In the operating environment of high-voltage transmission lines, traditional monitoring methods are often difficult to provide accurate micro-meteorological data due to electromagnetic interference, which seriously affects the safety and stability of the line. The application of insulated optical unit optical cables puts forward higher requirements for the line operating environment requirements. In order to overcome this problem, this study pioneered a new method for online monitoring of microclimate in insulated optical cables. This method achieves high-precision signal collection of micrometeorological parameters by deploying fiber grating sensors, and uses interference demodulation technology to finely process the collected signals. On this basis, we carefully extracted three key features of temperature, humidity and wind speed from the demodulated signals, which can keenly capture changes in micrometeorological conditions. In order to further improve the accuracy of monitoring, we used wavelet neural network to build an intelligent monitoring model. The model takes the extracted features as input and uses a deep learning algorithm to accurately identify and predict different types of micrometeorological conditions. After rigorous testing and verification, our research method has demonstrated excellent performance, with its Jaccard coefficient significantly close to 1. The high value of this indicator fully proves the accuracy and reliability of this method in the field of microclimate monitoring. This innovative monitoring method will provide environmental protection for the application of insulated optical unit optical cables in the power grid, and provide strong technical support for the safe operation of high-voltage transmission lines.