✍️Prompt Instructions for Effective Outcomes
Prompt instructions are crucial for optimizing the performance of spiderAI™. While spiderAI™ leverages Large Language Models (LLMs) for enhanced diagnostics, it relies heavily on context and the quality of input data to generate precise and actionable recommendations. By providing clear, complete, and structured inputs, field engineers can enhance diagnostic accuracy, improve context awareness, facilitate better communication, speed up decision-making, and enable in-depth reasoning. Here are a few prompting best practices to follow when using spiderAI™:
1. Provide Complete Machine Details
Ensure to include all relevant machine specifications, drive train configuration, mounting details, and operational observations. For detailed instructions, refer to the previous page, "User Input Information for vibration analysis."
2. Give Inputs in Orders
Provide frequency observations in a systematic order related to the machine's rotating speed. For detailed instructions, refer to the previous page, "User Input Information for vibration analysis."
3. Request Specific Faults
When you have a specific fault in mind, ask spiderAI™ directly for a precise diagnosis of that fault.
Example:
User: Given the vibration data, can you check for unbalance issues specifically?4. Request Possible Faults
If uncertain about the exact fault, ask spiderAI™ for a group of potential faults to get a broader analysis.
Example:
User: Based on the provided vibration and phase data, what are the possible faults?5. Include Phase Analysis
Providing phase angle data can significantly enhance the diagnostic accuracy, especially for identifying unbalance and misalignment issues.
Example:
6. Clarify Doubtful Situations
If you're unsure about the data provided or the diagnosis, ask spiderAI™ for clarification.
Example:
7. Seek More Details
Ask SpiderAI™ for more detailed analysis or further data requirements if the initial diagnosis is inconclusive.
Example:
8. Ask for Additional Information
Inquire what additional data might help in refining the diagnosis, such as high-resolution spectrum or phase data.
Example:
SpiderAI™ Response:
9. Request Detailed Reasoning and Evidence
For a clear understanding of the diagnosis, ask SpiderAI™ for detailed reasoning and evidence supporting the conclusion.
Example:
SpiderAI™ Response:
10. Include Doubts or Additional Machine Details
If you have any doubts or additional machine details to be considered, include them in the prompt. This helps in providing context and refining the analysis.
Example:
By following these prompting best practices, you can ensure SpiderAI™ provides precise and actionable recommendations, enhancing your diagnostic and decision-making processes.
Last updated