Purpose

Without damaging and consuming natural resources, green computing technology can meet the needs of society for a long time. This paper discusses how to realize the sustainable development of social economy through the innovation of green computing technology.

Design/methodology/approach

For the green computing technology and sustainable social and economic development problems, it builds back propagation (BP) neural network model and analyzes the topological structure of the network model as well as the impact of the training errors allowed by the network on its performance.

Findings

By optimizing the number of input nodes, the number of hidden nodes and the target value, the genetic algorithm (GA) can get the optimal neural network model. The simulation experiment proves that the proposed model is effective.

Originality/value

It can not only reduce the possibility of falling into local optimum, but also optimize the initial weights and thresholds of BP neural network and further improve the stability and test effect of BP neural network model.

中文翻译：

---

基于绿色计算技术和大数据的可持续社会经济发展启发式方法

目的

在不破坏和消耗自然资源的情况下，绿色计算技术可以长期满足社会需求。本文探讨了如何通过绿色计算技术的创新来实现社会经济的可持续发展。

设计/方法/方法

针对绿色计算技术和社会经济可持续发展问题，建立了反向传播（BP）神经网络模型，分析了网络模型的拓扑结构以及网络允许的训练误差对其性能的影响。

发现

通过优化输入节点数、隐藏节点数和目标值，遗传算法（GA）可以得到最优的神经网络模型。仿真实验证明所提出的模型是有效的。

原创性/价值

不仅可以降低陷入局部最优的可能性，还可以优化BP神经网络的初始权值和阈值，进一步提高BP神经网络模型的稳定性和测试效果。