13 results about "Sequence memory" patented technology

The invention relates to the technical field of wind power intelligent operation and maintenance, and discloses a wind turbine generator set temperature early warning method based on hybrid deep learning and a dynamic threshold value, and the method comprises the steps: carrying out the time sequence feature extraction and sequence memory enhancement of synchronous time sequence data through a constructed hybrid deep learning anomaly detection model; the method comprises the following steps: acquiring a wind turbine generator fusion feature fused by a fine-grained short-term feature and a coarse-grained long-term feature, predicting an abnormal probability at each moment, and generating a wind turbine generator temperature early warning signal by identifying a real-time working condition category, dynamically outputting a working condition specific threshold value and combining a working condition stability index. Therefore, on the basis of the hybrid deep learning and dynamic threshold technology, short-term peak detail features and long-term dependence features are fused, the peak fitting precision of the model is improved, it is ensured that the abnormal probability can accurately reflect the temperature abnormal degree, meanwhile, threshold self-adaptive dynamic adjustment under different working conditions is achieved, the false alarm rate under the strong turbulence working condition is reduced, and the working efficiency is improved. And accurate decision support is provided for operation and maintenance personnel.

The invention relates to the technical field of speech recognition, and discloses a speech recognition method and device, equipment and a storage medium. Aiming at the defects of fixed filter coefficient, static context modeling and difficulty in deep training and deployment of the existing deep feed-forward sequence memory network (DFSMN), the method comprises the following steps of: acquiring a frame-level acoustic feature sequence of input voice, and inputting an acoustic model containing a neural network unit; in the unit, the filter weight of the memory module is generated in real time through a parameter generation network based on the intermediate feature of the current frame; performing weighted aggregation on the historical and future frame features by using the weight to obtain a context enhancement feature; and outputting features based on the feature generation unit, and decoding to obtain an identification result. According to the method, context dynamic self-adaptive modeling is realized, the recognition precision and robustness are improved, the efficient reasoning characteristic of a pure feed-forward network is kept, the calculation overhead is not remarkably increased, and the method is adaptive to a server and terminal equipment and is suitable for multi-scene speech recognition of keywords, command words and the like.

The invention discloses a voice control optimization method and system based on big data. The method comprises the steps of voice data acquisition, voice recognition model construction, voice recognition model training and voice intelligent control. The invention relates to the technical field of voice data processing, in particular to a voice control optimization method and system based on big data, which innovatively adopts a three-channel parallel convolution structure, can simultaneously capture local frame association, medium-range dependency and long-range semantic information of voice signals, and improves the discrimination and robustness of voice feature representation; a deep feed-forward sequence memory neural network with improved hierarchical memory size design and interlayer jump connection structure is innovatively introduced, and the voice recognition continuity, context consistency and overall recognition precision are remarkably improved; a semantic understanding model is introduced in a speech recognition model training stage, and a joint loss function is constructed, so that the semantic consistency and speech recognition accuracy of the model in a complex context are remarkably improved.

A management system includes: a management sequence memory portion that stores, for each valve or each combination of valves to be managed among ON-OFF valves controlled by a sequence to be managed, definition information of a deviation in operation timing and a prescribed value of a deviation for an ON-OFF valve to be managed; an operation result information acquisition portion that acquires an actual value of an opening or closing time point of the valve to be managed; time difference determination portions that calculate a deviation in operation timing for each valve or each combination of valves to be managed based on the definition information and an actual value of the opening or closing time point, and determines whether a deviation in operation timing calculated is within a range of the prescribed value; and a determination result output portion that outputs a determination result by the time difference determination portions.

The application relates to a low-frame-rate noise adaptive speech enhancement system, method and server, and belongs to the technical field of communication, which comprises the following modules: a speech acquisition and preprocessing module, which is used for receiving and processing a call speech signal; a noise perception module, which is used for extracting and constructing feature information of background noise to form a noise environment feature vector based on the low-frame-rate speech data; a sequence memory enhancement module, which is used for combining the low-frame-rate speech data and the noise environment feature vector to perform noise adaptive speech enhancement processing; a semantic recognition and classification module, which is used for transcribing the enhanced speech signal into text and performing semantic classification; a rule decision module, which is used for performing corresponding interception operations according to preset interception rules; and an optimization module, which is used for updating and optimizing the interception rules and the classification model according to interception result feedback and combining artificial quality inspection marking. The system can be used for adaptive and fine speech enhancement.

The invention relates to the technical field of pulse discrimination, and provides an alpha / beta pulse discrimination method and system based on waveform trend difference, and the method comprises the steps: collecting alpha / beta pulse waveform data; inputting the alpha / beta pulse waveform data into a decay time constant adaptive multi-scale convolutional network for multi-scale waveform feature extraction to obtain a multi-scale waveform feature vector, and inputting the multi-scale waveform feature vector into a nuclear radiation pulse time sequence memory network for time sequence correlation analysis to obtain a preliminary alpha / beta classification probability; and combining the nuclear radiation pulse time sequence memory network with the attenuation time constant adaptive multi-scale convolutional network to obtain a teacher network, performing model compression on the teacher network through progressive knowledge distillation to obtain a lightweight teacher network model, deploying the lightweight teacher network model to an FPGA platform, inputting a to-be-discriminated alpha / beta pulse waveform into the lightweight teacher network model, and determining the to-be-discriminated alpha / beta pulse waveform. And performing real-time discrimination processing on the preliminary alpha / beta classification probability of the alpha / beta pulse waveform to be discriminated, and outputting a classification result and confidence. According to the invention, the accuracy and anti-interference capability of alpha / beta pulse discrimination are improved.

The invention discloses a low-complexity large-model long-sequence memory modeling method. The method comprises the following steps: firstly, constructing a fractional order state space model FOSSM by introducing a Capto fractional order derivative to replace an integer order differential operator; then, based on an asymptotic behavior of a Mittag-Leffler function, establishing a theoretical relationship between a power law attenuation property of an FOSSM memory kernel and a fractional order parameter; deducing the theoretical advantage boundary of the FOSSM in the aspect of power law data approximation compared with the standard SSM; and finally, the non-rational fractional order operator is converted into a rational transfer function through an Oustaloup rational approximation method, and efficient discretization of calculation complexity is realized. According to the method, power law long-range dependence can be modeled naturally, more matched mathematical representation is provided for a sequence with heavy-tailed time correlation, and the problem of path repetition caused by insufficient historical track memory in unmanned equipment navigation and the problem of poor interaction continuity caused by weak context keeping ability in man-machine interaction in a traditional method are solved.

The invention relates to a sequence memory guided multi-time sequence anomaly detection method, which comprises the following steps of: inputting a to-be-detected sequence X into a two-stage reconstruction model obtained by two-stage training to obtain a reconstruction sequence Y; taking the error of the last time node of the reconstructed sequence Y and the sequence X to be detected as a reconstruction error; and comparing the reconstruction error with an abnormal threshold value tau, if the reconstruction error is greater than or equal to the abnormal threshold value tau, judging the current time point as an abnormal time point, and if the reconstruction error is smaller than the abnormal threshold value tau, judging the current time point as a normal time point. According to the method, the abnormity in the multiple time sequences can be accurately detected by using a lightweight network, the method is suitable for various devices with resource limitation, and the model can be adjusted in real time to adapt to data distribution change of the multiple time sequences along with time lapse.

This invention discloses a low-instruction-overhead, reconfigurable normalization operator implementation method based on SIMD vector processing, relating to deep learning algorithm model technology and electronic information technology. The invention first divides the normalization operator into reduction and scalar operations, compiles the operations into a fixed sequence of M microcodes, preloads them into a microcode sequence memory, and configures a SIMD vector processing unit consisting of an instruction decoding unit, a hardware autonomous execution engine, and a memory access address generator. It continuously performs sequence length token and channel group scanning, performing P×N iterations on P groups of channels and N tokens, physically reusing the microcode sequence. The microcode pointer is automatically reset to zero, and intermediate result fusion is adaptively completed by scalar microcode operator encoding. Compared with existing GPU solutions, this method has better reconfigurability, lower execution cycle, and lower instruction overhead.

The invention discloses a non-intrusive dynamic acquisition method for WeChat 4.0 and above version database keys. The method comprises the following steps: acquiring a key memory address corresponding to a key dynamic library; obtaining a sequence memory address of a key dynamic library where the target instruction machine code sequence is located; obtaining an address relative offset according to the sequence memory address and the first base address; and obtaining a target breakpoint absolute address according to the address relative offset and the first base address, setting a breakpoint according to the target breakpoint absolute address, obtaining address data at the corresponding breakpoint, and performing hexadecimal conversion on the obtained address data to obtain a WeChat key, thereby realizing non-intrusive dynamic obtaining of the database key. According to the method, the problem that the Wechat 4.0 and above version database key cannot be effectively obtained by a traditional key obtaining method is solved.

The invention discloses a universal molecular pump life prediction method based on a neural network and related equipment. The method comprises the following steps: firstly, splitting a molecular pump into component-level faults, periodically collecting local displacement data, extracting strong correlation indexes by utilizing correlation analysis, forming a large-scale multi-working-condition time sequence data set with weak labels, and relieving data scarcity; and constructing a life prediction model by taking Transform as a core, overcoming CNN / RNN long sequence memory bottleneck and gradient problems, outputting life consumption percentage, and realizing accurate prediction.

The application relates to the technical field of wind power intelligent operation and maintenance, and discloses a wind turbine temperature early warning method based on hybrid deep learning and dynamic threshold values, wherein through a hybrid deep learning anomaly detection model, time sequence feature extraction and sequence memory enhancement are performed on synchronous time sequence data, wind turbine fusion features are obtained by fusing fine-grained short-term features and coarse-grained long-term features, the abnormal probability of each moment is predicted, the real-time working condition category is identified, the working condition specificity threshold value is dynamically output, the working condition stability index is combined, and a wind turbine temperature early warning signal is generated. Therefore, based on the hybrid deep learning and dynamic threshold value technology, the short-term peak detail features and the long-term dependence features are fused, the model peak fitting precision is improved, the abnormal probability can accurately reflect the temperature abnormality degree, meanwhile, the threshold value is adaptively and dynamically adjusted under different working conditions, the false positive rate of the strong turbulent flow condition is reduced, and accurate decision support is provided for operation and maintenance personnel.

A circuit and method of use thereof is provided, the circuit comprising: a first intermediate memory communicatively coupled with a vector processor and a RAM, wherein the vector processor is communicatively coupled with the RAM; an address sequence memory to store non-linear RAM addresses corresponding to linear positions in the first intermediate memory; a data sequencer to read a first frame of data from the RAM to the first intermediate memory based on addresses stored in the address sequence memory; and the first intermediate memory to provide the linearized frame of data to the vector processor to execute a vector instruction.