Abstract: This application discloses a data processing method performed by a computer device. Obtained service data is inputted into a pre-trained model and an initial service model to obtain first fully-connected layer outputs of the pre-trained model and second fully-connected layer outputs of the initial service model. The pre-trained model and the initial service model construct modeling units with different granularities, first modeling units of the pre-trained model are matched with second modeling units of the initial service model, fully-connected layer outputs of the second modeling units have the same sequence length as fully-connected layer outputs of matched target first modeling units so that with the assistance of the pre-trained model, the target service model can be obtained through the initial service model, the knowledge distillation of the models including modeling units with different granularities is realized, thereby effectively improving the service performance of the service model.

Abstract: A speech recognition text processing method includes acquiring an original text obtained by speech recognition, and performing at least one of punctuation addition processing and numerical normalization processing on the original text to obtain an intermediate result. The method further includes performing word removal processing on the original text through an attention interaction layer based on the intermediate result to obtain a processed text of the original text.

Abstract: Embodiments of the present invention provide a method and a device for detecting SOC of a battery. The method for detecting SOC of a battery includes: determining an initial SOC value of the battery according to a waiting time of the battery; calculating a current SOC value of the battery based on the initial SOC value and a current working current of the battery, denoting as a first SOC value. In embodiments of the present invention, the current of the battery is real-time measured and the SOC value of the battery is calculated by accumulating the current of the battery, in addition, the initial SOC value is calibrated by the waiting time, so that the error of the SOC detection result is reduced, thereby increasing accuracy of the SOC detection result.

Abstract: The embodiments of the present disclosure provide an equipment control processing method and device. The method includes: determining an action type according to an action of the component to be controlled; determining a target action logic code class according to the action type and a plurality of pre-created action logic code classes, wherein the action logic code class is created according to the action types extracted from the actions of a plurality of the components to be controlled; creating an action object corresponding to the component to be controlled according to the target action logic code class; correlating control signal corresponding to the action of the component to be controlled and the action object corresponding to the component to be controlled, and storing correlation between the control signal and the action object, so as to implement control processing on the component to be controlled according to the correlation between the control signal and the action object.

Abstract: The embodiments of the present disclosure provide an equipment control method, apparatus, electronic device and storage medium. The method includes determining one or more equipment control sub-processes involved in control process of equipment to be controlled and execution logic between each equipment control sub-process; obtaining code corresponding to the equipment control sub-process; generating an encoding configuration file corresponding to the control process according to the code corresponding to each equipment control sub-process and the execution logic between each equipment control sub-process; and controlling the equipment according to the encoding configuration file. By applying the solutions provided by the embodiments of the present disclosure, repetitive hard-coding are avoided, the debugging and development at a later stage are simplified, and the development efficiency is improved.

Abstract: The embodiments of the present disclosure provide a battery state of charge prediction method and system. The method includes obtaining voltages and currents of a battery during charge and discharge; obtaining optimized model parameters with genetic algorithm by optimizing model parameters in a second-order RC equivalent circuit model of the battery according to the voltages and currents of the battery during charge and discharge; obtaining a cubic spline fitting function of state of charge of the battery, building a state of charge prediction model of the battery with extended Kalman filter algorithm according to the optimized model parameters and the cubic spline fitting function; predicting the state of charge of the battery according to the state of charge prediction model. The battery state of charge prediction method and system provided by the embodiments of the present disclosure can improve the prediction accuracy of the state of charge of battery.

Abstract: The embodiments of the present disclosure provide a battery classification method and system. The methods includes obtaining circulatory charge and discharge data of a battery pack to be classified, extracting a characteristic data set of the battery pack from the charge and discharge data; reducing the characteristic data set with rough set theory to obtain a reduced characteristic data set of the battery pack; and classifying single batteries of the battery pack according to the reduced characteristic data set with fuzzy clustering algorithm. The battery classification method and system provided by the embodiments of the present disclosure can be applied to the retired power batteries, and improves the efficiency of classifying the retired power batteries.

Abstract: Embodiments of the present invention provide a method and a device for detecting SOC of a battery. The method for detecting SOC of a battery includes: determining an initial SOC value of the battery according to a waiting time of the battery; calculating a current SOC value of the battery based on the initial SOC value and a current working current of the battery, denoting as a first SOC value. In embodiments of the present invention, the current of the battery is real-time measured and the SOC value of the battery is calculated by accumulating the current of the battery, in addition, the initial SOC value is calibrated by the waiting time, so that the error of the SOC detection result is reduced, thereby increasing accuracy of the SOC detection result.

Abstract: Agglomerated fine solids, e.g., agglomerated pellets (2B), that are located within a restricted space or pinch point of a conveyance system (1), and that impede the free-flow of fine solids from one vessel (4) to another vessel (10), are detected and fragmented through the coordinated use of a flow detector (12), breaker plate (8) and press (6). The flow detector (12) senses an interruption or diminishment in the flow of the fine solids and signals and actuates the press (6) which in turn engages and fragments the agglomeration. The breaker plate (8) is sized and configured to trap major agglomerates.

Abstract: Agglomerated fine solids, e.g., agglomerated pellets (2B), that are located within a restricted space or pinch point of a conveyance system (1), and that impede the free-flow of fine solids from one vessel (4) to another vessel (10), are detected and fragmented through the coordinated use of a flow detector (12), breaker plate (8) and press (6). The flow detector (12) senses an interruption or diminishment in the flow of the fine solids and signals and actuates the press (6) which in turn engages and fragments the agglomeration. The breaker plate (8) is sized and configured to trap major agglomerates.