癫痫是一种反复发作的短暂性脑功能障碍综合征，其具体发病机制仍未完全阐明。为探究星形胶质细胞内钙离子振荡与谷氨酸浓度变化在癫痫发作中的关键作用，基于Watts-Strogatz小世界网络模型构建了神经元之间的连接关系，并建立了一种更侧重于生物学中细胞-细胞耦合机制的星形胶质细胞-神经元耦合网络模型。接着构建了基于星形胶质细胞的癫痫发作脑网络耦合动力学模型，并通过希尔伯特变换计算相位锁定值（phase-locking value, PLV），量化模型输出主频率的相位同步特性。研究结果表明，随着耦合强度的增加，网络同步性呈现出从高同步到去同步化，再到全局同步的非单调变化模式。此外，不同放电模式的组合可有效增强神经元群体的同步性，且放电模式与网络结构之间存在复杂的耦合关系。在规则网络中，胶质细胞内的谷氨酸浓度较高，而在随机网络中，这种关联性显著减弱。该研究不仅加深了对癫痫发作机制的理解，也为癫痫治疗策略的优化提供了理论依据。

Epilepsy is a recurrent transient brain dysfunction syndrome, and its underlying pathophysiological mechanisms remain incompletely understood. To investigate the critical role of calcium ion oscillations and glutamate concentration variations in astrocytes during epileptic seizures, a neuronal connectivity framework was established based on the Watts-Strogatz small-world network model. Subsequently, a neuron-astrocyte coupling network model was developed, with a particular emphasis on cell-to-cell coupling mechanisms in biological systems. Furthermore, an epilepsy brain network coupling dynamics model incorporating astrocytes was constructed. PLV (phase-locking value) was calculated using the Hilbert transform to quantify the phase synchronization of the dominant frequency in the model output. The results demonstrate that as coupling strength increases, the network exhibits a non-monotonic transition from high synchronization to desynchronization and then to global synchronization. Additionally, the combination of different discharge patterns effectively enhances neuronal population synchrony, and a complex coupling relationship is observed between discharge patterns and network structure. Specifically, higher glutamate concentrations are detected in astrocytes within regular networks, whereas this association is significantly weakened in random networks. This study provides deeper insights into the mechanisms underlying epileptic seizures and offers a theoretical foundation for optimizing epilepsy treatment strategies.