Sf Motors, Inc.

Chongqing Jinkang Powertrain New Energy Co., Ltd.

Chongqing Jinkang New Energy Vehicle Co., Ltd.

Chongqing Jinkang New Energy Automobile Co., Ltd.

Other

Santa Clara University

Kwai Inc.

Terawatt Technology Inc.

Chongqing Jinkang Energy Vehicle Co., Ltd.

Ying Liu

Yifan Tang

Brennan Campbell

Saeed Khaleghi Rahimian

Chien-Fan Chen

Sangwoo Han

Scott Quinlan Freeman Monismith

Derek Nathan Wong

Jeremy Andrew Elsberry

Mehdi Forouzan

Scott Monismith

Wenke Zhang

Lingzhi Liu

Nam Ling

Yongxiong Ren

Jun Hou

Junzhe Shi

Gang Yang

Ken Ogata

Yang Yang

Hajime Hoshi

Ming Kai Hsu

Bin Wu

Chen-Yen Yu

Yifei Pei

Ying Wang

Min Tian

Joseph Tolentino

Kent Wing Kin Chung

Pengli Du

A class-specific neural network for video compressed sensing and methods for training and testing the class-specific neural network are provided. The class-specific neural network includes a Gaussian-mixture model (GMM) and a plurality of encoders, where the GMM classifies video frame blocks with a plurality of clusters and assigns the video frame blocks to the plurality of clusters. Further, the plurality of encoders receive the video frame blocks and generate a plurality of compressed-sensed frame block vectors, where the plurality of encoders correspond to the plurality of clusters.

Class-specific neural network for video compressed sensing

A neural network system, a method and an apparatus for image compression are provided. The neural network may include a generator including an encoder, an entropy estimator, and a decoder, where the encoder receives an input image and generates an encoder output, a plurality of quantized feature entries are obtained based on the encoder output outputted at a last encoder block, the entropy estimator receives the plurality of quantized feature entries and calculates an entropy loss based on the plurality of quantized feature entries, and the decoder receives the plurality of quantized feature entries and generates a reconstructed image. Furthermore, the neural network may include a discriminator that determines whether the reconstructed image different from the input image based on a discriminator loss. Moreover, the generator may determine whether content of the reconstructed image matches content of the input image based on a generator loss including the entropy loss.

End-to-end deep generative network for low bitrate image coding

A method and an apparatus for video processing are provided. The method includes that a decoding terminal receives a plurality of coded video frames coded using one or more generative adversarial networks (GANs), receives network parameters related to the one or more GANs, and decodes the plurality of coded video frames using GANs based on the network parameters. Further, the one or more GANs respectively implement one or more video coding functions including reference-frame coding, motion-compensated frame prediction, and residue-frame coding.

Generative adversarial network for video compression

Various methods and techniques for enhancing a silicon-containing anode for a battery cell are presented. The methods may include providing a silicon-containing anode having reversible electrochemical capabilities including a silicon-containing material and an anode material compatible with a lithium-ion battery chemistry having porous and conductive mechanical properties. The methods may also include enriching a surface layer of the silicon-containing anode with sodium ions to intersperse the sodium ions between silicon atoms of the silicon-containing material. The methods may also include displacing the sodium ions with potassium ions to form a compression layer in the silicon-containing anode. The potassium ions may place the silicon atoms of the silicon-containing material in a pre-compressive state to counteract internal stress exerted on the silicon-containing material.

Systems and methods for potassium enhancing silicon-containing anodes for improved cyclability

Embodiments described herein generally relate to the modification of State of Charge (SoC) calculations within electric vehicles (EVs). A database of data points may be generated based on characteristics of a battery cell at various measured SoCs within a controlled environment. Subsequently, during the operation of an EV, a battery management system (BMS) within the EV may collect various operating data points. The collected operating data points may be utilized to reference similar data points stored in the database in order to determine an SoC value. The SoC value may be utilized to modify or alter the SoC calculations by the BMS for an EV in operation.

State of charge error estimation of battery cells background

Provided herein are systems, apparatuses, and methods of providing electrical energy for electric vehicles. A battery pack can be disposed in an electric vehicle to power the electric vehicle. A battery cell can be arranged in the battery pack. The battery cell can have a housing. The housing can define a cavity within the housing. The battery cell can have an electrode structure arranged within the cavity. The electrode structure can include a first polarity electrode plate, a second polarity electrode plate, and at least one separator. First and second polarity tabs are coupled with the first and second polarity electrode plates, respectively. The tabs include a flat surface, an electrode interface surface and at least one intermediate surface that extends therebetween. The at least one intermediate surface forms an acute angle with the electrode interface surface.

Electric vehicle battery cell structure

A battery cell for an electric vehicle battery pack is provided. A housing of the battery cell can define a cavity. An electrolyte material can be housed within the cavity. A first polarity terminal of the battery cell can be disposed at an open end of the housing. A first conductive tab can be disposed at a closed end of the housing and electrically coupled with a first polarity portion of the electrolyte material. A conductive rod can extend through a core of the electrolyte material and can include a first end disposed at the closed end of the housing and electrically coupled with the first conductive tab. A receptacle can be electrically coupled with the first polarity terminal and can extend towards the electrolyte material to engage with a second end of the conductive rod at the open end of the housing.

Battery cell for an electric vehicle battery pack

A battery cell providing a coated lithium reference lead includes at least one anode layer, at least one cathode layer, and a reference lead. The reference lead includes a conductive wire, a layer of lithium metal coupled to the conductive wire, and a polymer coating that covers the layer of lithium metal. The reference lead is inserted into the battery cell with the at least one anode layer and the at least one cathode layer.

Battery cell design with a coated lithium reference electrode

Apparatuses, systems, and methods of storing electrical energy for electric vehicles are provided. A battery pack can be disposed in an electric vehicle to power the electric vehicle. A battery cell can be arranged in the battery pack. The battery cell can have a housing. The housing can define a cavity within the housing. The battery cell can have a solid electrolyte arranged within the cavity. The battery cell can have a cathode disposed within the cavity along a first side of the solid electrolyte. The battery cell can have an anode. The anode can have a carbon nanotube structure. The anode can be disposed within the cavity along the second side of the solid electrolyte and separated from the cathode by the solid electrolyte. The carbon nanotube structure can have pores. The pores can be deposited with electrolyte material to retain lithium material received via the solid electrolyte.

Lithium-metal free anode for electric vehicle solid state batteries

A physics-based calendar life model for determining the state of health of a lithium ion battery cell. The model accounts for parasitic reactions on anode and cathode particles to accurately determine degradation of a battery. By incorporating electrolyte decomposition in a cathode, the present calendar model can predict capacity retention as well as the substantial rise of cell resistance at high state of charge (SOC) and temperatures. The present calendar model is a simple algorithm, utilizing only three parameters, and determines the capacity retention and resistance rise based on temperature, SOC and time.

Control-oriented physics-based calendar life model for lithium ion cells

A battery cell of a battery pack to power an electric vehicle can include a housing to at least partially enclose an electrode assembly is provided. The battery cell can include a vent plate coupled with the housing via a glass weld at a lateral end of the battery cell. The vent plate can include a scoring pattern to cause the vent plate to rupture in response to a threshold pressure. A first end of a polymer tab can be electrically coupled with the vent plate at an area within a scored region defined by the scoring pattern. A second end of the polymer tab can be electrically coupled with an electrode assembly. The polymer tab can melt in response to either a threshold temperature or a threshold current within the battery cell.

Battery cell for electric vehicle battery pack

An automatically generated and customized fast charging process results in reduced degradation in the battery cell. An algorithm for a particular battery cell profile is automatically generated and customized to minimize degradation due to fast charging for that particular batch. To generate the custom algorithm, battery cell information is retrieved for a profile of a battery, wherein each battery profile may have a particular manufacturer, model, type, electrode batch, and potentially other specific identification information. Each battery cell is charged from a particular SOC level and at a selected C-rate, and then discharged. During discharge, the battery cell is monitored for detection of lithium plating or other undesirable effects. A lookup table is automatically generated from the battery cell information, and can be provided to devices and/or battery management systems. The BMS then uses the lookup table to apply a charging process that is customized to the on-board battery.

Serial SOC testing for improved fast-charge algorithm

Techniques described herein relate generally to determining and applying threshold discharging C-rates for battery cells in low temperature environments. To combat internal resistance within a battery cell at low temperature, heat may be generated within a battery cell via a high discharge C-rate. A higher discharge C-rate may cause more heat generation with a battery cell and the higher temperature may mitigate the low temperature environment. As a result of the heat generation, a battery cell's capacity may be increased. Techniques described herein may identify, for a particular low temperature (0 degrees Celsius and below), a threshold discharge C-rate that if a battery cell is discharged above the threshold, the effect of temperature rising would be more dominant than the effect of the internal resistance and more capacity would be obtained from the battery cell.

Threshold discharging C-rates for battery cells in low temperature environments

Embodiments disclosed herein generally relate to a microporous separator with a pore geometry that creates a low or no tortuosity architecture. In one embodiment, a battery cell may comprise of an anode layer, a cathode layer, and a separator layer positioned between the cathode layer and the anode layer. The separator layer may be comprised of one or more block copolymers. The block copolymers that make up the separator layer may be materials that self-align into a vertical nanostructure. The vertical nanostructures may allow ions within the battery cell to flow in a vertical path between the cathode and anode. This vertical path my create a low or no tortuosity environment within the battery cell.

Growing community of inventors

Santa Clara, CA, United States of America

Ying Liu