论文部分内容阅读
针对马铃薯存储现状及其加工企业原材料和人力资源的严重浪费问题,结合新疆大罗素农业科技开发有限公司马铃薯储藏库温度数据的非线性、时变性等特点,采用常规控制方法控制效果不理想。提出了新的温度预测控制方法,它是用粒子群优化算法滚动优化输入控制量并得到实际输出值,用ESNs对储藏库内温度进行预测,利用预测输出和实际输出的偏差来对系统反馈较正。仿真结果表明:提出的温度预测控制方法是有效的,不仅在控制效果上优于LS-SVM预测控制,还有良好自适应性以及对扰动信号有较好的鲁棒性。
In view of the status quo of potato storage and the serious waste of raw materials and human resources in processing enterprises, the control effect of conventional control methods is not satisfactory based on the characteristics of non-linearity and time-varying of the temperature data of potato storage in Darongsu Agricultural Science and Technology Development Co., Ltd. in Xinjiang. A new temperature predictive control method is proposed. It uses the particle swarm optimization algorithm to scroll and optimize the input control volume and obtain the actual output value. ESNs are used to predict the temperature in the storage. By using the forecast output and the actual output deviation, positive. The simulation results show that the proposed temperature predictive control method is effective, which is superior to the LS-SVM predictive control in control effects, and has good adaptability and good robustness to disturbance signals.