Dynamic Multicast Filter for Substation Automation

The patent applies preliminary action by configuring packet filters in advance at secondary devices before multicast messages arrive. The filters are set up during system initialization or configuration phase, so when multicast messages flow through the network, the filtering is already in place. This prevents unnecessary processing load on devices while ensuring complete delivery of relevant messages, resolving the contradiction between message delivery completeness and processing efficiency.

The patent introduces packet filters as intermediary elements between the Ethernet network and application devices. These filters act as mediators that selectively pass or block multicast messages based on predefined criteria. This intermediary layer protects application devices from processing irrelevant messages while maintaining reliable delivery of necessary messages, thus improving processing efficiency without sacrificing message delivery completeness.

The patent applies dynamics by enabling packet filters to adapt their filtering criteria dynamically based on changing application states and operational conditions. Instead of using fixed static filters, the system continuously monitors application requirements and adjusts filter rules accordingly. This allows the network to maintain reduced load while adapting to varying message priorities, application states, and operational modes, thus resolving the contradiction between network load reduction and filter adaptability.

The patent implements feedback mechanisms where the system monitors application performance, message relevance, and operational states, then uses this information to dynamically adjust packet filter configurations. The feedback loop ensures that filters remain optimized for current conditions, maintaining effective network load reduction while adapting to changing requirements. This resolves the contradiction by making filters responsive to real-time conditions rather than static.

The patent applies the taking out principle by extracting and removing irrelevant multicast packets from the data stream before they reach the application processor. Packet filters at the network level selectively extract only the relevant messages needed by the application, discarding unnecessary packets. This extraction process prevents information loss of relevant data while eliminating the processing burden of irrelevant messages, thus resolving the contradiction between data completeness and processor capacity utilization.

The patent converts the potentially harmful effect of excessive processor load into a benefit by using packet filters to pre-process and reduce the message stream. The filtering mechanism transforms the problem of high processor demand into an opportunity to optimize network traffic flow, ensuring that only necessary messages reach the application. This conversion maintains data completeness for relevant messages while turning the processor capacity constraint into an advantage by eliminating wasteful processing, thus resolving the contradiction between data completeness and processor capacity.