关键词: 锂电产业链, 价格模态, FEEMD-NAR-HMM, 安全稳定, 同频共振

Abstract: In recent years, the global lithium-ion battery market has frequently encountered difficulties in supply and demand imbalances, leading to drastic fluctuations in the prices of key products within the industrial chain, having significant impacts on the security and stability of the lithium-ion battery industrial chain. Major new energy consuming and manufacturing countries, represented by the United States and China, have introduced various measures aimed at safeguarding the security and stability of the lithium-ion battery industrial chain and supply chain. Price, as the core signal in a market economy, plays a crucial role in the smooth transmission within the industrial chain, directly affecting its security. Therefore, from a macro perspective of the industrial chain, focusing on the degree of smooth transmission of price fluctuations across multiple nodes and delving into the intrinsic relationship between price fluctuations and the security and stability of the industrial chain, this research holds significant practical implications and provides valuable reference for promoting efficient coordination and sustainable development of the lithium-ion battery industrial chain.
   This paper views the lithium-ion battery industrial chain as a typical complex adaptive system with dissipative structures, dividing its dissipative structures into two subsystems: internal dissipation, i.e., the security and stability state of the industrial chain, and external dissipation, i.e., the degree of “synchronous resonance” in prices across nodes in the lithium-ion battery industrial chain. Based on the entropy flow model, the degree of synchronous resonance of the industrial chain with multiple prices is regarded as the external entropy flow, the safety and stability state as the internal entropy flow, and the collision results of the two entropy flows as the evolution results of the lithium-ion battery industrial chain system. Specifically, this paper first employs the fast ensemble empirical mode decomposition (FEEMD) algorithm and the nonlinear autoregressive (NAR) neural network to build a combined model, accurately predicting the market prices of three key nodes in the lithium-ion battery industrial chain: spodumene, lithium iron phosphate and lithium iron phosphate power batteries. Subsequently, the fluctuation states of the prices of these three nodes are combined into modal signals, and based on the Hidden Markov Model (HMM), conducting an in-depth assessment of the hidden state of safety and stability by observing the synchronous price resonance across multiple nodes in the industrial chain. Finally, this paper uses historical data from the Shanghai Metals Market in China as samples to verify the effectiveness and accuracy of the model.
   Through the prediction results of the FEEMD-NAR model, this paper finds that the price trends of spodumene, lithium iron phosphate and lithium iron phosphate power batteries all exhibit a trend of a decline first and then a rise, and as the supply and demand in downstream markets gradually balance, the market prices of midstream and upstream nodes will gradually return to a reasonable range. Furthermore, through modeling and solving using the HMM model, this paper reveals the following important findings: (1) Over the next period, with the prices of spodumene and lithium iron phosphate power batteries steadily rising, anode material lithium iron phosphate prices are more volatile and staggered fluctuations, leading to a reduction in the degree of synchronous resonance across the entire industrial chain and a decline in security and stability.(2) The degree of security and stability of the lithium-ion battery industrial chain is closely related to the smooth transmission of prices. When “synchronous resonance” occurs across the industrial chain, the level of security and stability will be high; whereas, when “asynchronous misalignment” occurs in the price mode combination, the level of security and stability of the industrial chain will be generally low.

Key words: lithium-ion battery industrial chain, price mode, FEEMD-NAR-HMM, security and stability, synchronous resonance