杏鲍菇温室开关式空调的无静差模糊控制器设计与试验

针对杏鲍菇工厂化生产中开关式空调温度控制精度较低的问题,考虑到常规模糊控制存在静态误差、静态模糊规则不适用于多温室及不断变化的外界环境的不足,该文提出了一种新型基于模糊控制的自适应变论域控制策略.在控制过程中,2个子模糊控制器受区域决策环节控制切换运行,自适应环节以开关式空调的启停周期作为滚动优化单元.当系统完成1个启停周期后,自适应环节根据该周期2个子模糊控制器所属区域的超调量分别校正各自模糊论域的中心点.通过多组对照试验表明,在北方夏季不同温室、自然因素等多变量试验条件下,设定1℃的范围温差,含自适应环节的模糊控制器均表现出良好的动态特性,与传统静态规则模糊控制器的效果相比,超调量平均下降6.5%,个别温室可下降15%以上,并保持在±5%以内.该方法在理论上实现了开关式空调的无静差模糊控制,试验中同样表现出了较好控制效果,且实现方法简单,普遍适用于开关式设备的温度控制.

Design and test of switch air-conditioner's zero static error fuzzy controller for eryngii greenhouse

Abstract: The greenhouse which breeds industrialized eryngii has the characteristics of large number and small size. Among all the environmental indicators in the greenhouse, the temperature is undoubtedly the most important. Because of cost reasons, most of the factory multi-greenhouse eryngii breeding adopts switch air-conditioning as its temperature control equipment Although the cost of this temperature control equipment is high, control is simple; however, control accuracy is lower and cannot meet the demand for the production of crops demanding temperature requirements. In addition, the control devices between multiple greenhouses exhibit individual differences, and the various uncertainties in the external environment continue to influence the greenhouse temperature changing parameters. The static control algorithm, therefore, will inevitably lead to the emergence of the static error. In view of this situation, this paper proposes a control strategy based on an adaptive variable universal fuzzy controller. This paper is organized as follows. Section 0 presents the purpose of this paper and current research situation. In Section 1, the change law of greenhouse's temperature, under the synthesized effect of Switch-air conditioner and variables, is analyzed. Fuzzy controller's inner structures and their design method are detailed in section 2. It also gives the membership grade of the fuzzy controller for the actuator. Section 3 proposes 3 steps to fulfill the adaptive algorithm under a list of limited conditions. In section 4, we do two sets of controlled trials in different conditions and then analyze the results. In section 5, a conclusion section summarizes the main point of the paper. The controlling structure includes the regional decision-making process, two sub-fuzzy controllers, and adaptive aspects. In the control process, the current temperature and the temperature change serve as the controller of the two input. Two sub-fuzzy controller region decision-making processes controls the switching operation, adaptive link to a start and stop cycle of switching the air-conditioning as the rolling optimization unit. At the end of a cycle, the adaptive link blurs according to the current cycle of the two sub-controllers which belong to regions respectively overshoot correction center point of each fuzzy domain to enter a new round of the control cycle. In this paper, we do two sets of controlled trials, the first group of 10 greenhouse before and after joining the adaptive link three cycles of fuzzy control overshoot controlled trials. The test results before and after joining the adaptive link the system overshoot average declined 6.5%, the largest decline was 15%, and 90% of the greenhouse in the second cycle overshoot can be decreased to less than ± 5%.The second set is static error for fuzzy control of a large greenhouse in controlled trials of 10 cycles before and after joining the static error; the test results show all open areas before joining the adaptive link overshoot in more than 15%, there are two periodic overshoot of 25%, and the closing region of the overshoot is substantially located in the vicinity of 10%. After adding adaptive link after a cycle, only cycle 2 and cycle 5 overshoot was10%, the opening of all the remaining cycle, and the close area overshoot maintained within ± 5%. The method to achieve this is simple, without the host computer doing a lot of computing work, and is particularly suitable for the control of the factory multiple greenhouse switch conditioning temperature.