基于模糊推理脉冲神经膜系统的电网故障通用诊断模型

为了提高故障诊断模型对拓扑结构经常变化的电网的适应能力,基于模糊推理脉冲神经膜系统,分别建立母线、线路和变压器3种元件电网拓扑结构变化而诊断模型结构保持的故障诊断模型.首先,采用模糊初始值表征可能不完备和不确定的告警数据.同时,根据故障区域拓扑结构和保护及断路器动作状态,对输入神经元进行\"归一\"预处理,以减小模型的复杂...     

Integrated neural and robotic simulations. Simulation of cerebellar neurobiological substrate for an object-oriented dynamic model abstraction process

Experimental studies of the Central Nervous System (CNS) at multiple organization levels aim at understanding how information is represented and processed by the brain’s neurobiological substrate. The information processed within different neural subsystems is neurocomputed using distributed and dynamic patterns of neural activity. These emerging patterns can be hardly understood by merely taking into account individual cell activities. Studying how these patterns are elicited in the CNS under specific behavioral tasks has become a groundbreaking research topic in system neuroscience. This methodology of synthetic behavioral experimentation is also motivated by the concept of embodied neuroscience, according to which the primary goal of the CNS is to solve/facilitate the body–environment interaction.With the aim to bridge the gap between system neuroscience and biological control, this paper presents how the CNS neural structures can be connected/integrated within a body agent; in particular, an efficient neural simulator based on EDLUT (Ros et al., 2006) has been integrated within a simulated robotic environment to facilitate the implementation of object manipulating closed loop experiments (action–perception loop). This kind of experiment allows the study of the neural abstraction process of dynamic models that occurs within our neural structures when manipulating objects.The neural simulator, communication interfaces, and a robot platform have been efficiently integrated enabling real time simulations. The cerebellum is thought to play a crucial role in human-body interaction with a primary function related to motor control which makes it the perfect candidate to start building an embodied nervous system as illustrated in the simulations performed in this work.     

On the power of elementary features in spiking neural P systems

Since their first publication in 2006, spiking neural (SN) P systems have already attracted the attention of a lot of researchers. This might be owing to the fact that this abstract computing device follows basic principles known from spiking neural nets, but its implementation is discrete, using membrane computing background. Among the elementary properties which confer SN P systems their computational power one can count the unbounded fan-in (indegree) and fan-out (outdegree) of each “neuron”, synchronicity of the whole system, the possibility of delaying and/or removing spikes in neurons, the capability of evaluating arbitrary regular expressions in neurons in constant time and some others. In this paper we focus on the power of these elementary features. Particularly, we study the power of the model when some of these features are disabled. Rather surprisingly, even very restricted SN P systems keep their universal computational power. Certain important questions regarding this topic still remain open.     

Characterizations of some classes of spiking neural P systems

We look at the recently introduced neural-like systems, called SN P systems. These systems incorporate the ideas of spiking neurons into membrane computing. We study various classes and characterize their computing power and complexity. In particular, we analyze asynchronous and sequential SN P systems and present some conditions under which they become (non-)universal. The non-universal variants are characterized by monotonic counter machines and partially blind counter machines and, hence, have many decidable properties. We also investigate the language-generating capability of SN P systems.     

Advances in Design and Application of Spiking Neural Networks

This paper presents new findings in the design and application of biologically plausible neural networks based on spiking neuron models, which represent a more plausible model of real biological neurons where time is considered as an important feature for information encoding and processing in the brain. The design approach consists of an evolutionary strategy based supervised training algorithm, newly developed by the authors, and the use of different biologically plausible neuronal models. A dynamic synapse (DS) based neuron model, a biologically more detailed model, and the spike response model (SRM) are investigated in order to demonstrate the efficacy of the proposed approach and to further our understanding of the computing capabilities of the nervous system. Unlike the conventional synapse, represented as a static entity with a fixed weight, employed in conventional and SRM-based neural networks, a DS is weightless and its strength changes upon the arrival of incoming input spikes. Therefore its efficacy depends on the temporal structure of the impinging spike trains. In the proposed approach, the training of the network free parameters is achieved using an evolutionary strategy where, instead of binary encoding, real values are used to encode the static and DS parameters which underlie the learning process. The results show that spiking neural networks based on both types of synapse are capable of learning non-linearly separable data by means of spatio-temporal encoding. Furthermore, a comparison of the obtained performance with classical neural networks (multi-layer perceptrons) is presented.     

基于脉冲神经网络的迁移学习算法与软件框架

使用脉冲序列进行数据处理的脉冲神经网络具有优异的低功耗特性,但由于学习算法不成熟,多层网络训练存在收敛困难的问题。利用反向传播网络具有学习算法成熟和训练速度快的特点,设计一种迁移学习算法。基于反向传播网络完成训练过程,并通过脉冲编码规则和自适应的权值映射关系,将训练结果迁移至脉冲神经网络。实验结果表明,在多层脉冲神经网络中,迁移学习算法能够有效解决训练过程中收敛困难的问题,在MNIST数据集和CIFAR-10数据集上的识别准确率分别达到98.56%和56.00%,且具有微瓦级别的低功耗特性。     

A biophysical implementation of a bidirectional graph search algorithm to solve multiple goal navigation tasks

The model presented here extends formal analysis (Hasselmo et al., Neural Networks, 15, pp. 689-707, 2002b) and abstract modelling (Gorchetchnikov and Hasselmo, Neurocomputing, 44-46, pp. 423-427, 2002a) of interactions within the hippocampal area (or other cortical areas), which can be flexibly used to navigate toward any arbitrary goal or multiple goals that change on a trial-by-trial basis. The algorithm is a version of a bidirectional breadth-first graph search implemented in simulated neurons using two flows of neural activity. The new model changes the continuous firing rate neuronal representations (Gorchetchnikov and Hasselmo 2002a) to more detailed compartmental versions with realistic parameters, while preserving the qualitative properties analysed previously (Hasselmo et al., 2002b, Gorchetchnikov and Hasselmo 2002a). The case of multiple goals being present in the environment is studied in this paper. The first set of simulations tests the algorithm in the selection of the closest goal. A small difference in distance between the simulated animal and different goals is sufficient for a correct selection. The second set of simulations studies the behaviour of the model when the goals have different saliences. A small salience-based difference between firing rates of the cells providing goal-related input to the model is sufficient for the selection of a more salient goal. This behaviour was tested in three types of environments: a linear track, a T-maze and an open field. Further investigation of quantitative properties of the model should allow it to handle cases when the exact location of the goal is uncertain.     

Long Short-Term Memory Spiking Neural Networks for Classification of Snoring and Non-snoring Sound Events

Snoring is a widespreadoccurrence that impacts human sleep quality.It is also one of the earliest symptoms of many sleep disorders.Snoring is accurately detected,making further screening and diagnosis of sleep problems easier.Snoring is frequently ignored because of its underrated and costly detection costs.As a result,this research offered an alternative method for snoring detection based on a long short-term memory based spiking neural network(LSTM-SNN)that is appropriate for large-scale home detection for snoring.We designed acquisition equipment to collect the sleep recordings of 54 subjects and constructed the sleep sound database in the home environment.And Mel frequency cepstral coefficients(MFCCs)were extracted from these sound signals and encoded into spike trains by a threshold encoding approach.They were classified automatically as non-snoring or snoring sounds by our LSTM-SNN model.We used the backpropagation algorithm based on an alternative gradient in the LSTM-SNN to complete the parameter update.The categorization percentage reached an impressive 93.4％,accom-panied by a remarkable 36.9％ reduction in computer power compared to the regular LSTM model.     

基于多层脉冲神经网络的非接触液位检测方法

尽管基于深度学习的非接触液位检测方法能够较好地完成检测任务,但其对计算资源的较高要求使其不适用于算力受限的嵌入式设备。为解决上述问题,该文首先提出了基于多层脉冲神经网络的非接触液位检测方法;其次,提出了单帧和帧差脉冲编码方法将视频流时间动态性编码成可重构的脉冲模式;最后在实际场景中对模型进行测试。实验结果表明,所提方法具有较高应用价值。