Abstract: A method for reusing farmland drainage is provided. A device used in the method includes a purifying pond, a filtering mechanism, a water pump and a drainage pipe. The method includes steps: (a) siting a wide and suitable place close to farmland to construct the purifying pond according to landform characteristics of the farmland, and performing an anti-seepage treatment on the purifying pond, and connecting the purifying pond with anti-seepage channels of the farmland; (b) adding flocculants into water through the filtering mechanism to remove particulate organic substances, suspended substances, colloid and phosphate pollutants thereof; and (c) pumping and discharging the purified water to an upstream of each of the anti-seepage channels through the water pump and the drainage pipe to refuse the water. The filtering mechanism arranged on a water inlet pipe of the purifying pond can filter impurities in farmland drainage to avoid water pollution.

Abstract: An apparatus includes processor(s) to: perform pre-processing operations including derive an audio noise level of speech audio of a speech data set, derive a first relative weighting for first and second segmentation techniques for identifying likely sentence pauses in the speech audio based on the audio noise level, and select likely sentence pauses for a converged set of likely sentence pauses from likely sentence pauses identified by the first and/or second segmentation techniques based on the first relative weighting; and perform speech-to-text processing operations including divide the speech data set into data segments representing speech segments of the speech audio based on the converged set of likely sentence pauses, and derive a second relative weighting based on the audio noise level for selecting words indicated by an acoustic model or by a language model as being most likely spoken in the speech audio for inclusion in a transcript.

Abstract: Methods and apparatuses for training model based on random forest are provided. The method includes: dividing worker nodes into one or more groups; performing random sampling, by worker nodes in each group, in the preset sample data to obtain the target sample data; and training, by the worker nodes in each group, one or more decision tree objects using the target sample data. Example embodiments of the present disclosure do not need to scan the complete sample data for once, thereby greatly reducing the amount of data to be read, the time cost, and further the iterative update time of the model. The efficiency of training is improved.

Abstract: An apparatus includes processor(s) to: perform pre-processing operations including derive an audio noise level of speech audio of a speech data set, derive a first relative weighting for first and second segmentation techniques for identifying likely sentence pauses in the speech audio based on the audio noise level, and select likely sentence pauses for a converged set of likely sentence pauses from likely sentence pauses identified by the first and/or second segmentation techniques based on the first relative weighting; and perform speech-to-text processing operations including divide the speech data set into data segments representing speech segments of the speech audio based on the converged set of likely sentence pauses, and derive a second relative weighting based on the audio noise level for selecting words indicated by an acoustic model or by a language model as being most likely spoken in the speech audio for inclusion in a transcript.

Abstract: An apparatus includes processor(s) to: perform preprocessing operations of a segmentation technique including divide speech data set into data chunks representing chunks of speech audio, use an acoustic model with each data chunk to identify pauses in the speech audio, and analyze a length of time of each identified pause to identify a candidate set of likely sentence pauses in the speech audio; and perform speech-to-text operations including divide the speech data set into data segments that each representing segments of the speech audio based on the candidate set of likely sentence pauses, use the acoustic model with each data segment to identify likely speech sounds in the speech audio, analyze the identified likely speech sounds to identify candidate sets of words likely spoken in the speech audio, and generate a transcript of the speech data set based at least on the candidate sets of words likely spoken.

Abstract: The present invention provides a hardware acceleration method and a communications system. The hardware acceleration method includes: sending, by a network functions virtualization orchestrator (NFVO), first request information to a virtualized infrastructure manager (VIM), wherein the first request information is configured to request the VIM to deploy the to-be-accelerated VNF onto a host in a management domain of the VIM, wherein a hardware resource of the host meets a requirement of the to-be-accelerated VNF, and the requirement of the to-be-accelerated VNF includes information indicating a type of a required hardware acceleration resource and a size of the required hardware acceleration resource in the to-be-accelerated VNF; receiving, by the VIM, the first request information from the NFVO; and deploying, by the VIM, the to-be-accelerated VNF onto the host in the management domain of the VIM.

Abstract: A joint control method with variable ZVS angles for dynamic efficiency optimization in a WPC system for EVs under ZVS conditions, including: adjusting a phase-shift duty cycle of a secondary active rectifier to control a charging voltage and a charging current of EV's batteries through a charging voltage closed loop and a charging current closed loop, respectively; adjusting a power angle of the secondary active rectifier to control a ZVS angle of the secondary active rectifier through a secondary ZVS angle closed loop; adjusting a phase-shift duty cycle of a primary inverter to control a ZVS angle of the primary inverter through the primary ZVS angle closed loop; determining the current operating case of the WPC system and adjusting the ZVS angles of the primary inverter and the secondary active rectifier to automatically identify an optimal operating point with a maximum charging efficiency through the P&O method.

Abstract: A hybrid offshore wind turbine includes a monopile, friction wheel, and suction bucket.

Abstract: An apparatus includes processor(s) to: generate a set of candidate n-grams based on probability distributions from an acoustic model for candidate graphemes of a next word most likely spoken following at least one preceding word spoken within speech audio; provide the set of candidate n-grams to multiple devices; provide, to each node device, an indication of which candidate n-grams are to be searched for within the n-gram corpus by each node device to enable searches for multiple candidate n-grams to be performed, independently and at least partially in parallel, across the node devices; receive, from each node device, an indication of a probability of occurrence of at least one candidate n-gram within the speech audio; based on the received probabilities of occurrence, identify the next word most likely spoken within the speech audio; and add the next word most likely spoken to a transcript of the speech audio.

Abstract: A method of assisted takeoff of a movable object includes increasing output to an actuator that drives a propulsion unit of the movable object under a first feedback control scheme, determining whether the movable object has met a takeoff threshold, and controlling the output to the actuator using a second feedback control scheme different from the first feedback control scheme in response to the movable object having met the takeoff threshold.

Abstract: Provided are a display panel and a driving method of a display panel. The display panel includes a substrate, a sub-pixel array located on the substrate, and at least one fingerprint identification unit group; each of the at least one fingerprint identification unit group includes a first fingerprint identification unit and a second fingerprint identification unit; the first fingerprint identification unit, which is in the fingerprint identification unit group to which the second fingerprint identification unit belongs, is electrically connected to the pixel circuit of at least one sub-pixel which is peripheral to the second fingerprint identification unit; the first fingerprint identification unit is configured to adjust, in a first fingerprint identification phase, the at least one sub-pixel electrically connected to the first fingerprint identification unit from a first luminance to a second luminance.

Abstract: The present disclosure discloses a method for optimizing equipment capacity and equipment power of an energy hub system. The method includes establishing an energy hub model containing natural gas boilers, electric boilers, coolers and heat pumps, establishing a bilevel optimized upper model to solve the optimal heat pump capacity, and establishing a bilevel optimized lower model to solve the optimal power utilization of each energy device based on the binary search algorithm of the quadratic function solves the upper model by using the multi-objective evolutionary algorithm NSGA-II to solve the lower model. The optimization method of the present invention can solve the multi-objective bilevel model problem without the help of commercial optimization software. Obtaining a reasonable, efficient and green planning scheme makes the total operating cost and total exhaust gas emissions of the energy hub relatively optimal.

Abstract: The present disclosure provides feature data processing methods and devices. One exemplary feature data processing method comprises: classifying features into an important feature set and an auxiliary feature set according to information attribute values of the features; converting features in the auxiliary feature set to hash features; and combining the hash features with features in the important feature set, and setting the combined features as fingerprint features. Training and prediction of to-be-processed data can be performed based on the fingerprint features. With the embodiments of the present disclosure, training dimensions can be more controllable and the amount training data amount can be reduced. Therefore, the efficiency of data processing can be improved.

Abstract: The present invention provides a current detection circuit applied to a SiC field effect transistor. The current detection circuit includes a current detection loop and an acquisition loop on the current detection loop. The current detection loop includes a voltage source, a capacitor, a first SiC field effect transistor, a second SiC field effect transistor, and a sampling resistor. The first SiC field effect transistor is connected to a power signal. The second SiC field effect transistor is connected to a pulse signal. The acquisition loop includes a compensating resistor and a compensating inductor. The compensating resistor and the compensating inductor are connected in series and then connected in parallel at two ends of the sampling resistor to counteract the influence of total parasitic inductance in the current detection loop.

Abstract: A mechanical reliability testing platform for tri-post insulators in a GIL device includes a horizontal-GIL-arrangement-form fixed-tri-post-insulator mechanical reliability verification testing platform for a horizontal dynamic insertion and extraction test, and a turning-GIL-arrangement-form fixed-tri-post-insulator mechanical reliability verification testing platform for a vertical dynamic insertion and extraction test. A driving unit is employed to realize the insertion and extraction of the conducting rod of the sliding-tri-post-insulator GIL form unit at the contact holder, so as to simulate the reciprocating forces on the fixed tri-post insulator induced by the thermal expansion and contraction of the pipe during the actual operation of the GIL, and simulate the working condition of the fixed tri-post insulator under abnormal forces when the GIL experience foundation settlement.

Abstract: Methods, systems, and computer-readable media for dynamic regulation of parallelism for job scheduling are disclosed. A job scheduler sends a resource manager a request to execute a first set of compute jobs using a set of computing resources. The number of jobs corresponds to a first parallelism value. The job scheduler receives a response indicating that the number of computing resources was sufficient or insufficient to schedule the jobs. The job scheduler sends another request to execute another set of compute jobs whose number corresponds to a new parallelism value determined based at least in part on the response. If the number of computing resources was sufficient, the new parallelism value represents an increase over the first parallelism value. If the number of computing resources was insufficient, the new parallelism value represents a decrease from the first parallelism value.

Abstract: The present invention relates to a three-dimensional integrated package device for a high-voltage silicon carbide power module, comprising a source substrate, first chip submodules, a first driving terminal, a first driving substrate, a ceramic housing, a metal substrate, a water inlet, a water outlet, second chip submodules, a second driving terminal, a second driving substrate and a drain substrate from top to bottom; and each first chip submodule is composed of a driving connection substrate, a power source metal block, a first driving gate metal post, second driving gate metal posts, a silicon carbide bare chip, an insulation structure and the like. A three-dimensional integrated half-bridge structure is adopted to greatly reduce corresponding parasitic parameters.

Abstract: A method, a system and a device for difference automatic calibration in cross modal target detection are provided. The method includes: obtaining a plurality of modals of color-thermal images to be detected, and constructing an input image set; acquiring a feature map of each modal image in the input image set, and extracting a shared candidate region; aligning the shared candidate region, and obtaining a feature representation of the each shared candidate region through pooling; obtaining a category of the feature representation of the shared candidate region based on a preset classifier, and obtaining a position of the shared candidate region through deviation regression; adopting non-maximum value suppression to obtain a shared candidate region with the highest probability according to the category of the feature representation of the shared candidate region, and returning to the image to be detected for marking according to the position of the shared candidate region.

Abstract: This application discloses a VNF service instantiation method and apparatus. The method is performed by a network functions virtualization orchestrator NFVO or a virtualized network function manager VNFM. The method includes: receiving a virtualized network function VNF instantiation request message; obtaining deployment information of a container service instance that needs to be invoked for VNF instantiation; determining one or more container as service clusters CaaS clusters based on the deployment information of the container service instance, and sending a container service instance creation request to a container as service manager CaaS manager in each of the one or more CaaS clusters; and receiving a container service instance creation success message from the CaaS manager. According to the foregoing solution, problems of cross-site deployment and management of a container service in an NFV scenario are resolved.

Abstract: An apparatus includes processor(s) to: use an acoustic model to generate a first set of probabilities of speech sounds uttered within speech audio; derive at least a first candidate word most likely spoken in the speech audio using the first set; analyze the first set to derive a degree of uncertainty therefor; compare the degree of uncertainty to a threshold; in response to at least the degree of uncertainty being less than the threshold, select the first candidate word as a next word most likely spoken in the speech audio; in response to at least the degree of uncertainty being greater than the threshold, select, as the next word most likely spoken in the speech audio, a second candidate word indicated as being most likely spoken based on a second set of probabilities generated by a language model; and add the next word most likely spoken to a transcript.