当前位置: X-MOL 学术Int. Trans. Electr. Energy Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Development of online demand response framework for smart grid infrastructure toward social welfare
International Transactions on Electrical Energy Systems ( IF 1.9 ) Pub Date : 2021-04-23 , DOI: 10.1002/2050-7038.12909
Marimuthu Krishna Paramathma 1 , Durairaj Devaraj 1 , Velusamy Agnes Idhaya Selvi 1 , Murugesan Karuppasamypandiyan 1
Affiliation  

Due to continuous growth in the demand for electricity with unmatched generation and transmission capacity expansion, the resource management and rescheduling of load without affecting the welfare of the market participants are the major concerns of the power market. As the demand changes continuously, the peak load consumers are unaware of the bidding cost and penalty. The Artificial Neural Network (ANN) based online Demand Response (DR) connectivity scheme is proposed for the smart power networks to obtain the equilibrium demand. The optimally rescheduled load, percentage increase of peak load, and time are considered the ANN input. Bidding cost and penalty of the peak load consumer are considered as the output. The data required to develop the ANN are generated using the Genetic Algorithm (GA) to maximize social welfare as the objective. The optimum load curtailment is taken as the decision variable. In this proposed method, the Curtailment Index (CI) is calculated and incorporated to utilize DR connectivity properly. This adopted method is tested with IEEE 30 bus system, and the GA results for CI and bidding cost have been compared with Particle Swarm Optimization (PSO) methodology. The ANN predicted bidding cost results are compared with GA optimized bidding cost. The result shows the accuracy of ANN for online DR techniques with minimum testing Mean Square Error (MSE) value of 1.72 × 10−3 and the training period of 45.98 seconds.

中文翻译:

面向社会福利的智能电网基础设施在线需求响应框架的开发

由于电力需求持续增长,发电和输电容量扩张不匹配,在不影响市场参与者福利的情况下进行资源管理和重新调度是电力市场的主要关注点。随着需求的不断变化,高峰负荷消费者并不知道投标成本和惩罚。提出了基于人工神经网络(ANN)的在线需求响应(DR)连接方案,用于智能电网以获得均衡需求。最佳重新调度的负载、峰值负载的百分比增加和时间被认为是 ANN 输入。峰值负载消费者的投标成本和惩罚被视为输出。开发 ANN 所需的数据是使用遗传算法 (GA) 生成的,以最大化社会福利为目标。将最佳负荷削减作为决策变量。在这个提议的方法中,计算并结合了弃电指数 (CI) 以正确利用 DR 连接。这种采用的方法在 IEEE 30 总线系统上进行了测试,并将 CI 和投标成本的 GA 结果与粒子群优化 (PSO) 方法进行了比较。ANN 预测的投标成本结果与 GA 优化投标成本进行比较。结果表明 ANN 对于在线 DR 技术的准确性,最小测试均方误差 (MSE) 值为 1.72 × 10 并且已经将 CI 和投标成本的 GA 结果与粒子群优化 (PSO) 方法进行了比较。ANN 预测的投标成本结果与 GA 优化投标成本进行比较。结果表明 ANN 对于在线 DR 技术的准确性,最小测试均方误差 (MSE) 值为 1.72 × 10 并且已经将 CI 和投标成本的 GA 结果与粒子群优化 (PSO) 方法进行了比较。ANN 预测的投标成本结果与 GA 优化投标成本进行比较。结果表明 ANN 对于在线 DR 技术的准确性,最小测试均方误差 (MSE) 值为 1.72 × 10−3和 45.98 秒的训练周期。
更新日期:2021-07-02
down
wechat
bug