CN109197539B - Model predictive control algorithm-based automatic control method for irrigation multi-stage channels - Google Patents
Model predictive control algorithm-based automatic control method for irrigation multi-stage channels Download PDFInfo
- Publication number
- CN109197539B CN109197539B CN201811250103.8A CN201811250103A CN109197539B CN 109197539 B CN109197539 B CN 109197539B CN 201811250103 A CN201811250103 A CN 201811250103A CN 109197539 B CN109197539 B CN 109197539B
- Authority
- CN
- China
- Prior art keywords
- control
- channel
- constraint
- matrix
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002262 irrigation Effects 0.000 title claims abstract description 56
- 238000003973 irrigation Methods 0.000 title claims abstract description 56
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 52
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 148
- 238000013461 design Methods 0.000 claims abstract description 29
- 238000005457 optimization Methods 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 81
- 230000008569 process Effects 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 6
- 230000007423 decrease Effects 0.000 claims description 3
- 238000005096 rolling process Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000012384 transportation and delivery Methods 0.000 abstract description 15
- 238000007726 management method Methods 0.000 abstract description 8
- 230000009471 action Effects 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 239000003621 irrigation water Substances 0.000 description 2
- 238000004886 process control Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- IUTDVGSJRKTQPM-UHFFFAOYSA-N [4-(1,3-benzothiazol-2-yl)phenyl]boronic acid Chemical compound C1=CC(B(O)O)=CC=C1C1=NC2=CC=CC=C2S1 IUTDVGSJRKTQPM-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012887 quadratic function Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01G—HORTICULTURE; CULTIVATION OF VEGETABLES, FLOWERS, RICE, FRUIT, VINES, HOPS OR SEAWEED; FORESTRY; WATERING
- A01G25/00—Watering gardens, fields, sports grounds or the like
- A01G25/16—Control of watering
Landscapes
- Engineering & Computer Science (AREA)
- Water Supply & Treatment (AREA)
- Life Sciences & Earth Sciences (AREA)
- Environmental Sciences (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides an automatic control method for irrigation multilevel channels based on a model predictive control algorithm, and belongs to the field of automatic control of irrigation and water delivery and distribution and agricultural water resource management. Firstly, acquiring design data and operating condition data of an irrigation multi-stage channel, establishing a channel control model and converting the channel control model into a state space equation form; then, predicting the future output quantity of the irrigation channel water delivery and distribution system, and establishing a model predictive control algorithm target function and identifying constraint conditions according to the operating conditions; and finally, obtaining the optimal control quantity through optimization solution, and realizing safe and effective automatic control on the irrigation multi-stage channel. The model predictive control algorithm is designed based on the irrigation multi-stage channel design data and the operation conditions, the known water intake change and the constraint condition in the operation of the irrigation multi-stage channel can be effectively responded, the model predictive control algorithm can be used for the automatic control design of the irrigation multi-stage channel, the safe and reliable water supply service can be ensured, and the efficient management and utilization of water resources in irrigation areas can be effectively realized.
Description
Technical Field
The invention belongs to the field of irrigation, water delivery and distribution automatic control and agricultural water resource management, and particularly relates to an irrigation multi-stage channel automatic control method based on a model predictive control algorithm.
Background
In the current water resource utilization structure, the proportion of agricultural water is the largest, the agricultural water accounts for about 70% of the total water consumption in the world, however, the agricultural irrigation water consumption mode is rough, the management and scheduling are unscientific, and the control mode is lagged behind, so that a large amount of waste of water resources is caused, meanwhile, the efficient and reliable water supply service cannot be provided for water consumers, so that the water resource allocation is inefficient, the utilization rate is low, and about 20% -30% of water loss is caused by unreasonable scheduling operation in the irrigation channel water delivery process. The irrigation channel water delivery and distribution automatic control can ensure that the channel water delivery process is carried out safely and reliably through real-time monitoring and comprehensive scheduling management of channel hydraulic information, water is supplied according to needs, water loss and waste of a canal system are effectively reduced, the operation management and service level of the canal system are improved, and the irrigation channel water delivery and distribution automatic control method is a necessary trend of modern agricultural irrigation development.
The PID feedback control algorithm is the most classical, but the control effect of the PID algorithm based on single input and single output in a multi-cascade water delivery channel is not ideal enough, the optimization control algorithm of multiple input and multiple output needs to be developed, and a linear quadratic form (L QR) has more achievements as one of the optimization control algorithms.
1) Establishing an irrigation multi-stage channel control model; the method comprises the following specific steps:
1-1) determining a channel to be controlled, and collecting design data and operating condition data of the channel;
assuming that the multi-level channel to be controlled consists of f trench ponds, the channel design data to be collected includes the length L of each trench pondiLongitudinal slope SbiRoughness niDesign flow rate QiAnd section form data and the like, wherein the operation condition data to be collected comprises the water taking flow q of each channel pooliDesign running water depth h of control pointspiDesigned operating water level yspiSystem output reference value yriAnd safe operating range + -ri(riRepresenting the maximum value of allowable fluctuation of the water level operation process of the control point corresponding to the ith canal pond), and the like.
1-2) establishing a channel control model by using the data collected in the step 1-1); the expression is as follows:
in the formula, yiThe water level of a downstream control point corresponding to the ith canal pond is relative to yspiAmount of change (c), unit: m; t is time, unit: s; a. thesiFor the return water district area that ith canal pond corresponds, the unit: m is2;qini、qoutiAnd q isdiThe canal pond that corresponds for ith canal pond respectively is gone into flow, is gone out flow and is fetched water the flow and be corresponding to initial steady state's variation, unit: m is3/s;τiThe unit of the lag time corresponding to the ith channel pool is as follows: and s.
Obtaining the water return area A of each channel pool by theoretical formula calculation or numerical simulation identification method under the condition of designed flowsiAnd lag time τiTwo parameters are used for determining the hydraulic characteristics of each channel pool of the irrigation multi-stage channel and combining the water taking flow q of each channel pooliThe channel control model expression (1-1) can be obtained.
2) Converting the channel control model established in the step 1) into a state space equation form;
constructing a discrete state space equation of the multilevel channel according to the channel control model formula (1-1) established in the step 1), and assuming that the irrigation channel is a steady system, the discrete state space equation is shown in the formula (1-2):
x(k+1)=Ax(k)+Bu(k)+Dd(k) (1-2)
wherein k is time in discrete form; x is a state variable; u is a control variable; d is a disturbance variable; a is a system matrix; b is a control matrix; d is a disturbance matrix.
3) Constructing an objective function of a linear quadratic algorithm and solving an optimal control quantity;
3-1) setting the square sum of the state variable and the control variable in the formula (1-2) as an objective function formula (1-3) of the operation of the channel. In the process of controlling the operation of the channel, the target of controlling the water level to be finally stabilized at a set value is realized by properly adjusting and changing the flow of the control structure, and the expression of a target function of the operation of the channel is as follows:
in the formula, J is an objective function, Q is an n × n-dimensional state weighting matrix (n is a state variable dimension) which is a symmetrical positive definite (or semi-positive definite) matrix, and R is an f × f-dimensional control weighting matrix (f is a control variable dimension which is equal to the number of the channel pools) which is a symmetrical positive definite matrix.
The first term on the right of the equation of the objective function (1-3) is a function for measuring the dynamic deviation of the system, and the second term is used for measuring the control energy consumption. The linear quadratic optimal control can be understood as keeping a small output deviation through a small control quantity, so that the comprehensive optimization of the dynamic deviation and the energy consumption of a controlled system is achieved.
3-2) finding the optimal control variable u x k under the constraint of the formula (1-2) to minimize the formula (1-3). As can be known from the control theory, under the constraint of the state space equation, when the performance index is minimum, the optimal control variable satisfies the form of the formula (1-4):
u*(k)=-Kx(k) (1-4)
wherein K is (B)TPB+R)-1BTPA, referred to as optimal feedback matrix; p is the solution of the algebraic Riccati equation as shown in equations (1-5):
P=ATPA-ATPB(BTPB+R)-1BTPA+Q (1-5)
equations (1-5) can be solved by numerical methods, where K and P are solved in MAT L AB with the callable function dlqr based on the system matrix a, control matrix B, state weighting matrix Q, and control weighting matrix R.
As can be known from the linear quadratic optimal control quantity formula (1-4), the control variables are linear combinations of state variables and are determined by the optimal feedback matrix K, and the linear quadratic feedback control algorithm carries out real-time feedback control according to the current state of the system, so that the constraint conditions of water taking and operation in a plan cannot be effectively met.
The model predictive control algorithm (MPC) has feedback and feedforward functions at the same time, is a hotspot of research and application in the field of current process control, is designed based on a linear control model, predicts the future output of the system through the linear control model, and performs optimization control through an established finite time domain objective function, so that the MPC can effectively cope with determined or predictable external disturbance, and can consider the constraint conditions suffered by the system. However, the design and solution of the model predictive control algorithm under the constraint condition are not sufficient at present, and the effective application of the MPC in the actual engineering is limited.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an automatic control method for irrigation multi-stage channels based on a model predictive control algorithm. The invention can effectively deal with the known water intake change and constraint conditions in the channel operation, can be used for the design of an automatic control system of a multilevel irrigation channel, can realize safe and reliable water supply service, and effectively realizes the high-efficiency management and utilization of water resources in irrigation areas.
The invention provides an automatic control method for an irrigation multi-stage channel based on a model predictive control algorithm, which is characterized by comprising the following steps of:
1) establishing an irrigation multi-stage channel control model, which comprises the following specific steps:
1-1) determining a channel to be controlled, and collecting design data and operating condition data of the channel;
selecting the irrigation multi-stage channel system of any irrigation area as the channel to be controlled, and assuming that the channel system is composed of f canal pools, the channel design data comprises the length L of each canal pool of the f canal poolsiLongitudinal slope SbiRoughness niDesign flow rate QiAnd section form data, and operation condition data including water intake flow q of each channel pooliDesign running water depth h of control pointspiDesigned operating water level yspiOutput reference y of the systemriAnd safe operating range + -ri(ii) a Wherein r isiRepresenting the maximum value of allowable fluctuation of the water level operation process of a control point corresponding to the ith canal pond;
1-2) establishing a channel control model by using the data collected in the step 1-1); the expression is as follows:
in the formula, yiThe water level of the downstream control point corresponding to the ith canal pond is opposite to the designed operation water level yspiAmount of change (c), unit: m; t is time, unit: s; a. thesiFor the return water district area that ith canal pond corresponds, the unit: m is2;qini、qoutiAnd q isdiThe canal pond that corresponds for ith canal pond respectively is gone into flow, is gone out flow and is fetched water the flow and be corresponding to initial steady state's variation, unit: m is3/s;τiThe unit of the lag time corresponding to the ith channel pool is as follows: s;
2) converting the channel control model established in the step 1) into a state space equation form;
according to the formula (1), the discrete state space equation of the multilevel channel is constructed as shown in the formula (2) and the formula (3):
x(k+1)=Ax(k)+Bu(k)+Dd(k) (2)
y(k)=Cx(k) (3)
wherein k is time in discrete form; x is a state variable; u is a control variable; d is a disturbance variable; y is an output variable; a is a system matrix; b is a control matrix; c is an output matrix; d is a disturbance matrix;
3) predicting the future output quantity of the irrigation multi-stage channel system, and constructing a target function of a model prediction control algorithm; the method comprises the following specific steps:
3-1) determining a prediction time domain p and a control time domain c according to a system operation control requirement, and performing rolling prediction on a system state variable and an output variable by time intervals by using a formula (2) and a formula (3) to obtain an output result of the system at the tail end of the prediction time domain;
in the control time domain c, the predicted values of the state variables and the output variables of the channel system are respectively as follows:
when the control time domain is finished, predicting that the responses of the rest part in the time domain p are free responses;
the system output predicted value is arranged into the following matrix form expression:
Y(k+1|k)=Sxx(k)+SuU(k)+SdD(k) (19)
3-2) comparing the formula (18) obtained in step 3-1) with a reference amount yrCarrying out quadratic summation on the deviation and the control variable u to obtain a target function of the model predictive control algorithm;
in the formula (I), the compound is shown in the specification,predicting the output quantity, y, for the systemrIs a reference quantity of system output, u is a control variable, QjWeighted penalty matrix for jth water level deviation, RjA weighted penalty matrix for the jth control variable;
the output reference amount and the weighting matrix are expressed in the form of a matrix,
Yr(k+1|k)=[yr(k+1);yr(k+2);…;yr(k+p)](23)
Q=diag(Q1,Q2,...,Qp),R=diag(R1,R2,...,Rc) (24)
combining the matrix form of equations (23) and (24) for the system output prediction and control variables, objective function equation (22) is represented as a matrix simplified form as follows:
J=[Q(Y(k+1|k)-Yr(k+1))]2+[RU(k)]2(25)
4) based on the result of the step 3), identifying system constraint conditions, and obtaining optimal control quantity through optimization solution calculation; the method comprises the following specific steps:
4-1) identifying system constraints; the method comprises the following specific steps:
controlling structural flow amplitude constraints;
Qlb≤Q0(t-1)+ΦU(k)≤Qub(26)
where Φ is the control variable transformation matrix, Q0For controlling the current time flow of the structure, QubAnd QlbMaximum and minimum flow rates of the control structure, respectively;
controlling the structural flow variable amplitude constraint; the control structure flow amplitude variation constraint comprises a control variable maximum amplitude variation constraint and a control structure flow minimum amplitude variation constraint;
the expression form of the maximum amplitude constraint of the control variable is as follows:
Ulb≤U(k)≤Uub(27)
the expression form of the minimum variable amplitude constraint of the control structure flow is as follows:
|U(k)|≥Udb(28)
in the formula of UubAnd UlbThe maximum amount of flow variation, U, permitted by the control structure as it increases and decreases, respectivelydbA minimum flow variation required for adjustment of the control structure;
water level amplitude constraint;
Ylb≤Y(k+1|k)≤Yub(29)
in the formula, YubAnd YlbRespectively the maximum allowable water level of a control point in the operation process of irrigating the multistage channelA minimum value;
water level amplitude variation restriction;
ΔYlb≤RX(k+1|k)≤ΔYub(30)
where R is a sparse coefficient matrix, Δ YubAnd Δ YlbMaximum water level variation allowed when the water level of the control point rises and falls is respectively set;
4-2) according to the formula (25), combining the identification of the system operation constraint conditions in the step 4-1), converting the objective function of the model predictive control algorithm into a standard form of quadratic programming, and solving by utilizing a quadratic programming optimization algorithm to obtain an optimal control sequence in a prediction time domain; the specific method comprises the following steps:
the standard form of the quadratic programming problem objective function is organized as equation (25) as equation (31):
in the formula, HuThe sea-son matrix is a symmetric semi-positive definite matrix; grIs a gradient vector; e.g. of the type0Is a constant term; t is a transpose operator;
substituting formula (19) for formula (25) and finishing to obtain formula (32):
wherein the hessian matrix, the gradient vector and the constant term are respectively shown as formulas (33), (34) and (35):
Hu=(QSu)TQSu+RTR (33)
Gr(k+1)=-(QSu)TQ[Yr(k+1)-Sxx(k)-SdD(k)](34)
e0=0.5×{Q[Yr(k+1)-Sxx(k)-SdD(k)]}T{Q[Yr(k+1)-Sxx(k)-SdD(k)]} (35)
4-2-1) performing unconstrained optimization solution;
when the input and the output of the system are not influenced by the constraint condition, the quadratic programming problem has an analytic solution; and (3) obtaining an analytical expression (36) of an optimal solution by taking the derivative of the control variable U (k) of the formula (31) as 0:
obtaining an unconstrained optimal control sequence U in the control time domain by solving equation (36)uc(k) The control amount is the optimum control amount U for minimizing the expression (25) in the control time domainuc(k):
4-2-2) carrying out constrained optimization solution;
the quadratic programming expression with linear constraints is as follows:
s.t.U(k)≤b (39)
where b is a matrix and vector defining a linear constraint;
the standard form of the flow amplitude constraint arrangement of the control structure into quadratic programming linear constraint is as follows:
the standard form of the maximum amplitude-variation constraint of the control structure flow to be arranged into quadratic programming linear constraint is as follows:
wherein I is an identity matrix of dimension c × c;
the standard form of the water level amplitude constraint which is organized into quadratic programming linear constraint is as follows:
the standard form of the water level amplitude variation constraint which is organized into quadratic programming linear constraint is as follows:
in the formula, coefficient matrix Sux,Sxx,SdxObtained by dividing equation (21) by the output matrix C;
the model prediction control algorithm predicts the future output according to the established channel control model and outputs a predicted value and a reference quantity yrAnd the sum of the quadratic forms of the deviation and the control variable U is an objective function, under the constraint condition, the optimal control quantity U in the control time domain is obtained by identifying the system operation constraint condition, and the control quantity is the optimal control quantity which enables the objective function represented by the formula (22) to be minimum in the control time domain.
The invention has the characteristics and beneficial effects that:
the invention provides an automatic control method for irrigation multilevel channels based on a model predictive control algorithm, which is characterized in that a channel control model is constructed by utilizing channel design data and operating conditions, the future output of a channel is predicted through a state space equation, and constraint conditions in the operation of a channel system are considered in the solution of an objective function, so that the model predictive algorithm can effectively cope with the known water taking process and the operation constraint conditions, and the algorithm is more practical. The model predictive control algorithm can effectively solve the problems that a single feedback control algorithm cannot effectively cope with the hysteresis and the coupling characteristics and cannot consider the known water taking process and the operation constraint conditions in the operation of the multistage irrigation channel, provides an effective technical scheme for the automatic control of the water delivery and distribution of the irrigation channel and the refined agricultural water management, and has the advantages of strong operability and convenience in practical application and popularization.
Drawings
FIG. 1 is an overall flow diagram of the method of the present invention.
Fig. 2 is a schematic diagram of a water level deviation change process according to an embodiment of the present invention.
Detailed Description
The invention provides an automatic control method of an irrigation multi-stage channel based on a model predictive control algorithm, which is further described in detail below by combining the attached drawings and concrete implementation.
The invention provides an automatic control method of an irrigation multi-stage channel based on a model predictive control algorithm, the overall flow of the method is shown in figure 1, and the method comprises the following steps:
1) establishing an irrigation multi-stage channel control model, which comprises the following specific steps:
1-1) determining a channel to be controlled, and collecting design data and operating condition data of the channel;
in the invention, the channel to be controlled is an irrigation multi-stage channel system of any irrigation area, and for a channel with a trapezoidal section consisting of f channel ponds, the channel design data to be collected comprises the length L of each channel pond of the f channel pondsiLongitudinal slope SbiRoughness niDesign flow rate QiAnd section form data (for trapezoidal sections, the section form data comprises the bottom width b of each ditch pooliSlope coefficient SliDesign the depth h of the trenchi) The collected operation condition data comprises the water intake flow q of each ditch pooliDesign running water depth h of control pointspiDesigned operating water level yspiOutput reference y of the systemriAnd safe operating range + -ri(riRepresenting the maximum value of allowable fluctuation of the water level operation process of the control point corresponding to the ith canal pond), and the like.
One embodiment of the invention is a two-part main canal of the southwest Changmu Manan irrigation district in China, the canal system comprises 4 canal ponds, and the parameter value scheme of each canal pond is shown in table 1.
TABLE 1 channel design parameters of embodiments of the present invention
1-2) establishing a channel control model by using the data collected in the step 1-1);
the invention adopts an integral time lag model shown in formula (1) as a channel control model, the model takes a channel pool i between adjacent control structures as a control unit, and is a lumped parameter linear equation obtained after the saint-wien equation is linearized, and the expression is as follows:
in the formula, yiThe water level of the downstream control point corresponding to the ith canal pond is opposite to the designed operation water level yspiAmount of change (c), unit: m; t is time, unit: s; a. thesiFor the return water district area that ith canal pond corresponds, the unit: m is2;qini、qoutiAnd q isdiThe canal pond that corresponds for ith canal pond respectively is gone into flow, is gone out flow and is fetched water the flow and be corresponding to initial steady state's variation, unit: m is3/s;τiThe unit of the lag time corresponding to the ith channel pool is as follows: and s.
Obtaining the water return area A of each channel pool by theoretical formula calculation or numerical simulation identification method under the condition of designed flowsiAnd lag time τiTwo parameters are used to determine the hydraulic characteristics of each channel pool of the multistage channel, and the water intake flow q of each channel pool is combinediThe channel control model formula (1) can be obtained.
In this embodiment, the integral time lag model parameter calculation theoretical formula method: the two main channel channels of the Changmanan irrigation area are prismatic channels, the section forms are uniform, and a theoretical formula can be adopted for calculation.
The method for calculating the area of the water return area corresponding to each channel pool comprises the following steps of;
the horizontal assumption of a water surface line of the water return area of the integral time-delay model is adopted, the length of the water return area can be determined through a geometric relation, the geometric shape of the water surface of the water return area under the assumption is determined by a section form, and the area parameter of the water return area can be rapidly calculated by combining geometric shape characteristics and the length of the water return area. Assuming that the prismatic water delivery channel is composed of f canal ponds, the canal pond i is at the design flow rate QiThe normal water depth of the running uniform flow ishniThe water depth set value of the downstream control point of the ditch pool is hspi(hspi>hni) When the cross section of the channel is trapezoidal, the area of the water return area corresponding to the ith channel pool can be obtained by approximate calculation, and the calculation formula is as follows:
in the formula, biThe channel bottom width m is the channel bottom width of the trapezoid cross section corresponding to the ith channel pool; sliThe coefficient of the slope of the trapezoidal section corresponding to the ith channel pond. SbiIs a channel longitudinal slope corresponding to the ith channel pond,
the method for calculating the lag time corresponding to each channel pool comprises the following steps:
according to the assumption of an integral time-lag model, when the whole ditch pool is in a water return area, the lag time is 0; when the ditch pool part is in a backwater area, the lag time is the propagation time of disturbance waves at the upstream end in the uniform flow part, and an integral time lag model developer Schuurmans et al (1995) suggests that a motion wave model is adopted to simulate the water flow propagation process of an open channel water conveying process. In the motion wave model, the propagation velocity of the water flow is a function of the water depth, and the formula of the propagation velocity of the motion wave of the water flow is as follows:
wherein Q is the water delivery flow rate, unit: m is3/s;TiFor the corresponding surface of water width in ith canal pond, the unit: m; h is water depth, unit: and m is selected.
For the uniform flow water delivery part of the prismatic channel, the lag time can be calculated according to the length of the uniform flow area and the wave velocity of the motion wave, and the calculation formula of the lag time is as follows:
after a linearized Saint-Vietnam equation set is adopted to approximate the dQ/dh term, the calculation formula of the lag time is as follows:
in the formula, LuiThe length of a water return area corresponding to the ith canal pond is m; p is wet week, m; a. theiThe flow area, m, corresponding to the ith canal pit2;TiThe water surface width m corresponding to the ith canal pond; h is water depth m; v. ofiAverage flow velocity, m, for the ith channel3S; the subscript 0 indicates the value of the parameter at initial steady state.
Calculating the area A of a water return area of each channel pool of the multistage channel under the condition of design flow according to the step 1-2)siAnd lag time τiAnd (3) determining a water intake flow change scheme according to the channel design operation flow scheduling scheme by using the parameters as shown in the table 2, so as to obtain the channel control model as shown in the formula (1).
Table 2 case channel design flow integral model parameters
2) Converting the channel control model established in the step 1) into a state space equation form;
there is lag time in the channel control model, and control system control action takes place at certain time interval, according to equation (1), constructs discrete state space equation (2) and equation (3) of multistage channel, assumes that irrigation channel is the system of normality, and its discrete form state space equation can be expressed as:
x(k+1)=Ax(k)+Bu(k)+Dd(k) (2)
y(k)=Cx(k) (3)
wherein k is time in discrete form; x is a state variable; u is a control variable; d is a disturbance variable; y is an output variable; a is a system matrix; b is a control matrix; c is an output matrix; d is a disturbance matrix.
Taking the channel pool i as an example, if the control time step is selected to be Δ t, the discrete form of the integral time-lag model is as follows:
in the formula, Δ t is a control time step, unit: s; k is a radical ofτiThe delay time under the discrete time model corresponding to the ith channel pool is obtained by the ratio of the actual delay time to the control time step, and the rest parameters are consistent with the continuous form.
Defining the variation range delta y of water level deviation corresponding to the ith canal pondi(k)=yi(k)-yi(k-1), the controlled flow amplitude variation delta q corresponding to the ith channel pooli(k)=qi(k)-qi(k-1), and obtaining the following formula after finishing deformation of the formula (4):
deviation y of selected water leveli(k) And its amplitude Δ yi(k) And controlling the flow amplitude Deltaq in the early stageini(k-i)(i=1...kτ) Is a state variable, i.e. x (k) ═ yi(k),Δyi(k),Δqini(k-1),…,Δqini(k-kτi)]TSelecting the amplitude of the control flow in the current time interval as a control variable, namely u (k) ═ delta qini(k)]Selecting a future time interval to change the flow quantity of the water intake into a disturbance variable, namely d (k) ═ delta qdi(k)]Selecting the water level deviation as an output variable, namely y (k) [ -yi(k)]Therefore, the state space equation of the ditch pool i can be obtained.
When considering the channel system composed of all f channels, taking the state variable x (k) ═ x1(k);x2(k);…;xf(k)]Control variable u (k) is [ u ]1(k);u2(k);…;uf(k)]Disturbance variable d (k) ═ d1(k);d2(k);…;df(k)]Output variable y (k) ═ y1(k);y2(k);…;yf(k)]Can obtain the discrete shape of the multi-level channelEquation (2) and equation (3) for the state space of the equations, wherein the system matrix a, the control matrix B, the output matrix C and the disturbance matrix D are determined by the channel control model parameters of each trench pool.
According to the calculation method, the state variable x, the control variable u, the disturbance variable D and the output variable y of the case channel, the system matrix A, the control matrix B, the output matrix C and the disturbance matrix D are obtained as follows.
3) Predicting the future output quantity of the irrigation multi-stage channel system, and constructing a target function of a model prediction control algorithm; the method comprises the following specific steps:
3-1) determining a prediction time domain p and a control time domain c according to the system operation control requirement, and performing rolling prediction on the system state variable and the output variable by time intervals by using the formula (2) and the formula (3) to obtain an output result of the system at the end of the prediction time domain. The predicted values of the state variable and the output variable of the channel system at the time k +1 can be respectively expressed as follows:
the predicted value of the channel system state variable at the moment k +2 is as follows:
similarly, the predicted value of the channel system output variable at the time k +2 is as follows:
the prediction process is carried out until a control time domain c, and the predicted values of the state variables and the output variables of the channel system are respectively as follows:
the above steps are all prediction of system state variables and output variables in the control time domain c, and the responses of the rest part in the prediction time domain p are all free responses after the control time domain is finished.
The system output prediction value can be expressed in a matrix form by sorting as follows:
Y(k+1|k)=Sxx(k)+SuU(k)+SdD(k) (19)
3-2) comparing the formula (18) obtained in step 3-1) with a reference amount yrDeviation of (2)And performing quadratic summation on the control variable u to obtain a target function of the model predictive control algorithm.
Wherein J is an objective function, p is a prediction time domain, c is a control time domain,predicting output variables, y, for the systemrIs a reference quantity of system output, u is a control variable, QjWeighted penalty matrix for jth water level deviation, RjA weighted penalty matrix for the jth control variable.
The output reference amount and the weighting matrix are expressed in the form of a matrix,
Yr(k+1|k)=[yr(k+1);yr(k+2);…;yr(k+p)](23)
Q=diag(Q1,Q2,...,Qp),R=diag(R1,R2,...,Rc) (24)
and combining the matrix form of equations (23) and (24) for the system output prediction and control variables, objective function equation (22) can be expressed as a matrix simplified form as follows:
J=[Q(Y(k+1|k)-Yr(k+1))]2+[RU(k)]2(25)
4) recognizing system constraint conditions and proposing a processing scheme by using the formula (25) constructed in the step 3), and obtaining optimal control quantity through optimization, solution and calculation; the method comprises the following specific steps:
4-1) identifying system constraint conditions and providing a processing scheme according to the overflow characteristic of the control structure and the water level safety operation range of the control point. The main constraints to be considered in the operation of an irrigation delivery and distribution system include: the flow of the control structure is limited within an allowable range according to the overcurrent capacity of the control structure, so that the performability of control actions is ensured; the amplitude variation of the flow of the control structure is limited within a certain range so as to reduce consumption and avoid damage and the like; the water level is maintained between the set safety range limits to realize safe and efficient water delivery and prevent the overflow of the canal levee; according to the protection requirement of the channel, the water level change rate is limited within a certain range, and the damage of the lining of the channel caused by the rapid change of the water level in the water delivery operation process of the channel is prevented. The constraint conditions can be summarized into two types of constraints of control variables and output variables, and each type of constraint can be divided into two types of constraints of amplitude and allowable amplitude. The specific case constraints for control variables and output variables in the multi-stage irrigation water distribution system of the present invention can be classified as:
1. controlling variable amplitude constraint: maximum and minimum flow capacity of the control structure (gate);
2. controlling variable amplitude constraint: maximum and minimum flow amplitude constraints allowed by the control structure;
3. output variable amplitude constraint: controlling the safe operating range of the point water level;
4. output variable amplitude constraint: and controlling the safe amplitude variation constraint of the water level of the point.
The above constraint may be specifically expressed in the following form:
① control structural flow magnitude constraints;
the control structure flow amplitude constraint in the multi-level irrigation channel system of the present invention is the maximum and minimum flow capacity of the control structure. In the control time domain, the expression of the flow amplitude constraint of the control structure is as follows:
Qlb≤Q0(t-1)+ΦU(k)≤Qub(26)
where Φ is the control variable transformation matrix, Q0For controlling the current time flow of the structure, QubAnd QlbRespectively the maximum and minimum overflowness of the control structure.
② controlling structural flow amplitude variation constraint;
the control structure flow variation constraint is the maximum and minimum flow variation allowed by the control structure in the irrigation system of the present invention. In the control time domain, the expression form of the maximum amplitude constraint of the control variable is as follows:
Ulb≤U(k)≤Uub(27)
in the control time domain, the expression form of the minimum variable amplitude constraint of the flow of the control structure is as follows:
|U(k)|≥Udb(28)
in the formula of UubAnd UlbThe maximum amount of flow variation, U, permitted by the control structure as it increases and decreases, respectivelydbThe minimum amount of flow change required to adjust the control structure.
③ water level amplitude constraints;
the water level amplitude is constrained in the multi-level irrigation channel system of the invention to be the maximum value and the minimum value allowed by the water level of the control point in the operation process. In the prediction time domain, the expression of the water level amplitude constraint is as follows:
Ylb≤Y(k+1|k)≤Yub(29)
in the formula, YubAnd YlbRespectively the maximum value and the minimum value allowed by the water level of a control point in the operation process of the multi-stage irrigation channel.
④ water level amplitude variation restriction;
similar to the flow variable amplitude constraint of the control structure, the water level also has variable amplitude constraint, the difference is that the water level only has maximum variable amplitude constraint, and in the prediction time domain, the expression of the maximum variable amplitude constraint of the water level is as follows:
ΔYlb≤RX(k+1|k)≤ΔYub(30)
in the formula, R is a sparse coefficient matrix, and the full-state variable is converted into a vector containing only water level deviation change through the action of R, wherein delta Y isubAnd Δ YlbThe maximum water level variation allowed when the water level at the control point rises and falls respectively.
The type, nature and processing method of the different constraints are shown in the following table.
TABLE 3 multistage irrigation channel model prediction algorithm constraint type and processing method
4-2) according to the formula (25) constructed in the step 3), in combination with the identification of the system operation constraint conditions in the step 4-1), the objective function of the model predictive control algorithm can be converted into a standard form of quadratic programming, and the optimal control sequence in the prediction time domain is obtained by utilizing the quadratic programming optimization algorithm to solve. The specific method comprises the following steps:
and (5) sorting the equation (25) into a quadratic programming problem and solving by adopting a correlation algorithm. The quadratic programming is a nonlinear programming problem with an objective function being a quadratic function and constraint conditions being linear forms, and the standard form of the quadratic programming problem objective function is as shown in formula (31):
in the formula, HuThe sea-son matrix is a symmetric semi-positive definite matrix; grIs a gradient vector; e.g. of the type0Is a constant term; t is the transpose operator.
In order to sort the objective function into a standard form of a quadratic programming problem objective function, a model prediction output result (19) is substituted into an equation (25) and is sorted to obtain an equation (32):
the objective function equation (32) is a typical quadratic programming problem, in which the hessian matrix, gradient vectors and constant terms are shown as equations (33), (34) and (35), respectively:
Hu=(QSu)TQSu+RTR (33)
Gr(k+1)=-(QSu)TQ[Yr(k+1)-Sxx(k)-SdD(k)](34)
e0=0.5×{Q[Yr(k+1)-Sxx(k)-SdD(k)]}T{Q[Yr(k+1)-Sxx(k)-SdD(k)]}(35)
4-2-1) unconstrained optimization solution
When the input and the output of the system are not limited by the constraint condition, the quadratic programming problem has an analytic solution. In order to obtain the analytical expression formula of the optimal solution, the control variable U (k) is derived by the formula (31), and the derivative of the control variable U (k) is 0 to obtain the analytical expression (36) of the optimal solution:
obtaining an unconstrained optimal control sequence U in the control time domain by solving equation (36)uc(k) The control amount is an optimum control amount for minimizing the expression (25) in the control time domain, and in practical application, the control structure is based on the optimum control amount U under the unconstrained conditionuc(k) Executing a control action:
4-2-2) constrained optimization solution
In an actual process control system, system input and system output are generally restricted by various external conditions, so that the design and solution of the model predictive control algorithm under the restriction conditions are more practical and have application value. The objective function of the model predictive control algorithm adopts a finite time domain form, so that the relevant constraints applied to the input and the output of the controlled system can be directly added in the solving of the objective function, the system constraints can be considered, and the optimization solving is carried out under the condition that the constraints exist, which is also one of the most important advantages of the model predictive control algorithm.
The constraint types of system input and output can be classified into linear and nonlinear. When the constraint is linear, the control problem is a typical quadratic programming problem, and the solution and the analysis can be carried out through a quadratic programming related algorithm; when the constraint is nonlinear, a correlation solving algorithm of nonlinear programming is needed to solve. Aiming at the situation that the constraint is linear constraint, the quadratic programming containing the linear constraint can be expressed as follows:
s.t.U(k)≤b (39)
where b is a matrix and vector defining a linear constraint.
In addition to the dead zone constraint, the remaining constraints can be expressed in the form of a linear constraint shown in equation (39).
The standard form of the flow amplitude constraint arrangement of the control structure into quadratic programming linear constraint is as follows:
the standard form of the maximum amplitude-variation constraint of the control structure flow to be arranged into quadratic programming linear constraint is as follows:
in the formula, I is an identity matrix of dimension c × c.
The standard form of the water level amplitude constraint which is organized into quadratic programming linear constraint is as follows:
① the standard form of the water level amplitude variation constraint which is organized as quadratic programming linear constraint is:
in the formula: coefficient matrix Sux,Sxx,SdxObtained by dividing equation (21) by the output matrix C.
The model prediction control algorithm predicts the future output according to the established channel control model and outputs a predicted value and a reference quantity yrThe sum of quadratic forms of the deviation and the control variable u is an objective function, under the constraint condition, the system operation constraint condition is identified, and the MAT L AB active set optimization algorithm is utilized to solve and obtain the optimal control quantity U (k) in the control time domain, wherein the control quantity is the optimal control quantity of each control structure in the control time domain, which enables the objective function represented by the formula (31) to be minimum, and the optimal control quantity is poured into the control structure in practical applicationAnd the control structure in the irrigation multi-stage channel executes corresponding control action according to the optimal control quantity U (k).
The channel operation process is simulated based on the established model predictive control algorithm, the result is shown in figure 2, the abscissa in the figure is simulation time, the ordinate is control point water level deviation, the upper straight dotted line and the lower straight dotted line represent the control point water level change allowable range, and the curve in the figure is the control point water level change process. The simulation result shows that before water intake changes, the model predictive control algorithm can be adjusted and controlled in advance, and after water intake disturbance, the model predictive control algorithm is controlled and adjusted, so that the water level of a channel control point operates in an allowable range, and finally the water level of the channel control point is recovered to be close to a water level operation set value. The results show that the model predictive control algorithm has the capability of coping with known water intaking disturbance and processing constraint conditions, can effectively control the two-section main channel of the Changmanan irrigation district, and can be popularized and applied to the automatic control of irrigation multilevel channels of the irrigation district.
Claims (1)
1. An automatic control method for irrigation multi-stage channels based on a model predictive control algorithm is characterized by comprising the following steps:
1) establishing an irrigation multi-stage channel control model, which comprises the following specific steps:
1-1) determining a channel to be controlled, and collecting design data and operating condition data of the channel;
selecting the irrigation multi-stage channel system of any irrigation area as the channel to be controlled, and assuming that the channel system is composed of f canal pools, the channel design data comprises the length L of each canal pool of the f canal poolsiLongitudinal slope SbiRoughness niDesign flow rate QiAnd section form data, and operation condition data including water intake flow q of each channel pooliDesign running water depth h of control pointspiDesigned operating water level yspiOutput reference y of the systemriAnd safe operating range + -ri(ii) a Wherein r isiRepresenting the maximum value of allowable fluctuation of the water level operation process of a control point corresponding to the ith canal pond;
1-2) establishing a channel control model by using the data collected in the step 1-1); the expression is as follows:
in the formula, yiThe water level of the downstream control point corresponding to the ith canal pond is opposite to the designed operation water level yspiAmount of change (c), unit: m; t is time, unit: s; a. thesiFor the return water district area that ith canal pond corresponds, the unit: m is2;qini、qoutiAnd q isdiThe canal pond that corresponds for ith canal pond respectively is gone into flow, is gone out flow and is fetched water the flow and be corresponding to initial steady state's variation, unit: m is3/s;τiThe unit of the lag time corresponding to the ith channel pool is as follows: s;
2) converting the channel control model established in the step 1) into a state space equation form;
according to the formula (1), the discrete state space equation of the multilevel channel is constructed as shown in the formula (2) and the formula (3):
x(k+1)=Ax(k)+Bu(k)+Dd(k) (2)
y(k)=Cx(k)(3)
wherein k is time in discrete form; x is a state variable; u is a control variable; d is a disturbance variable; y is an output variable; a is a system matrix; b is a control matrix; c is an output matrix; d is a disturbance matrix;
3) predicting the future output quantity of the irrigation multi-stage channel system, and constructing a target function of a model prediction control algorithm; the method comprises the following specific steps:
3-1) determining a prediction time domain p and a control time domain c according to a system operation control requirement, and performing rolling prediction on a system state variable and an output variable by time intervals by using a formula (2) and a formula (3) to obtain an output result of the system at the tail end of the prediction time domain;
in the control time domain c, the predicted values of the state variables and the output variables of the channel system are respectively as follows:
when the control time domain is finished, predicting that the responses of the rest part in the time domain p are free responses;
the system output predicted value is arranged into the following matrix form expression:
Y(k+1|k)=Sxx(k)+SuU(k)+SdD(k) (19)
3-2) comparing the formula (18) obtained in step 3-1) with a reference amount yrCarrying out quadratic summation on the deviation and the control variable u to obtain a target function of the model predictive control algorithm;
in the formula (I), the compound is shown in the specification,predicting the output quantity, y, for the systemrIs a reference quantity of system output, u is a control variable, QjWeighted penalty matrix for jth water level deviation, RjA weighted penalty matrix for the jth control variable;
the output reference amount and the weighting matrix are expressed in the form of a matrix,
Yr(k+1|k)=[yr(k+1);yr(k+2);…;yr(k+p)](23)
Q=diag(Q1,Q2,...,Qp),R=diag(R1,R2,...,Rc) (24)
combining the matrix form of equations (23) and (24) for the system output prediction and control variables, objective function equation (22) is represented as a matrix simplified form as follows:
J=[Q(Y(k+1|k)-Yr(k+1))]2+[RU(k)]2(25)
4) based on the result of the step 3), identifying system constraint conditions, and obtaining optimal control quantity through optimization solution calculation; the method comprises the following specific steps:
4-1) identifying system constraints; the method comprises the following specific steps:
controlling structural flow amplitude constraints;
Qlb≤Q0(t-1)+ΦU(k)≤Qub(26)
where Φ is the control variable transformation matrix, Q0For controlling the current time flow of the structure, QubAnd QlbMaximum and minimum flow rates of the control structure, respectively;
controlling the structural flow variable amplitude constraint; the control structure flow amplitude variation constraint comprises a control variable maximum amplitude variation constraint and a control structure flow minimum amplitude variation constraint;
the expression form of the maximum amplitude constraint of the control variable is as follows:
Ulb≤U(k)≤Uub(27)
the expression form of the minimum variable amplitude constraint of the control structure flow is as follows:
|U(k)|≥Udb(28)
in the formula of UubAnd UlbThe maximum amount of flow variation, U, permitted by the control structure as it increases and decreases, respectivelydbA minimum flow variation required for adjustment of the control structure;
water level amplitude constraint;
Ylb≤Y(k+1|k)≤Yub(29)
in the formula, YubAnd YlbAre respectively asMaximum and minimum values allowed by the water level of a control point in the operation process of the irrigation multi-stage channel;
water level amplitude variation restriction;
ΔYlb≤RX(k+1|k)≤ΔYub(30)
where R is a sparse coefficient matrix, Δ YubAnd Δ YlbMaximum water level variation allowed when the water level of the control point rises and falls is respectively set;
4-2) according to the formula (25), combining the identification of the system operation constraint conditions in the step 4-1), converting the objective function of the model predictive control algorithm into a standard form of quadratic programming, and solving by utilizing a quadratic programming optimization algorithm to obtain an optimal control sequence in a prediction time domain; the specific method comprises the following steps:
the standard form of the quadratic programming problem objective function is organized as equation (25) as equation (31):
in the formula, HuThe sea-son matrix is a symmetric semi-positive definite matrix; grIs a gradient vector; e.g. of the type0Is a constant term; t is a transpose operator;
substituting formula (19) for formula (25) and finishing to obtain formula (32):
wherein the hessian matrix, the gradient vector and the constant term are respectively shown as formulas (33), (34) and (35):
Hu=(QSu)TQSu+RTR (33)
Gr(k+1)=-(QSu)TQ[Yr(k+1)-Sxx(k)-SdD(k)](34)
e0=0.5×{Q[Yr(k+1)-Sxx(k)-SdD(k)]}T{Q[Yr(k+1)-Sxx(k)-SdD(k)]} (35)
4-2-1) performing unconstrained optimization solution;
when the input and the output of the system are not influenced by the constraint condition, the quadratic programming problem has an analytic solution; and (3) obtaining an analytical expression (36) of an optimal solution by taking the derivative of the control variable U (k) of the formula (31) as 0:
obtaining an unconstrained optimal control sequence U in the control time domain by solving equation (36)uc(k) The control amount is the optimum control amount U for minimizing the expression (25) in the control time domainuc(k):
4-2-2) carrying out constrained optimization solution;
the quadratic programming expression with linear constraints is as follows:
s.t. U(k)≤b (39)
where b is a matrix and vector defining a linear constraint;
the standard form of the flow amplitude constraint arrangement of the control structure into quadratic programming linear constraint is as follows:
the standard form of the maximum amplitude-variation constraint of the control structure flow to be arranged into quadratic programming linear constraint is as follows:
wherein I is an identity matrix of dimension c × c;
the standard form of the water level amplitude constraint which is organized into quadratic programming linear constraint is as follows:
the standard form of the water level amplitude variation constraint which is organized into quadratic programming linear constraint is as follows:
in the formula, coefficient matrix Sux,Sxx,SdxObtained by dividing equation (21) by the output matrix C;
the model prediction control algorithm predicts the future output according to the established channel control model and outputs a predicted value and a reference quantity yrAnd the sum of the quadratic forms of the deviation and the control variable U is an objective function, under the constraint condition, the optimal control quantity U in the control time domain is obtained by identifying the system operation constraint condition, and the control quantity is the optimal control quantity which enables the objective function represented by the formula (22) to be minimum in the control time domain.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811250103.8A CN109197539B (en) | 2018-10-25 | 2018-10-25 | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811250103.8A CN109197539B (en) | 2018-10-25 | 2018-10-25 | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109197539A CN109197539A (en) | 2019-01-15 |
CN109197539B true CN109197539B (en) | 2020-07-31 |
Family
ID=64997000
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811250103.8A Active CN109197539B (en) | 2018-10-25 | 2018-10-25 | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109197539B (en) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109765942B (en) * | 2019-03-04 | 2020-05-08 | 中南大学 | Open channel water level control method based on input-output feedback and generalized predictive control |
CN110689198A (en) * | 2019-09-27 | 2020-01-14 | 中南大学 | Open channel system prediction control method based on layered prediction iteration |
US12037262B2 (en) * | 2020-01-29 | 2024-07-16 | Xylem Vue Inc. | Systems and methods relating to effective management of fluid infrastructure |
CN111474965B (en) * | 2020-04-02 | 2021-10-26 | 中国水利水电科学研究院 | Fuzzy neural network-based method for predicting and controlling water level of series water delivery channel |
CN112434424B (en) * | 2020-11-24 | 2022-07-05 | 中国矿业大学 | Method for constructing irrigation mode of yam in sandy land |
CN113093531B (en) * | 2021-03-01 | 2022-03-15 | 武汉大学 | Large pipe-channel combined system emergency dispatching control method based on model predictive control |
CN114757464B (en) * | 2021-12-31 | 2023-09-26 | 水利部南京水利水文自动化研究所 | Water level flow fitting method under influence of irrigation demand |
CN114417631B (en) * | 2022-01-26 | 2023-10-03 | 清华大学 | Modeling method of infusion area water delivery and distribution system based on observation data |
CN114415520B (en) * | 2022-01-26 | 2023-08-25 | 清华大学 | Full-canal system self-adaptive predictive control method for water networking irrigation areas |
CN115456422B (en) * | 2022-09-16 | 2024-02-02 | 中国水利水电科学研究院 | Dynamic previewing correction method for irrigation area water distribution plan based on computational hydrodynamics |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8311786B2 (en) * | 2008-10-02 | 2012-11-13 | Geza Kisch | Accurate methods for modeling the spatial distribution for irrigation systems for landscapes |
CN102890732A (en) * | 2012-09-10 | 2013-01-23 | 清华大学 | Hydrodynamic condition parallel numerical simulation method for channel with various structures |
CN103651064A (en) * | 2013-11-12 | 2014-03-26 | 浙江工业大学 | Large-scale irrigation system control method based on distributed model prediction control |
CN103886519A (en) * | 2014-04-10 | 2014-06-25 | 北京师范大学 | Economic water consumption data spatial discretization method based on RBF neural network |
CN104714564A (en) * | 2014-12-17 | 2015-06-17 | 浙江工业大学 | Distributed liquid level system control method based on extendstate observer |
CN104833347A (en) * | 2015-05-22 | 2015-08-12 | 清华大学 | Irrigated area canal system surveying method and system based on GPS (Global Position System) positioning |
CN106223394A (en) * | 2016-08-12 | 2016-12-14 | 扬州大学 | Single pumping plant list water reservoir system water resource optimal allocation method in storehouse is directly mended under the conditions of insufficient irrigation |
KR20170005553A (en) * | 2015-07-06 | 2017-01-16 | 주식회사 유일기연 | Floods, drought assessment and forecasting techniques development for intelligent service |
CN107194607A (en) * | 2017-06-09 | 2017-09-22 | 清华大学 | A kind of restoration of the ecosystem key object decision method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7996192B2 (en) * | 2005-05-28 | 2011-08-09 | Dblive Corporation | Method and apparatus for generating an environmental element prediction for a point of interest |
-
2018
- 2018-10-25 CN CN201811250103.8A patent/CN109197539B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8311786B2 (en) * | 2008-10-02 | 2012-11-13 | Geza Kisch | Accurate methods for modeling the spatial distribution for irrigation systems for landscapes |
CN102890732A (en) * | 2012-09-10 | 2013-01-23 | 清华大学 | Hydrodynamic condition parallel numerical simulation method for channel with various structures |
CN103651064A (en) * | 2013-11-12 | 2014-03-26 | 浙江工业大学 | Large-scale irrigation system control method based on distributed model prediction control |
CN103886519A (en) * | 2014-04-10 | 2014-06-25 | 北京师范大学 | Economic water consumption data spatial discretization method based on RBF neural network |
CN104714564A (en) * | 2014-12-17 | 2015-06-17 | 浙江工业大学 | Distributed liquid level system control method based on extendstate observer |
CN104833347A (en) * | 2015-05-22 | 2015-08-12 | 清华大学 | Irrigated area canal system surveying method and system based on GPS (Global Position System) positioning |
KR20170005553A (en) * | 2015-07-06 | 2017-01-16 | 주식회사 유일기연 | Floods, drought assessment and forecasting techniques development for intelligent service |
CN106223394A (en) * | 2016-08-12 | 2016-12-14 | 扬州大学 | Single pumping plant list water reservoir system water resource optimal allocation method in storehouse is directly mended under the conditions of insufficient irrigation |
CN107194607A (en) * | 2017-06-09 | 2017-09-22 | 清华大学 | A kind of restoration of the ecosystem key object decision method |
Also Published As
Publication number | Publication date |
---|---|
CN109197539A (en) | 2019-01-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109197539B (en) | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels | |
CN103591637B (en) | A kind of central heating secondary network runing adjustment method | |
CN114415520B (en) | Full-canal system self-adaptive predictive control method for water networking irrigation areas | |
CN105305468B (en) | Thermal power generation unit primary frequency modulation parameter optimization method based on particle cluster algorithm | |
CN114417631B (en) | Modeling method of infusion area water delivery and distribution system based on observation data | |
CN112598195A (en) | Building type comprehensive energy system operation optimization method and device and terminal equipment | |
CN105573123B (en) | A kind of boiler coordination control method for thermal power unit based on the modeling of improved T-S fuzzy predictions | |
CN115167308A (en) | Water transfer engineering multi-target prediction control algorithm for guiding gate regulation and control | |
CN115456422B (en) | Dynamic previewing correction method for irrigation area water distribution plan based on computational hydrodynamics | |
CN110210114A (en) | Highest temperature prediction algorithm for roller compacted concrete gravity dam construction period | |
CN112733367A (en) | Water transfer project ice period water delivery scheduling method based on seven-day weather forecast | |
Kumari et al. | Dual degree branched type-2 fuzzy controller optimized with a hybrid algorithm for frequency regulation in a triple-area power system integrated with renewable sources | |
CN117314062B (en) | Multi-stage gate combined multi-target optimized water distribution scheduling method | |
CN113093531B (en) | Large pipe-channel combined system emergency dispatching control method based on model predictive control | |
CN103259258A (en) | Micro-grid, micro-grid control method and control device | |
Sun et al. | The application of new adaptive PSO in AGC and AFC combination control system | |
CN109888841A (en) | A kind of generating set load optimal distribution method | |
CN114268102B (en) | Power system running state quantization method based on analytic probability power flow model | |
Sambariya et al. | Application of FOPID Controller for LFC using elephant herding optimization technique | |
Moosavian et al. | Sequential quadratic programming and analytic hierarchy process for nonlinear multiobjective optimization of a hydropower network | |
CN113641101A (en) | Multi-channel pool control parameter optimization algorithm based on numerical simulation | |
Emami et al. | Presentation of a method based on gray wolf optimizer and imperialist competitive algorithms in optimal operation of dam reservoir | |
Wang et al. | Iterative adp optimal control for ggbs production based on dynamic target optimization | |
King et al. | Using a fuzzy inference system to control a pumped storage hydro plant | |
CN112180731B (en) | Energy equipment operation control method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
OL01 | Intention to license declared | ||
OL01 | Intention to license declared |