基于改进萤火虫算法优化SVM的变电工程造价预测

变电工程造价水平直接关系到电网工程的整体经济性,造价水平预测是控制造价、提高造价合理性的重要手段。在传统萤火虫算法的基础上,采用高斯扰动技术改进萤火种算法的位置更新公式,提高萤火从算法的寻优性能从而优化SVM预测模型的参数。通过Schaffer函数测试发现,高斯扰动萤火虫算法具有收敛速度快、搜索能力强等优点。实测结果表明：该模型具有较高的预测精度和有效性。

Substation Engineering Cost Forecasting Method Based on Modified Firefly Algorithm and Support Vector Machine

The cost level of substation engineering is closely related to the integrated economy of power grid projects, and the cost level forecasting is a crucial tool for controlling cost and improving cost rationality. Based on the conventional firefly algorithm, the Gaussian Disturbance is introduced into the firefly algorithm to improve the update equation, which aims to improve the searching ability and optimize the SVM parameters. By operating the Schaffer testing function, it is discovered that the Gaussian disturbance firefly algorithm has better convergence rate and searching ability. The case study of substation engineering in Guangdong Province further proves that the proposed model has higher forecasting accuracy and effectiveness