-Gaussian serial number sequences of the pixel points in the first category; determining the possible abnormal degree of each pixel point in the current video frame according to the dislocation differences between the sub-Gaussian serial number sequences of the pixel points in the second category; and adjusting the acquisition frequency level of the surveillance video according to the magnitudes of the possible abnormal degrees of each pixel point. The present invention dynamically adjusts the sampling frequency level, achieving a balance between monitoring accuracy and system efficiency.

The invention discloses an image segmentation method and system based on attention fusion of a multi-scale double-branch encoder, and the method comprises the steps: introducing a data enhancement module to carry out the feature enhancement of an input image, and extracting the structure information of different frequency levels. And the enhanced features are respectively sent into a double-branch encoder composed of a local feature extraction branch and a multi-axis attention branch so as to realize parallel modeling of local detail features and global context information. The features output by the double-branch encoder are further input into a Transform module, and a long-range dependency relationship is modeled through a self-attention mechanism. And then, a fusion module combining a channel compression excitation module and a coordinate attention module is adopted to carry out joint re-calibration on channel importance and spatial position information, and fusion features with higher discrimination capability are obtained. And finally, recovering the spatial resolution layer by layer through a decoder, and outputting a pixel-level segmentation result of the target area.

Transceiver (SE) for transmitting and receiving an electromagnetic signal (EM), wherein the electromagnetic signal (EM) is provided for exchange with a sensor for object detection (OD), wherein the transceiver (SE) has an analog part (ANA) which is designed to to convert a first signal (S1) in at least one intermediate frequency level into a second signal (S2) in a transmission frequency level and to output the second signal (S2) as an electromagnetic signal (EM) via an output (OUT), to receive a third signal (S3) as an electromagnetic signal (EM) via an input (EIN), and converting the third signal (S3) in the transmission frequency level into a fourth signal (S4) in the at least one intermediate frequency level, wherein the third signal (S3) is derived from the second signal (S2), wherein the transceiver (SE) is configured to generate a test signal (TS) and feed it into the analog part (ANA) as the first signal (S1) and to test the analog part (ANA) by comparing the test signal (TS) and the fourth signal (S4) with each other.

The invention belongs to the technical field of voice signal processing, and particularly relates to a voice enhancement method and system based on deformable embedding and time-frequency channel combination, and the method comprises the steps: employing multi-scale dense connection processing, and capturing a context relation between voice and noise at different time-frequency levels; adaptive feature embedding is realized in a multi-scale space through deformable convolution, so that the robustness to voice structure change and the generalization ability to different noise conditions are improved; through time-frequency channel combined processing, the calculation advantage of lightweight convolution is reserved, the structure perception capability is enhanced, a voice dominant region can be focused in a complex noise environment, and irrelevant interference is inhibited.

The invention discloses a data transmission method and device, electronic equipment and a medium, and belongs to the technical field of computers. The method is applied to a first device and comprises the following steps: receiving a first input; in response to the first input, displaying a plurality of application category groups on the first device; wherein each group of application categories comprises at least one application on the first equipment, and the use frequency levels of the applications under different application category groups are different; and sending application migration data to the second equipment, wherein the application migration data is migration data of the applications in the at least one group of application categories.

Embodiments dynamically measure latency for a plurality of functions with a plurality of corresponding frequency levels within a serverless computing cluster; measure a transition latency from an idle state to an active state for the plurality of functions; determine whether a target response time to perform a service level objective (SLO) within the serverless computing cluster is going to be missed; dynamically reallocate at least one core and changing a frequency level across the plurality of functions by scaling down in response to a determination that the target response time to perform the SLO within the serverless computing cluster is going to be met; and dynamically reallocate the at least one core and changing the frequency level across the plurality of functions by scaling up in response to a determination that the target response time to perform the SLO within the serverless computing cluster is going to be missed.

The application discloses an interface light control method based on audio data and a computer readable storage medium, and the method comprises the following steps: acquiring audio data in real time, and identifying drum points in the audio data in real time; acquiring the number of drum points in a current drum point frequency level detection period according to a preset drum point frequency level detection period, and determining a drum point frequency level according to the number of drum points; acquiring the maximum spectral energy sum of drum points in a current drum point energy level detection period according to a preset drum point energy level detection period, and determining a drum point energy level according to the maximum spectral energy sum; controlling the interface light in a next drum point frequency level detection period according to the drum point frequency level, and / or controlling the interface light in a next drum point energy level detection period according to the drum point energy level. The application can enrich the visual effect of a display screen.

This invention discloses a compatibility testing method for frequency-using devices based on a recording and playback device, belonging to the field of wireless communication testing technology. The method constructs a joint action set consisting of desired signal input power levels and interference frequency levels. Under each joint action, it collects and stores IQ sample data of the frequency-using device under test, calculates the bit error rate and the proportion of demodulated effective data, and forms a test trajectory. The IQ sample data is replayed, and a state-action reward function is constructed based on the test trajectory to generate a random testing strategy. During online testing, the candidate interference frequency set is dynamically eliminated based on the threshold power estimate and confidence margin, and backoff control is performed in the unlocked state. Finally, the most unfavorable interference frequency level is located, and the minimum desired signal input power that satisfies the preset bit error rate threshold and the preset demodulated effective data proportion threshold is determined, outputting the digital sensitivity result. This invention can improve the efficiency, stability, and consistency of compatibility testing.

The invention belongs to the technical field of image processing, and particularly relates to a video optimization acquisition method for a smart factory based on machine vision, and the method comprises the steps: carrying out the mixed Gaussian background modeling of a production line region in a monitoring video, and dividing pixel points into a plurality of first types according to the distance between the mixed Gaussian models of different pixel points; constructing a sub Gaussian sequence number sequence of each pixel point, and dividing the pixel points in the first category into a plurality of second categories according to the difference between the sub Gaussian sequence number sequences of each pixel point in the first category; determining the possible abnormal degree of each pixel point in the current video frame according to the dislocation difference between the sub Gaussian sequence number sequences of each pixel point in the second category; and adjusting the acquisition frequency level of the monitoring video according to the possible abnormal degree of each pixel point. According to the invention, the sampling frequency level is dynamically adjusted, and the balance between monitoring precision and system efficiency is realized.

The invention relates to an online education-oriented multi-modal sentiment analysis method, which comprises the following steps of: obtaining a text comprehensive modal representation, a voice comprehensive modal representation and a visual comprehensive modal representation in to-be-analyzed data, carrying out multi-modal fusion, carrying out analysis in a multi-modal mode, and utilizing complementarity among different modals to analyze the sentiment of the to-be-analyzed data. The comprehensiveness and accuracy of emotion recognition are improved; determining an optimal wavelet basis function and an optimal decomposition layer number according to a fusion result, performing discrete wavelet transform on the fusion result to obtain high-frequency characteristics and low-frequency characteristics, and performing fine separation on local frequency characteristics; according to the high-frequency features and the low-frequency features, the attention weight of a multi-frequency graph is determined, feature fusion is carried out on the high-frequency features and the low-frequency features according to the attention weight of the multi-frequency graph, a multi-modal sentiment analysis result is obtained, the synergistic effect between modals is captured on different frequency levels, heterogeneous data are effectively fused, complex sentiment expression is more comprehensively understood, and the sentiment analysis accuracy is improved. And the accuracy of multi-modal emotion recognition is improved.

Aspects of the present invention relate to a method for controlling active power output of a renewable energy power plant comprising one or more renewable energy generators. Aspects of the invention may be used to control the active power output of a wind energy power plant and to determine accurate estimates of the available active power when the frequency level of the connected power network is outside of the frequency deadband.

1. The name of the design product: phase sequence relay. 2. The use of the design product: for comparing phase sequence voltage amplitude, frequency level and phase. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective view.

Embodiments of the present disclosure disclose an image compression and image decompression method, device, equipment and storage medium, wherein the image compression method comprises: obtaining a first image; transforming the first image from a spatial domain to a frequency domain to obtain first transform coefficients corresponding to each pixel in the first image; and performing compression processing on the corresponding first transform coefficients by using a target quantization step corresponding to each pixel to obtain first image compression data; the target quantization step corresponding to each pixel is determined according to a frequency level corresponding to the pixel. According to the embodiments of the present disclosure, different quantization steps can be designed according to the frequency level corresponding to each pixel to perform compression processing on the corresponding first transform coefficients, so that the flexibility in the image compression process can be improved.

This invention belongs to the field of high-integration SOC chip technology and discloses a dynamic voltage and frequency adjustment method, device, and terminal equipment. The method includes: obtaining a mapping table; determining the voltage level and frequency level to be adjusted of the subsystem based on the load strength of the subsystem; sending the voltage level and frequency level to be adjusted of multiple subsystems to the DVFS driver; determining, in the mapping table, the target voltage register value corresponding to the maximum voltage value among the multiple voltage values ​​corresponding to the multiple voltage levels to be adjusted, and the target frequency register value corresponding to the maximum frequency value among the multiple frequency values ​​corresponding to the multiple frequency levels to be adjusted; writing the target voltage register value into the register of the power chip and the target frequency register value into the register of the frequency chip; and adjusting the voltage of multiple subsystems to the target voltage value and the frequency to the target frequency value through the DVFS driver. This invention completes the DVFS process without delay.