渠道自动运行鲁棒控制

渠道运行控制系统是根据渠道系统的信息来控制渠道水位、流量的基本系统。目的在于提高渠系的全面调度运行水平,改善输水效率,实现适时适量地供水,避免供水的不足与浪费,降低调度运行费用,从而达到提高水资源利用率的目的,并为用水单位提供较好的服务。 渠道闸门关启时,带来很大的回波和扰动,传感器测量误差,自然界突发事件、渠道控制模型参数变量等都会带来系统的不确定性。作者首次将鲁棒控制应用在渠道运行控制当中,旨在系统性能最大满足和不确定性的抑制之间寻求调节及优化方案。 本论文根据渠道运行系统多输入多输出、运行状况多变、特点,推导了渠道运行控制系统的数学模型。描述渠道系统的控制过程采用的是圣·维南方程,为了便于用鲁棒控制理论解决渠道运行控制问题,对圣·维南方程作了离散化和线性化的处理。离散化的方法是明渠非恒定流数值解中的插值方法,线性化方法采用的是控制论中常用的台劳级数展开方法。为了实现渠道鲁棒控制,将水流的过渡过程用状态空间方法进行描述,由于多个渠池耦合时,系统状态矩阵过大,为后面鲁棒控制的优化与计算造成困难,本文采用了基于截断的均衡模型降阶方法,从而为后面渠道鲁棒控制铺路,为了对控制过程有更清楚的了解,并用状态空间进行了仿真。 然后,本论文介绍了H∞控制思想:这种摄动系统可以理解成取自某一个集合,那么实际被控对象可以描述为一个系统集（Σ0,△Σ）。其中,Σ0是模型的理想部分成为标称系统;△Σ表示不确定性因素所构成的某个可描述集。这样,实际系统Σ可以看成是标称系统Σ0与不确定因素△Σ的集合。本论文讨论了渠系闸门驱动环节的参数摄动,着重研究了水流流量与节制闸闸孔流量系数、闸孔开度Um、上下游水位差△h之间的摄动关系,看出由于闸孔开度Um,上下游水位差的扰动而导致水流流量△h的不确定性,从而可将不确定因素分离,实施抑制手段,使流量的不确定性限制在一定的范围之内;同时本文也讨论了非参数摄动界函数的确定,各种处理系统摄动机
逆变焊机
The canal operating control system is the basic system which controls the canal water level and the current according to the canal system information. The goal lies in enhancing the comprehensive dispatch movement level of ditch department, the improvement of water distribution efficiency, realization of timely water supply, avoiding the insufficiency and the waste of water supply, reducing the dispatch operating cost, thus we raise use efficiency of water, and provide good services for the water-used unit.When strobe of canal is closed, there are many echoes and disturbs. The sudden event in nature, parameter variation in the canal control model all will bring the system uncertainty. The author introduces the robust control into the control of canal operation for the first time, in order to find the adjusting and optimization method between the maximum satisfaction and restrain of uncertainty.According to multi-inputs, multi-outputs, variable characteristics of the canal running system, the present paper gives the mathematical model of the canal operation control system. Canal system’s controlling process is described by Saint Uygur equation. In order to handle the canal operation control problem by robust control theory, the equation is discreted and linearized. The discretization is done with interpolation method in non-steady flow numerical solution, the linearized is done with Taylor fatigue series expansion method in the cybernetics. In order to realize the canal’s robust control, the fluent transient process is described by the state space method . As when many ditches ponds couple, system mode matrix will be oversized , which brings the difficultyies in the robust control’s the optimization and the computation. Therefore , the step method based on the interruption balanced model is used in this article .thus paves the way for the following canal’s robust control so as to to have a clearer understanding on the controlled process, and the simulation with the state space, is also carried on next.Then, this paper discusses the control thought of H∞. This kind of perturbation system may be understood as coming from some set, then the real-controlled objects can be described as a system set （∑0, △∑） .Among them, ∑0 is the model’ ideal part which is called the nominal system; The △∑ expresses some describable set which is formed by uncertain factors. Thus, the actual system ∑ may be regard as the nominal system∑0 and the set formed by uncertain factor AS . This paper discusses parameter perturbation of the strobe ditch drive, studies the fluent current and coefficient of floodgate hole discharge, openness Um of floodgate, perturbation relations of difference between the upstream water level and downstream water level.Knowing that Floodgate openness Um , and perturbation relations of difference between the upstream and downstream water level result the uncertainty of A/i, so we can separate the uncertain factors, and do some compression method to restrict the uncertainty of flood.At the same time this article also discusses the determination of non-parameter perturbation function determination, each kind of processing system perturbation mechanism, the composition perturbation system structural style, then we proposed the mix sensitivity design method, when designing controller, the system performance should be requested to have: robust stability of system, the foreign disturbance sensitivity, the system’s parameter robustness and so on, these two are especially important. If we want to reduces the feedback control system’s the sensitivity to foreign disturbance and enhance the system’s robust stability, we needs to makes up the reasonable choice between sensitivity function T and in the effective band sensitivity function S . The optimized design method of HJs merit lies in that it can choose the reasonable selection weight function according to the different frequency band control systems’ task, it may cause the system’s the sensitivity function and makes- up sensitivity function change, in order to satisfy the performance index requiry which the system needs and to balance the relations between the stability and sensitivity of the system.This method is applied to design the controller HM, the control system simulation on //? is carried on , in order to test the controller and the system performance and effect design. In the simulation, follows are done: The weight function design, the augmentation system model’s establishment, design of the controller H^, the mix sensitivity design and the control system performance adjustment , the vote performance parameter gamma y , test of controller’s stablity, system simulation repeatedly design, simulation. Compared with the state space, this system performance is more stable.For the usage of H^ control in the canal control can bring very large conservative property, therefore u analysis synthesis design method and its thought are studied next. The process of designing controller is designing stable controller K in order to keep system’s stability and capacity called comprehensive u. This paper discuss the three capacity: （1） nominal performance （2）system’s robust stability （3）system’s robust ability.Then the paper optimizes the controller Hm and does w analysis at the sametime through the effective approximation method "D-K iteration" based on the u integration, first of all system’s ability y is optimized through //analysis, then controller is designed on J iteration. Through simulation , we can see that the robust of system is strong than W. control.In the end, using the linear matrix inequality （LMI） control method, we discuss the control of the ditch system, the controller optimization, and the type of uncertainty model. Through the discussion of the model, the canal operating control system may be designed more effectively and precisely. A complex dynamic system may be described using a simple model, for example south saint Uygur equation’s linearization, but the difference between this kind of simplification model and the actual object is model’s uncertainty in fact. Simplification of model will bring the model’s uncertainty, besides, unknown to system certain characteristics and tache, as a result of the natural environment change, the users randomly fetch the water, and open sub-drainage, these may creates the perturbation, and various ditches ponds’ water level’s change with time and other unknown factors also will bring the system’s change, then it causes more model’s uncertainty. The model’s uncertainty mainly conclude the multi- butchers mold and the affine parameter dependence model:（1） system’s uncertainty: For example dynamic characteristic which is neglected in the linear model, because the neglect of time-change characteristic or the simplification of non-linearity factor of input causes the dynamic behavior’s change（2） parameter’s uncertainty: Some difficulty with precise description physical parameter which is difficult to be precisely described, Or changes occurs in the movement process, it’s difficult to determine the change of the parameter.Through the research on uncertain system model, such as uncertain system’s robust analysis, we mainly consider to add some non-linearity into the linear model, even one kind of dynamic canal control system whose domain of time doesn’t change, where linear model is the system’s nominal, just the model which is obtained neglecting the uncertainty.The description on system nominal model and system’s uncertainty affect the system’s ability directly, in addition if we can know more uncertain information, the system’s ability achieved is more real.Through the robust stability analysis, and the introduction of the concept of affine square stability, the definite range of uncertainty’s variation, the change speed and its definition method have been studied. A set of complete design method of robust control on canal operation is established.In the summary and the forecast, the discussion on water conservation large-scale systems disperser’s control theory is carried on, when the large-scale water conservation canal moves, like the northward rerouting of southern river project, each ditch pond running status affects mutually, running time relatively lag, system space is largely scaled, structure is complicated. Regarding these large-scale systems, the centralized control will cause the entire control system’s complexity of information exchange , thus it raises the cost of the system integration and its running, system’s reliability also reduces. With system’s timeliness, reliability and efficiency considered, large-scale system’s disperser control method can be adopted.The disperser control is using partial subsystems which compose large-scale systems to design certain local controllers, so as to realise the entire ditch department’s control Each subsystem may by a controller which is composed by three ditches sections, these controllers are controlled by the master control-room.