Leaky Integrator Echo State Network (Leaky-ESN) is a useful training method for handling time series prediction problems. However, the singular coupling of all neurons in the reservoir makes Leaky-ESN less effective for sophisticated learning tasks. In this paper, we propose a new improvement to the Leaky-ESN model called the Multiple-Reservoir Hierarchical Echo State Network (MH-ESN). By introducing a new mechanism for constructing the reservoir, the efficiency of the network in handling training tasks is improved. The hierarchical structure is used in the process of constructing the reservoir mechanism of MH-ESN. The MH-ESN consists of multiple layers, each comprising a multi-reservoir echo state network model. The sub-reservoirs within each layer are linked via principal neurons, which mimics the functioning of a biological neural network. As a result, the coupling among neurons in the reservoir is decreased, and the internal dynamics of the reservoir are improved. Based on the analysis results, the MH-ESN exhibits significantly better prediction accuracy than Leaky-ESN for complex time series prediction.
In the prediction of a nonlinear time series based on a leaky integrator echo state network (leaky-ESN), building a reservoir related to the specific problem is a key step. For problems such as poor performance of randomly generated reservoirs, it is tough to determine the parameter values of the reservoirs. The work in this paper uses the gravitational search algorithm (GSA) to optimize the global parameters of a leaky-ESN, such as the leaking rate, the spectral radius, and the input scaling factor. The basic GSA has some problems, such as slow convergence and poor balance between exploration and exploitation, and it cannot solve some complex optimization problems well. To solve these problems, an improved gravitational search algorithm (IGSA) is proposed in this paper. First, the best agent and elite agents were archived and utilized to accelerate the exploration phase and improve the convergence rate in the exploitation phase. Second, to improve the effect of the poor fitness agents on the optimization result, a differential mutation strategy was proposed, which generated new individuals to replace original agents with worse fitness, increasing the diversity of the population and improving the global optimization ability of the algorithm. Finally, two simulation experiments showed that the leaky-ESN optimized by the IGSA had better prediction accuracy.
As energy supply units, lithium-ion batteries have been widely used in the electric vehicle industry. However, the safety of lithium-ion batteries remains a significant factor limiting their development. To achieve rapid fault diagnosis of lithium-ion batteries, this paper presents a comprehensive fault diagnosis process. Firstly, an interleaved voltage sensor topology structure is utilized to acquire battery voltage data. An improved complete ensemble empirical mode decomposition with adaptive noise method is introduced to process data. Then, the reconstructed voltage data sequence is used to eliminate the influence of noise. A fault location is performed using dichotomy correlation coefficient and time window correlation coefficient. Afterwards, principal component analysis is used to select the principal components with high contribution rate as classification features. The gray wolf optimization algorithm is used to find the parameters of the least squares support vector machine, constructing an optimal classifier for fault classification. A fault experiment platform is established to realize the physical triggering of faults such as external short circuit, internal circuit, and connection of experimental battery packs. Finally, the accuracy and reliability of the method are verified by the results of fault localization and fault type determination.