The scientific and reasonable evaluation of the carrying capacity of water resources is of guiding significance for solving the issues of water resource shortages and pollution control. It is also an important method for realizing the sustainable development of water resources. Aiming at an evaluation of the carrying capacity of water resources, an evaluation model based on the cloud model theory and evidential reasoning approach is studied. First, based on the existing indicators, a water resources evaluation index system based on the pressure-state-response (PSR) model is constructed, and a classification method of carrying capacity grade is designed. The cloud model theory is used to realize the transformation between the measured value of indicators and the degree of correlation. Second, to obtain the weight of the evaluation index, the weight method of the index weights model based on the entropy weight method and evidential reasoning approach is proposed. Then, the reliability distribution function of the evaluation index and the graded probability distribution of the carrying capacity of water resources are obtained by an evidential reasoning approach. Finally, the evaluation method of the carrying capacity of water resources is constructed, and specific steps are provided. The proposed method is applied to the evaluation of water resources carrying capacity for Hunan Province, which verifies the feasibility and effectiveness of the method proposed in the present study. This paper applies this method of the evaluation of the water resources carrying capacity of Hunan Province from 2010 to 2019. It is concluded that the water resources carrying capacity of Hunan Province belongs to III~V, which is between the critical state and the strong carrying capacity state. The carrying capacity of the province’s water resources is basically on the rise. This shows that the carrying capacity of water resources in Hunan Province is in good condition, and corresponding protective measures should be taken to continue the current state.

The launch of large-scale bike sharing in China has effectively met the demand for low-carbon short-distance travel, but it has also led to the emergence of a large number of abandoned bikes, which is called the “bike-sharing siege”. In order to achieve the sustainable development of the bike-sharing industry, this paper discusses how to encourage bike-sharing companies to recycle and reuse abandoned bikes and improve the efficiency of local government financial resources based on the promotion of the EPR system. We apply the evolutionary game to investigate the interaction mechanism of complex behaviors between local governments and bike-sharing companies under two subsidy policies: a fixed recycling subsidy and regressive recycling subsidy. The results show that both recycling subsidy policies have diminishing marginal effects. In addition to incentives, local governments need to unify bike-sharing parking, establish bike-sharing monitoring platforms, and reasonably allocate bike parking spaces to ensure the quality of the urban environment. Under effective regulation, by implementing a regressive recycling subsidy policy and setting a rational amount for the recycling subsidy, local governments can effectively promote the EPR system, strengthen the environmental responsibility of companies, and avoid the “free-rider phenomenon”. Based on the current situation in China, this research provides references for each local government to formulate management policies from the perspective of subsidy policy.

In modern high-demand applications, such as real-time video communication, cloud gaming, and high-definition live streaming, achieving both superior transmission speed and high visual fidelity is paramount. However, unstable networks and packet loss remain major bottlenecks, making accurate and low-latency video error concealment a critical challenge. Traditional error control strategies, such as Forward Error Correction (FEC) and Automatic Repeat Request (ARQ), often introduce excessive latency or bandwidth overhead. Meanwhile, receiver-side concealment methods struggle under high motion or significant packet loss, motivating the exploration of predictive models. SimVP-v2, with its efficient convolutional architecture and Gated Spatiotemporal Attention (GSTA) mechanism, provides a strong baseline by reducing complexity and achieving competitive prediction performance. Despite its merits, SimVP-v2’s reliance on 2D convolutions for implicit temporal aggregation limits its capacity to capture complex motion trajectories and long-term dependencies. This often results in artifacts such as motion blur, detail loss, and accumulated errors. Furthermore, its single-modality design ignores the complementary contextual cues embedded in the audio stream. To overcome these issues, we propose A3DSimVP (Audio- and 3D-Enhanced SimVP-v2), which integrates explicit spatio-temporal modeling with multimodal feature fusion. Architecturally, we replace the 2D depthwise separable convolutions within the GSTA module with their 3D counterparts, introducing a redesigned GSTA-3D module that significantly improves motion coherence across frames. Additionally, an efficient audio–visual fusion strategy supplements visual features with contextual audio guidance, thereby enhancing the model’s robustness and perceptual realism. We validate the effectiveness of A3DSimVP’s improvements through extensive experiments on the KTH dataset. Our model achieves a PSNR of 27.35 dB, surpassing the 27.04 of the SimVP-v2 baseline. Concurrently, our improved A3DSimVP model reduces the loss metrics on the KTH dataset, achieving an MSE of 43.82 and an MAE of 385.73, both lower than the baseline. Crucially, our LPIPS metric is substantially lowered to 0.22. These data tangibly confirm that A3DSimVP significantly enhances both structural fidelity and perceptual quality while maintaining high predictive accuracy. Notably, A3DSimVP attains faster inference speeds than the baseline with only a marginal increase in computational overhead. These results establish A3DSimVP as an efficient and robust solution for latency-critical video applications.

Federated learning has won a lot of interest in recent years, due to its capability in collaborative learning and privacy preservation. To ensure the accuracy of outsourced training tasks, task publishers prefer to assign tasks to task workers with a high reputation. However, existing reputation-based incentive mechanisms assume that task publishers are honest, and only task workers would probably behave dishonestly to pollute the federated learning model. Different from existing work, we argue that task publishers would also behave dishonestly, where they would benefit from colluding with task workers to help task workers obtain a high reputation. In this paper, we propose a collusion-resistant and reliable incentive mechanism for federated learning. First, to measure the credibility of both task publishers and task workers, we devise a novel metric named reliability. Second, we devise a new method to compute the task publisher reliability, which is obtained by computing the deviation of reputation scores given by different task publishers, i.e., low reliability is assigned to a task publisher once its deviation is far away from that of other publishers. Third, we propose a bidirectional reputation calculation method based on the basic uncertain information model to compute reputation and reputation reliability for task workers. Furthermore, by integrating an incentive mechanism, our proposed scheme not only effectively defends against collusion attacks but also ensures that only task workers with high reputation, reputation reliability, and the capability to accomplish complex tasks can win a high reward. Finally, we conduct extensive experiments to verify the efficiency and efficacy of our proposed schemes. The results demonstrate that our proposed schemes are not only collusion-resistant but also achieve 6.31% higher test accuracy compared with the state of the art on the MNIST dataset.