CN113093531B - Large pipe-channel combined system emergency dispatching control method based on model predictive control - Google Patents
Large pipe-channel combined system emergency dispatching control method based on model predictive control Download PDFInfo
- Publication number
- CN113093531B CN113093531B CN202110226122.2A CN202110226122A CN113093531B CN 113093531 B CN113093531 B CN 113093531B CN 202110226122 A CN202110226122 A CN 202110226122A CN 113093531 B CN113093531 B CN 113093531B
- Authority
- CN
- China
- Prior art keywords
- gate
- pressure
- inverted siphon
- model
- pipeline
- 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
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Flow Control (AREA)
Abstract
The invention discloses a large pipe-channel combined system emergency dispatching control method based on model predictive control, when an accident occurs, an open channel check gate is controlled according to a traditional emergency dispatching plan, a plurality of groups of opening increment schemes of long inverted siphon/pipeline outlet gates/valves are set for the long inverted siphon/pipeline outlet gates/valves, an open channel integral-time lag model and a large inverted siphon pressure linear model are taken as prediction models, hydraulic response in a certain prediction time domain in the future is predicted at intervals of a certain control interval time, the safety limiting condition of the long inverted siphon/pipeline is considered, and the optimization control of pipe internal pressure fluctuation and the outlet gates/valves is considered by utilizing an objective function of the model predictive control. The invention fully considers the stability of the outlet gate/valve for controlling the pressure fluctuation in the pipe, reduces the action frequency of the gate/valve as much as possible, reduces the accident loss of the gate/valve, and improves the rapidity, the safety and the stability of emergency treatment response.
Description
Technical Field
The invention belongs to the field of water conservancy dispatching, and designs a water conservancy dispatching emergency control method, in particular to a model prediction control-based large pipe-channel combined system emergency dispatching control method.
Background
In the prior art, hydraulic engineering is designed by considering the form of open channel non-pressure flow water transmission, but with the deep development of water resources, more and more large pipe-channel combined water transmission engineering appears, namely pressure flow sections and non-pressure flow sections alternately appear along the line, for example, north-jaw water resource allocation engineering contains a section of ultra-long inverted siphon with the length of 72km, and the pressure flow section of the secondary water diversion engineering in Yunnan accounts for 68% of the total length of the line. Compared with open channel water delivery engineering, the operation scheduling of the pipe-channel combined system is more complicated.
Because the response speed of the open channel system mainly based on gravity waves and the response speed of the pipeline system mainly based on elastic waves have great difference, the water flow conditions are very complex, the operation regulation and the control of the open channel system and the pipeline system are difficult to match, and the real-time regulation and the control of the whole water quantity are more difficult due to the uncertainty of links such as flow control, information acquisition and the like of the water delivery and distribution system. Particularly under the accident condition, if the pressure flow section has secondary accidents (such as pipe explosion caused by pressure overrun or cavitation caused by negative pressure) due to unreasonable or wrong emergency scheduling, the influence and loss are more difficult to bear than the accidents on the open channel system.
How to deal with the accident when the large-scale water transfer project has the accident is an important subject for ensuring the project safety, and the current common means is to formulate an emergency dispatching plan, namely, to assume various possible accident conditions in advance, to formulate an optimal action scheme for controlling buildings (such as a gate, a water diversion port, a water outlet gate and the like) along the line through numerical simulation and to formulate a plan library, and to carry out emergency dispatching management according to the emergency plan and the experience of a manager when the accident occurs. This treatment method has two major disadvantages: firstly, it is time-consuming and labor-consuming to assume various accident conditions in advance and compare and select a scheduling plan through numerical simulation; secondly, the method is more suitable for emergency dispatching of open channel control buildings (such as a check gate). In a large-scale pipe duct combination system, an outlet gate/valve of a long inverted siphon or pressure pipeline is positioned at the junction of a pressure flow section and a non-pressure flow section, and the hydraulic response laws at two sides of the gate/valve are completely different due to the huge difference of the wave velocity of pressure flow and non-pressure flow. On one hand, strong nonlinearity and coupling exist in the overflowing of the outlet gate/valve; on the other hand, if the inverted siphon or pressure pipe section is long (e.g. northwest water resource allocation project), the water hammer phenomenon will cause the pressure flow on the upstream side and the non-pressure flow on the downstream side of the outlet gate/valve to fluctuate and aggravate simultaneously. Thus a large variation in hydraulic response may be caused by a long inverted siphon or a small deviation in the movement of the inlet/outlet gate/valve of the pressure conduit. The unpredictability and randomness of accidents make the emergency scheduling plan of the pipe-channel combined system difficult to exert expected effects, and no mature emergency scheduling experience exists at home and abroad for the projects, so that the risk required to be borne by the experience judgment of managers is large. In this context, it may be more suitable to introduce advanced automation control techniques to the long inverted siphon or pressure pipe outlet gates/valves.
In summary, one of the technical problems that needs to be urgently solved by those skilled in the art is: the emergency dispatching control mode of the large-scale pipe-channel combined system is provided, and the requirements of the following aspects can be met by using an advanced automatic control technology: firstly, the outlet gate/valve of the long inverted siphon/pressure pipeline can be reasonably controlled, the pressure limit in the long inverted siphon/pressure pipeline is considered, and secondary accidents such as pipe explosion due to pressure overrun or cavitation due to over-low pressure are prevented; and secondly, the optimization control can be realized on the basis of ensuring the engineering safety, so that the pressure fluctuation in the pipe and the opening and closing frequency of an outlet gate/valve are as small as possible, the accident loss is reduced, and the safety and the stability of the canal system control are improved. Based on the model prediction control, the invention provides a model prediction control-based emergency scheduling control mode of a large-scale pipe duct combined system.
Disclosure of Invention
The invention aims to solve the technical problem of providing a model predictive control-based emergency scheduling control mode for a large pipe-channel combined system aiming at the defects in the prior art. By providing a specific modeling step and an automatic control method, safety limiting conditions and control optimization conditions which need to be considered in emergency scheduling are considered, so that when an accident happens at a certain position of the pipe-channel combined system, the long inverted siphon/pipeline outlet gate/valve can timely and reasonably act, secondary accidents such as pipe explosion due to over-pressure or cavitation due to over-low pressure are prevented, and the system can meet the requirements of stability, safety, robustness and the like in the emergency scheduling of the pipe-channel combined system.
The purpose of the invention is realized as follows: when an accident occurs at a certain position of the pipe-channel combined system, the open channel check gate is controlled according to a traditional emergency scheduling plan, for the long inverted siphon/pipeline outlet gate/valve, a plurality of groups of opening increment schemes of the long inverted siphon/pipeline outlet gate/valve are set based on a model prediction technology, an open channel integral-time lag model (ID model) and a large inverted siphon pressure linear model are used as prediction models, and T is a control interval time at regular intervalsintervalPredicting time domain T for certain futurepredictPredicting the hydraulic response in the pipeline, considering the safety limit condition of the long inverted siphon/pipeline, and considering the pressure fluctuation in the pipeline and the optimal control of an outlet gate/valve by using an objective function of model prediction control.
The technical scheme adopted by the invention is as follows:
a large pipe-channel combined system emergency dispatching control method based on model predictive control is characterized by comprising the following steps:
Pmin≤P≤Pmax
in the formula:
Pmin: the lower pressure limit of the outlet section of the long inverted siphon/pressure pipeline is expressed in m;
Pmax: representing the upper pressure limit of the outlet section of the long inverted siphon/pressure pipeline, unit m;
p: represents the instantaneous pressure of the outlet section of the long inverted siphon/pressure pipeline, in m;
step 3, constructing a linear prediction model of the pipe-channel combination system by using an open channel integral-time lag model (ID model) and a large-scale inverted siphon pressure linear model, and rapidly predicting the opening increment scheme of each group of outlet gates/valves in a set prediction time domain TpredictInternal hydraulic response processes;
step 4, excluding the pipeline pressure safety constraint condition in the step 2 from the prediction time domain TpredictAn unreasonable scheme that the pressure is over-limited and the tube is burst or the pressure is too low and the cavitation is corroded can be caused in the process; the optimal outlet gate/valve increment scheme is selected by considering the optimal control on the pressure fluctuation and the gate opening and closing frequency through an objective function J;
step 5, selecting the optimal long inverted siphon/pipeline outlet gate/valve according to the model predictive control algorithm in step 4The outlet gate/valve increment scheme acts, and the open channel gate is opened and closed according to the existing emergency scheduling plan in the step 1; when the long inverted siphon/pipeline outlet gate/valve is opened and closed at the maximum opening and closing speed VmaxAfter finishing the action, keeping the opening unchanged until the next prediction control is started;
step 6, controlling the interval time TintervalThen repeating the step 3 to the step 5;
step 7, when the long inverted siphon/pipeline outlet section is in a safe time TsafeThe pressure fluctuation in the device is always maintained in a safe range, and the long inverted siphon/pipeline outlet gate/valve is at a target opening value etarget0Closing the model predictive control module;
step 8, through a gate opening reverse calculation module, the minimum opening and closing speed V of the gate is usedminAnd controlling the long inverted siphon/pipeline outlet gate/valve until the target stable flow is reached.
Further, the prediction time domain TpredictIs 5-30 min; the control interval time TintervalIs 5-30 min.
Further, the safety time TsafeIs 0.5-2 h.
Further, in step 3, the Schuurmans et al proposed that an open channel integral-time-lag model in 1995 generalizes the channel pool into a uniform flow area and a return water area, wherein the water depth of the uniform flow area is normal, and it is considered that the existence of the uniform flow area will cause the downstream water level response to lag the upstream flow change, and the channel lag time completely in the return water area is 0, the model simulates the return water area with an integral link, and the uniform flow area with a lag link has an expression:
in the above formula:
Asis the water surface area of the backwater area in unit of m2；
e is the downstream water level deviation in m;
x is a position of a water return area;
qinthe variation of the upstream inflow flow of the channel pond in m3/s；
qoutIs the variation of the downstream outflow of the channel pond in m3/s；
t is time, unit s;
tau is lag time, namely the time for transmitting the upstream flow change to the downstream water return area through the uniform flow area, and the unit is s;
the large-scale inverted siphon pressure linear model relates the pressure change at the downstream end of the inverted siphon pipe with the flow change at the two ends, and the pressure change in the inverted siphon pipe is considered to be divided into a low-frequency part and a high-frequency part which are respectively caused by the deformation of the wall of the siphon pipe and the reflection of sound waves; the expression is as follows:
in the above formula:
h2lowis the low-frequency pressure increment at the downstream of the inverted siphon/pressure pipeline, and has the unit of m;
h2highis the high-frequency pressure increment at the downstream of the inverted siphon/pressure pipeline, and the unit is m;
a is the wave speed of pressure waves in m/s;
l is the length of the inverted siphon/pressure pipeline, and the unit is m;
Bslis the slot width used in the pleiman slot method, in m;
Afis the section wet circumference, unit m;
the open channel section channel pool takes an integral-time lag model as a prediction process model, and the long inverted siphon/pressure pipeline takes the large inverted siphon pressure linear model as a prediction process model. The sluice flow formula of the gate/valve is coupled with the upstream and downstream hydraulic connection among the ditches, and the Henry formula is selected as the flow of the sluice used in the invention. The Henry formula is a set of unified calculation method for calculating the free outflow and the submerged outflow brake flow, which is proposed by Henry et al in 1950, and is well known to be more reliable, and the calculation formula is as follows:
in the formula:
q is the flow of the passing gate, m3/s；
e is the gate opening, m;
Cdis the brake-passing flow coefficient;
b is the gate width, m;
g is the acceleration of gravity, m/s2；
HupThe depth of water before the gate, m;
Hdownthe water depth after the gate, m.
Further, in step 4, the objective function is as follows:
minJ＝G×NISE+W×Δej
in the above formula: j is an objective function, the hydraulic response process of each group of outlet gate increment schemes is predicted according to the internal process model, and a gate opening increment scheme which enables the J value of the objective function to be minimum is selected; NISE is a non-dimensionalized water level error square integral, and measures the pressure control stability of the long inverted siphon/pipeline outlet section; n is the number of selection schemes; delta ejThe opening increment scheme of the jth outlet gate/valve is adopted; hdIs longA pressure head, m, of the inverted siphon/pipe outlet section; G. w is a penalty coefficient and is obtained by trial calculation.
The invention has the following beneficial effects: (1) by a model predictive control technology, the hydraulic response process in the long inverted siphon/pressure pipeline is predicted in advance, and the phenomenon that the pressure in the pipeline is over-limited or too low due to unreasonable operation is avoided; (2) the stability of pressure fluctuation control in the pipe by the outlet gate/valve is fully considered, the action frequency of the gate/valve is reduced as much as possible, and the accident loss of the gate/valve is reduced; (3) the emergency control effect of the pipe-channel combined system is improved through an advanced automatic control technology, and the rapidity, the safety and the stability of response are improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a control flow diagram of an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an embodiment of the present invention.
Fig. 3 is a schematic flow chart of a gate opening back-calculation module in the embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in further detail with reference to the drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
FIG. 2 is a schematic view of a channel system of a conventional large pipe-channel combination system, in which a long inverted siphon is connected to an upstream open channel and a downstream open channel. In the system, the length of the inverted siphon is long, and the pressure change in the siphon tube in the regulation and control process has the characteristics of strong nonlinearity and strong coupling property and is mutually influenced with the upstream and downstream open channels. The emergency control of the long inverted siphon in the emergency dispatching is a major difficulty in ensuring the safety of the whole project. As shown, the generally long inverted siphon pool can be divided into an inlet section, a trunk section, and an outlet section. The inlet section is an open channel maintenance gate (No. i-1 gate) and an inlet pressure regulating pool; the pipe body section is generally a plurality of large-caliber prestressed steel cylinder concrete pipes (PCCP pipes); the outlet section is a check gate and a stilling pool. The check gate of the outlet section is positioned at the junction of the pressure flow section and the non-pressure flow section, and the hydraulic response laws at two sides of the gate are completely different due to the great difference of the wave velocity of the pressure flow and the non-pressure flow. Small variations in the movement of the outlet gate can result in large differences in the hydraulic response in the siphon.
Fig. 1 is a control flow chart of an embodiment of the present invention, which takes the occurrence of a flow blockage accident at a downstream of a long inverted siphon at a certain time as an example. Initial flow rate of Q0When a sudden downstream flow blockage accident (such as gate failure, channel blockage, etc.), other buildings along the line need to be adjusted to a new stable flow state (Q) as soon as possibletarget) So as to reduce water loss or avoid secondary accidents such as downstream overtopping and the like. When the control center knows that the emergency scheduling control module is opened immediately when an emergency accident happens, the open channel check gate can make an action time schedule according to an open channel emergency scheduling plan made in the previous period (such as an emergency scheduling plan recorded by hydraulic response and emergency scheduling (tree brocade and the like, south-to-north water transfer and water conservancy science, 10 th volume 10, 5 th period, P161-165) of accident working conditions of a large water delivery channel) or according to the experience of an administrator, the control of the long inverted siphon outlet gate is more complicated, the safety transition of a long inverted siphon pressure section cannot be guaranteed through the plan or the experience control, and the model prediction control module needs to be started.
The model predictive control consists of four major parts, namely an internal process model, a rolling time domain, a constraint condition and an objective function.
The internal process model is a pipe-channel combined system full-line modeling through a linear model, and can quickly obtain the hydraulic response change of a control point through a small amount of information (such as the flow increment of an inlet and an outlet of a channel pool) so as to determine the action of a controller and achieve the aim of real-time online control. The linear model in the invention is as follows:
the linear prediction model of the open channel pool uses an integral-time lag model (ID model) which uses an integral link to simulate a backwater area and a hysteresis linkSimulating a uniform flow area. Area of return water AsAnd the lag time tau is two important parameters of the model and can be obtained by a parameter identification technology (methods such as a least square method or a maximum likelihood method) in the early stage. The model expression is as follows:
in the formula: a. thesIs the water surface area of the water return area m2(ii) a e is the downstream water level deviation, m; x is a position of a water return area; q. q.sinThe variation of the upstream inflow rate of the channel, m3/s；qoutIs the variation of the outflow, m, downstream of the channel3S; t is time, s; tau is lag time, i.e. the time, s, taken by the upstream flow change to pass through the uniform flow region to the downstream return region.
Discretizing the formula (1) to obtain a formula (2)
In the formula, hi1The water level increment of the upstream of the ditch pool i is in unit m; h isi2Is the water level/pressure increment of the downstream of the canal pond i in the unit of m; q. q.si1For the increment of the inflow flow at the upstream of the ditch pool i, the unit m3/s；qi2Is the downstream outflow flow increment of the canal pit i in m3S; t is the calculation time step length, and is taken for 1 min; k is a calculation step; a. thesiIs the area of the backwater area of the canal pool i in unit m2；kdiIs the lag time step for the trench pool i. Wherein A of four canal ponds (canal pond i-2, canal pond i-1, canal pond i +1 and canal pond i +2) issAnd kdThe early stage is obtained by a parameter identification technology (methods such as a least square method or a maximum likelihood method).
The inverse siphon pressure prediction uses the linear model newly proposed in 2020 by scholars of mao zhonghao, guan xuan and the like. The model considers that the pressure variation in the inverted siphon is divided into a low frequency part and a high frequency part, and the expression is as follows:
in equation (3): h is2lowIs the low-frequency pressure increment, m, of the downstream end of the inverted siphon/pressure pipeline; h is2highIs the high-frequency pressure increment, m, of the downstream end of the inverted siphon/pressure pipeline; a is the wave speed of pressure wave, which is 1000m/s in the invention; l is the length of the inverted siphon/pressure pipeline, m; b isslIs the slot width, m, used in the pleisman slot process; a. thefSection wet circumference, m.
Discretizing the formula (3) to obtain a formula (4)
In the formula (4), hi1Is the water level increment of the upstream of the ditch pool i, m; h isi2Is the water level/pressure increment of the downstream of the canal pond i, m; q. q.si1For the increment of the inflow flow upstream of the channel i, m3/s；qi2For the increment of the outflow flow m downstream of the channel i3/s；hf(ii) an on-way head loss, m; t is simulation time step length, and is taken for 1 min; a is the cross-sectional area of the pipeline, m2(ii) a L is the length of the pipeline, m; a is the wave speed of pressure wave, and is 1000 m/s; g is the acceleration of gravity, m/s2；kdiIs the lag time step for the trench pool i.
From the discretization form of the linear model, it can be seen that the ID model suitable for the open channel or the linear model suitable for the large inverted siphon is based on the calculation/actual measurement result of the k step, and the water level/pressure increment downstream of the channel pool of the k +1 step is calculated.
And the linear prediction model of each channel pool is coupled upstream and downstream through the gate passing flow at the gate. Wherein the brake passing flow of each gate is calculated by using a Henry formula. The Henry formula requires input of the water level before the gate, the water level after the gate, and the gate opening. The water level before the gate, namely the water level at the downstream end of the upstream channel pond of the gate, is calculated by a linear model; the water level after the gate, namely the water level at the upstream end of the downstream canal pond, cannot be predicted by the linear model, but the water level after the gate can be assumed to be constant all the time under the condition that the prediction domain is small (for example, only 5-10 steps in the future are predicted); the gate opening, the control gates (i-2, i-1 and i +1) on the open channel control the opening change process according to the pre-established emergency scheduling plan, and the overlength inverted siphon outlet gate (i) controls the opening change according to the proposed model prediction control method.
The sampling period of the sensor and the calculation time step length of the prediction model are both 1min, the sensor is arranged at the upstream end and the downstream end of each channel pool, and the upstream water level, the downstream water level, the inflow flow and the outflow flow of each channel pool are recorded. Assuming that an accident occurs somewhere at the time T ═ 1h (k ═ 60, i.e., step 60), the subsequent opening processes of the gates i-2, i-1, and i +1 can be obtained according to a pre-established emergency scheduling plan, and the opening change process of the ultra-long inverted siphon outlet gate (gate i) needs to be determined by a model predictive control technique. Hypothesis prediction time domain TpredictFor 5min (i.e. predicting the hydraulic response for 5 steps in the future), the prediction process is as follows:
and (61) predicting:
taking i as an example of 3, at this time, the trench pool i-2, the trench pool i-1, the trench pool i +1, and the trench pool i +2 are respectively denoted as trench pool 1, trench pool 2, trench pool 3, trench pool 4, and trench pool 5; gate i corresponds to gate 3, and so on.
S1, for the ditch pool 1, increasing the inflow flow rate by q11(60-kd1) Outflow delta q12(60) Downstream water level increment h12(60) And recording results according to the sensor. Calculation of h by ID model12(61) Therefore, the water level value H of the downstream of the 61 st step ditch pond 1 can be predicted12(61). The water level after the gate of the gate 1 is assumed to be constant (i.e. H)21(61)＝H21(60)). The opening degree of the gate 1 in the 61 st step can be obtained according to an open channel emergency scheduling plan, and the passing gate flow Q of the gate 1 in the 61 st step can be calculated according to a Henry formula1(61). The outflow from the channel 1 is the inflow of the channel 2, so that q can be calculated12(61)、q21(61)。
S2, for the ditch pond 2, the same principle as the ditch pond 1.
S3, for the ditch pool 3, increasing the inflow flow rate by q31(60-kd3) And q is31(60) Upper part ofWater level increment h of swimming pressure regulating pool31(60) And recording results according to the sensor. Calculating the pressure increment h of the long inverted siphon outlet section through a large inverted siphon pressure linear model32(61) From this, the pressure value H of the 61 st step inverted siphon outlet section can be predicted32(61). The water level after the gate of the gate 3 is assumed to be constant (i.e. H)41(61)＝H41(60)). The opening degree of the brake 3 in the 61 st step has 10 opening degree increment schemes, and each scheme can calculate the passing brake flow Q of the brake 3 in the 61 st step through a Henry formula3(61). The outflow from the channel 3 is the inflow to the channel 4, and q can be calculated32(61)、q41(61)。
S4, the same applies to the ditch pond 4 as to the ditch pond 1.
S5, the same applies to the ditch pond 5 as to the ditch pond 1.
The prediction process of the 62 th to 65 th steps is the same as the prediction process of the 61 st step.
Finally, the prediction time domain T of the ultra-long inverted siphon outlet section under each opening increment scheme can be obtainedpredictAnd in the internal pressure change process, selecting an opening increment scheme which is most suitable for the 61 st step of the long inverted siphon outlet gate according to the constraint condition of the model prediction technology and an objective function.
The rolling time domain may be decomposed into a predicted time domain TpredictAnd a control interval TintervalTo understand. In this example TpredictAnd TintervalAre all taken as 10min, namely at the current time TcAnd predicting the hydraulic response process of each group of outlet gate increment schemes within 10min in the future, and selecting the optimal outlet gate increment scheme to implement through the following constraint conditions and an objective function. In the invention, 10 groups of gate opening increment schemes are selected for model prediction, and are [ -1.0m, -0.75m, -0.5m, -0.25m,0m,0.5m,1m,1.5m,2.0m and 2.5m]. At the maximum opening and closing speed V of the gatemaxThe action of the outlet gate is executed, and the maximum opening and closing speed V of the gate is realizedmaxThe flow rate was taken to be 0.5 m/min. After the execution of the optimal outlet gate increment scheme is finished, the opening of the long inverted siphon outlet gate is kept constant until a certain control interval time TintervalThen the model prediction process is carried out again, namely at Tc+TintervalThe prediction and comparison selection of multiple groups of schemes are carried out again at the momentAnd (6) carrying out the process.
The constraint condition is the characteristic of the model prediction technology and is also the key point of the invention. By considering the safety constraint condition of the long inverted siphon, the occurrence of secondary accidents such as pipe explosion due to pressure overrun or cavitation due to too low pressure and the like can be avoided in the process of emergency control. In general emergency control, the long inverted siphon/pipeline outlet end can cause violent water hammer phenomenon because gate/valve operation, for guaranteeing the engineering safety of long inverted siphon, regard the pressure limit of export section as the safety constraint condition, promptly:
Pmin≤P≤Pmaxformula (5)
In the formula: pminRepresents the lower pressure limit, m, of the outlet section of the long inverted siphon/pressure pipeline; pmaxRepresents the upper pressure limit, m, of the outlet section of the long inverted siphon/pressure pipeline; p represents the instantaneous pressure of the outlet section of the long inverted siphon/pressure pipeline, m;
and the target function is the core of the optimal control technology such as model prediction control and the like, and the optimal outlet gate opening increment scheme is selected through the target function. The objective function used in the present invention is as follows:
minJ＝G×NISE+W×Δejformula (6)
In the formula: j is an objective function, the hydraulic response process of each group of outlet gate increment schemes is predicted according to the internal process model, and a gate opening increment scheme which enables the J value of the objective function to be minimum is selected; NISE is a non-dimensionalized water level error square integral, and measures the pressure control stability of the long inverted siphon/pipeline outlet section; n is a selection scheme number, and n is 10; delta ejThe opening increment scheme of the jth outlet gate/valve is adopted; hdIs the pressure head of the long inverted siphon/pipeline outlet section, m; G. w is a penalty coefficient obtained by trial calculationObtaining, in the invention, G is 10;
because many simplifications and assumptions are made in the derivation process of linear prediction models (such as an integral-time lag model, a large-scale inverted siphon pressure linear model and the like), a prediction result of the linear prediction models has certain deviation from the actual hydraulic response of the system. If the model prediction module is continuously opened, the outlet gate is likely to continuously operate, and the system cannot be stable, so that the closing condition of the model prediction control module needs to be established. As shown in fig. 1, the shutdown model prediction module needs to satisfy two conditions simultaneously: first, the pressure at the outlet section of the long inverted siphon is T in the pastsafeIs always maintained within a safe pressure range in time, T in the inventionsafeThe pressure of the long inverted siphon outlet section is taken as 1h, namely the pressure of the long inverted siphon outlet section is always in a safe range within a certain duration of 1h after an accident occurs, the maximum water hammer pressure wave crest can be basically considered to be safely transited, the long inverted siphon is basically safe, and then the new stable state can be gradually reached by depending on the self characteristics of water in a canal system; and secondly, the outlet gate reaches the target opening degree under the regulation of the model predictive controller. And when the two conditions are simultaneously met, the model prediction control module automatically closes and opens the outlet gate opening back calculation module.
The target opening degree set by the model prediction control module is only a preliminary value, and may have a certain deviation with the final opening degree of the actual steady state, so that the gate opening degree back-calculation module is required to be further finely adjusted, so that the system reaches the target steady state as soon as possible. In the invention, the front and rear water heads H of the gate of the long inverted siphon outlet gate are obtained in real timeupAnd HdownCarrying out inverse calculation on a Henry formula by an iterative trial calculation method and obtaining a target flow QtargetAnd outputting the gate action after the corresponding target opening value is obtained. In order to control the opening and closing frequency of the outlet gate, the gate opening degree back-calculation module is arranged at intervals of TintervalThe time is performed once and the influence of the dead zone of the gate is taken into account. In the invention, the dead zone of the gate is 5cm, namely, when the difference value between the output gate opening of the gate opening back-calculation module and the current gate opening is less than 5cm, the opening of the outlet gate is not changed. To prevent the pressure in the inverted siphon from severely fluctuating due to the action output of the gate opening reverse calculation module and limit the outlet gate at the momentDoor opening and closing at minimum speed VminRegulating, the minimum opening and closing speed V of the gate in the inventionminThe concentration was taken to be 0.1 m/min.
The emergency dispatching control mode can predict the hydraulic response process in the long-reverse siphon/pressure pipeline in advance through a model prediction control technology, avoid the over-limit or over-low pressure in the pipeline caused by wrong commands, fully consider the stability of the pressure fluctuation control of the outlet gate/valve to the pipeline, reduce the action frequency of the gate/valve and reduce the accident loss of the gate/valve. The advanced automatic control technology can improve the emergency control effect of the pipe-channel combined system and improve the rapidity, safety and stability of response.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
1. In the model prediction control module, the target opening value e of the long inverted siphon/pipeline outlet gate/valvetarget0Computing
The target opening value of the long inverted siphon/pipeline outlet gate/valve refers to the target opening value required to be determined in model predictive control. There may be two calculation methods: calculating an on-way water surface line by using the flow and the downstream water level in a stable state after an accident through an energy equation, reading the water depth before and after an outlet gate/valve, and calculating the target gate opening through a gate opening back-calculation module; and secondly, directly determining a target opening value according to the proportion of the target steady-state flow and the initial-state flow. Due to the existence of the long inverted siphon/pressure pipe, different adjustment processes may cause different pressure states in the pipe, thereby affecting the water depths of the upstream and downstream open channels, and causing the final stable opening degree of the outlet gate/valve to be difficult to determine in advance. The target opening value calculated by the method I or the method II is a preliminary value, and after the model predictive control module is closed, the gate opening back-calculation module is used for further fine adjustment, so that the system reaches a target stable state as soon as possible.
The target opening degree calculated by the two methods is a preliminary value, but the comparison shows that the first method is more complex, so the method adopts the second method to calculate the target opening degree value of the outlet gate/valve required by model prediction control, and the calculation formula is as follows:
in the formula:
Q0for initial state flow before accident, m3/s；
QtargetFor the target steady state flow m after the emergency dispatching of the accident3/s；
e0The initial opening of an outlet gate/valve m before an accident occurs;
etarget0an outlet gate/valve target opening value m required for model predictive control;
2. gate opening degree back calculation module
Target steady state flow Q after known accident emergency dispatching is finishedtargetAnd calculating the gate opening according to the real-time water heads at the upstream and the downstream of the outlet gate/valve, and performing back calculation of a gate passing flow formula. The brake flow coefficient C can be known from the Henry formuladAnd the calculation formulas of the passing gate flow Q both contain the opening e, the reverse calculation cannot be carried out through simple formula transformation, the calculation needs to be determined through iterative trial calculation, and the trial calculation flow is shown in fig. 3.
The above embodiments are merely illustrative of the present invention and are not to be construed as limiting the invention. Although the present invention has been described in detail with reference to the embodiments, it should be understood by those skilled in the art that various combinations, modifications or equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention, and the technical solution of the present invention is covered by the claims of the present invention.
Claims (3)
1. A large pipe-channel combined system emergency dispatching control method based on model predictive control is characterized by comprising the following steps:
step 1, determining the type and the occurrence time of a sudden accident of a pipe-channel combined system, and starting a large-scale pipe-channel combined system emergency dispatching control module based on model predictive control; the control buildings on the open channel are controlled according to the existing emergency scheduling plan, and the long inverted siphon/pipeline outlet gate/valve is regulated by the model prediction control module;
step 2, setting a plurality of groups of long inverted siphons/pipeline outlet gates/valve opening increment schemes; determining the upper pressure limit P of the outlet section of the long-reverse siphon/pressure pipelinemaxAnd a lower pressure limit PminAs a pipeline pressure safety constraint;
step 3, constructing a linear prediction model of the pipe-channel combination system by using an open channel integral-time lag model and a large-scale inverted siphon pressure linear model, and rapidly predicting the opening increment scheme of each group of outlet gates/valves in a set prediction time domain TpredictInternal hydraulic response processes;
step 4, excluding the pipeline pressure safety constraint condition in the step 2 from the prediction time domain TpredictAn unreasonable scheme that the pressure is over-limited and the tube is burst or the pressure is too low and the cavitation is corroded can be caused in the process; the optimal outlet gate/valve increment scheme is selected by considering the optimal control on the pressure fluctuation and the gate opening and closing frequency through an objective function;
step 5, the long inverted siphon/pipeline outlet gate/valve acts according to the optimal outlet gate/valve increment scheme selected by the model predictive control algorithm in the step 4, and the open channel gate is opened and closed according to the existing emergency scheduling plan in the step 1; when the long inverted siphon/pipeline outlet gate/valve is opened and closed at the maximum opening and closing speed VmaxAfter finishing the action, keeping the opening unchanged until the next prediction control is started;
step 6, controlling the interval time TintervalThen repeating the step 3 to the step 5;
step 7, when the long inverted siphon/pipeline outlet section is in a safe time TsafeThe pressure fluctuation in the device is always maintained in a safe range, and the long inverted siphon/pipeline outlet gate/valve is at a target opening value etarget0Closing the model predictive control module;
step 8, through a gate opening reverse calculation module, the minimum opening and closing speed V of the gate is usedminControl long inverted siphonPipeline outlet gate/valve until target steady flow is reached;
in the step 3, the open channel integral-time lag model generalizes the channel pool into a uniform flow area and a water return area, wherein the water depth of the uniform flow area is normal, the existence of the uniform flow area is considered to lead the response of the downstream water level to lag the change of the upstream flow, and the lag time of a channel completely positioned in the water return area is 0, the model uses an integral link to simulate the water return area, uses a lag link to simulate the uniform flow area, and the expression is as follows:
in the above formula:
Asis the water surface area of the backwater area in unit of m2；
e is the downstream water level deviation in m;
x is a position of a water return area;
qinthe variation of the upstream inflow flow of the channel pond in m3/s；
qoutIs the variation of the downstream outflow of the channel pond in m3/s；
t is time, unit s;
tau is lag time, namely the time for transmitting the upstream flow change to the downstream water return area through the uniform flow area, and the unit is s;
the large-scale inverted siphon pressure linear model relates the pressure change at the downstream end of the inverted siphon pipe with the flow change at the two ends, and the pressure change in the inverted siphon pipe is considered to be divided into a low-frequency part and a high-frequency part which are respectively caused by the deformation of the wall of the siphon pipe and the reflection of sound waves; the expression is as follows:
in the above formula:
h2lowis the low-frequency pressure increment at the downstream of the inverted siphon/pressure pipeline, and has the unit of m;
h2highis the high-frequency pressure increment at the downstream of the inverted siphon/pressure pipeline, and the unit is m;
a is the wave speed of pressure waves in m/s;
l is the length of the inverted siphon/pressure pipeline, and the unit is m;
Bslis the slot width used in the pleiman slot method, in m;
Afis the section wet circumference, unit m;
the open channel section channel pool takes an integral-time lag model as a prediction process model, and the long inverted siphon/pressure pipeline takes the large inverted siphon pressure linear model as a prediction process model; the upper and lower water power connections are coupled by a gate/valve gate flow formula among the channels and ponds, and the gate flow is selected from a Henry formula; henry's formula is calculated as follows:
in the formula:
q is the flow of the passing gate, m3/s；
e is the gate opening, m;
Cdis the brake-passing flow coefficient;
b is the gate width, m;
g is the acceleration of gravity, m/s2；
HupThe depth of water before the gate, m;
Hdownthe water depth after the gate is m;
in step 4, the objective function is as follows:
minJ＝G×NISE+W×Δej
in the above formula: j is an objective function, the hydraulic response process of each group of outlet gate increment schemes is predicted according to the internal process model, and a gate opening increment scheme which enables the J value of the objective function to be minimum is selected; NISE is a non-dimensionalized water level error square integral, and measures the pressure control stability of the long inverted siphon/pipeline outlet section; n is the number of selection schemes; delta ejThe opening increment scheme of the jth outlet gate/valve is adopted; hdIs the pressure head of the long inverted siphon/pipeline outlet section, m; G. w is a penalty coefficient and is obtained by trial calculation.
2. The emergency dispatching control method of the large pipe-channel combined system according to claim 1, wherein: the prediction time domain TpredictIs 5-30 min; the control interval time TintervalIs 5-30 min.
3. The emergency dispatching control method of the large pipe-channel combined system according to claim 1, wherein: the safety time TsafeIs 0.5-2 h.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110226122.2A CN113093531B (en) | 2021-03-01 | 2021-03-01 | Large pipe-channel combined system emergency dispatching control method based on model predictive control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110226122.2A CN113093531B (en) | 2021-03-01 | 2021-03-01 | Large pipe-channel combined system emergency dispatching control method based on model predictive control |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113093531A CN113093531A (en) | 2021-07-09 |
| CN113093531B true CN113093531B (en) | 2022-03-15 |
Family
ID=76667698
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110226122.2A Active CN113093531B (en) | 2021-03-01 | 2021-03-01 | Large pipe-channel combined system emergency dispatching control method based on model predictive control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113093531B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113550273B (en) * | 2021-09-03 | 2022-08-19 | 北京奥特美克科技股份有限公司 | Gate opening control method and device and electronic equipment |
| CN114415742B (en) * | 2021-12-13 | 2024-04-02 | 中国电建集团华东勘测设计研究院有限公司 | Automatic regulation and control device and method for stabilizing water level of regulating tank at front end of long open channel |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3695783A1 (en) * | 2019-02-15 | 2020-08-19 | Origin Wireless, Inc. | Method, apparatus, and system for wireless gait recognition |
| CN112036092A (en) * | 2020-07-10 | 2020-12-04 | 江苏省海洋资源开发研究院(连云港) | River flow prediction method based on relation between river intermediate surface speed and river width |
| CN112232138A (en) * | 2020-09-25 | 2021-01-15 | 天津大学 | Channel slope damage intelligent identification method based on superpixel characteristics |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| RU2325683C2 (en) * | 2006-07-19 | 2008-05-27 | Государственное образовательное учреждение высшего профессионального образования "Кузбасская государственная педагогическая академия" (ГОУ ВПО "КузГПА") | Method of objects identification in operating systems |
| KR20130090278A (en) * | 2012-02-03 | 2013-08-13 | 이메트릭스 주식회사 | Real time cotrolling system and method based area modeling |
| AU2013243238B2 (en) * | 2012-04-05 | 2017-05-18 | Rubicon Research Pty Ltd | Supervisory control of automated irrigation channels |
| CN105279704B (en) * | 2015-09-25 | 2019-02-01 | 天津大学 | The regulation method for generating plan of long distance water transfer project reply burst water contamination accident |
| EP3313134B1 (en) * | 2016-10-21 | 2020-09-30 | ABB Schweiz AG | A method of scheduling field devices in a wireless network of an industrial process system |
| CN106647289B (en) * | 2017-03-06 | 2019-09-10 | 武汉大学 | The feed forward control method of gate delay opening and closing time is dispatched in open channel water delivery |
| CN108181940B (en) * | 2017-12-27 | 2020-08-04 | 中国水利水电科学研究院 | Gate regulating and controlling method suitable for downstream emergency water cut-off condition of series channel |
| CN108519783B (en) * | 2018-03-30 | 2020-10-16 | 武汉大学 | Pre-gate control point variable target water level control system and method based on channel pool storage balance |
| CN109197539B (en) * | 2018-10-25 | 2020-07-31 | 清华大学 | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels |
| 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 |
| CN110647039B (en) * | 2019-10-08 | 2022-03-25 | 黄河勘测规划设计研究院有限公司 | Synchronous control self-adaptive balance scheduling method for long-distance open channel water delivery project |
| 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 |
| CN112418546B (en) * | 2020-12-04 | 2023-04-07 | 四川大学 | System for predicting front floating ice state of open channel gate and construction method and application thereof |
-
2021
- 2021-03-01 CN CN202110226122.2A patent/CN113093531B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3695783A1 (en) * | 2019-02-15 | 2020-08-19 | Origin Wireless, Inc. | Method, apparatus, and system for wireless gait recognition |
| CN112036092A (en) * | 2020-07-10 | 2020-12-04 | 江苏省海洋资源开发研究院(连云港) | River flow prediction method based on relation between river intermediate surface speed and river width |
| CN112232138A (en) * | 2020-09-25 | 2021-01-15 | 天津大学 | Channel slope damage intelligent identification method based on superpixel characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113093531A (en) | 2021-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113093531B (en) | Large pipe-channel combined system emergency dispatching control method based on model predictive control | |
| CN113236581B (en) | Intelligent parallel pump system and optimal adjustment method | |
| JP7346705B2 (en) | Water level prediction and control method for series water channel based on fuzzy neural network | |
| CN110647039B (en) | Synchronous control self-adaptive balance scheduling method for long-distance open channel water delivery project | |
| CN104865830B (en) | Dual-intelligent-optimization control method for unit load | |
| CN109197539B (en) | Model predictive control algorithm-based automatic control method for irrigation multi-stage channels | |
| CN108519783B (en) | Pre-gate control point variable target water level control system and method based on channel pool storage balance | |
| DE102010022763A1 (en) | Method for automatic hydraulic balancing in fluid-carrying systems | |
| CN108518204A (en) | A kind of gap gas lift liquid drainage gas production control method and system | |
| CN111642449A (en) | Slope-variable water tank for testing swimming capacity of fishes | |
| CN113343595B (en) | Inversion model of open channel water delivery system accident and method for determining accident flow and position | |
| CN113361214A (en) | Open channel control model parameter identification method based on water level flow data | |
| CN112016754A (en) | Power station boiler exhaust gas temperature advanced prediction system and method based on neural network | |
| CN112596386A (en) | Matlab-based urban drainage system simulation control mixed model with mechanism model, concept model and data model | |
| CN114936742A (en) | Water supply system scheduling agent decision method | |
| CN103412479B (en) | Paired running oxygen-eliminating device intelligent control method | |
| CN112733367A (en) | Water transfer project ice period water delivery scheduling method based on seven-day weather forecast | |
| CN106200381A (en) | A kind of according to processing the method that water yield control by stages water factory runs | |
| CN103088783B (en) | Reservoir regulation scheme generating method facing ecology | |
| CN109706906A (en) | A kind of triangle gate ship lock opens the control method of lock operation | |
| CN105926550A (en) | Scheduling method for preventing pressure forebay of hydropower station with overlong open diversion channel from being empty | |
| CN110043744A (en) | A method of determining high lift water-carriage system termination of pumping accident pump late gate Closing Law | |
| CN116070386A (en) | Intelligent design and evaluation method and system for water network engineering | |
| CN213203971U (en) | City landscape water quality regulation and control device based on slowly hold put soon | |
| Kong et al. | Automatic feedback control algorithm for canal for a quick upstream water supply interruption in the case of an emergency |
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 |