Transaction fees are a crucial revenue source for miners in public and consortium blockchains. However, while public blockchains have additional revenue streams, transaction fees serve as the primary income for miners in consortium blockchains formed by various financial institutions. These miners allocate different levels of computing resources to process transactions and earn corresponding fees. Nonetheless, relying solely on transaction fees can lead to significant volatility and encourage non-standard mining behaviors, thereby posing threats to the blockchain's security and integrity. Despite previous attempts to mitigate the impact of transaction fees on illicit mining behaviors, a comprehensive solution to this vulnerability is yet to be established. To address this gap, we introduce a novel approach that leverages Dynamic Transaction Storage (DTS) strategies to effectively minimize block incentive volatility. Our solution implements a Verkle tree-based storage mechanism to reduce bandwidth consumption. Moreover, to configure the DTS strategies, we evaluate several optimization algorithms and formulate the challenge as a Vehicle Routing Problem. Our experiments conducted using historical transactions from Bitcoin and remittance data from the Industrial and Commercial Bank of China reveal that the strategy focusing on time-based transaction incorporation priority, while excluding a designated space for small-fee transactions, as discovered by the gradient-based optimizer algorithm, proves most effective in reducing volatility. Hence, the DTS strategy can sustain stable block incentives irrespective of transaction types or user bidding behavior. Furthermore, the inclusion of higher-fee transactions, often smaller in size, can alleviate propagation delays and the occurrence of forks.