CN109164704A - A kind of air compressor machine group Optimization Scheduling based on mixed model - Google Patents
A kind of air compressor machine group Optimization Scheduling based on mixed model Download PDFInfo
- Publication number
- CN109164704A CN109164704A CN201810888070.3A CN201810888070A CN109164704A CN 109164704 A CN109164704 A CN 109164704A CN 201810888070 A CN201810888070 A CN 201810888070A CN 109164704 A CN109164704 A CN 109164704A
- Authority
- CN
- China
- Prior art keywords
- air compressor
- compressor machine
- air
- energy consumption
- compression station
- 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.)
- Granted
Links
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)
- Control Of Positive-Displacement Pumps (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to a kind of air compressor machine group Optimization Scheduling based on mixed model, belongs to information technology field.The present invention establishes the mixed model that imitation technology is combined with combined schedule Optimal Decision-making to solve blindness present in current air compressor group group scheduling.The present invention updates the air compressor machine energy consumption model sample set using expertise building using the real-time sample dynamic of industry spot, then on-line study is carried out using relevant parameter of the least-squares algorithm to air compressor machine energy consumption model, and application simulation emulation technology and layering tree search algorithm solve air compressor machine group Multiobjective Optimal Operation model, overall merit finally is carried out to the scheduling scheme of acquisition using entropy assessment and Technique for Order Preference by Similarity to an Ideal Solution, to assist the scheduling scheme of spot dispatch personnel formulation safety, energy-saving and environmental protection and economy.The method also has a wide range of applications in different industrial circles.
Description
Technical field
The invention belongs to information technology field, be related to air compressor machine dynamic modeling based on data, model parameter it is online
It updates and the built-up pattern Optimization Solution scheduling theory based on imitation technology and layering tree search algorithm, is a kind of based on mixed
The air compressor machine group Optimization Scheduling of molding type.The present invention is updated using the real-time sample dynamic of industry spot utilizes expertise
Then the air compressor machine energy consumption model sample set of building is carried out using relevant parameter of the least-squares algorithm to air compressor machine energy consumption model
On-line study, and application simulation emulation technology and layering tree search algorithm solve air compressor machine group Multiobjective Optimal Operation model,
Overall merit finally is carried out to the scheduling scheme of acquisition with entropy assessment and Technique for Order Preference by Similarity to an Ideal Solution, to assist spot dispatch personnel
Formulate the scheduling scheme of safety, energy-saving and environmental protection and economy.The method is also widely used valence in different industrial circles
Value.
Background technique
Atmospheric control is using compressed gas as working media, with structure is simple, system maintenance amount is small, component price
The features such as cheap, plays an important role in terms of improving industrial production efficiency and Automation of Manufacturing Process rate.But it compresses
The production of air is the main energy consumption unit in the industrial productions links such as metallurgy, industrial manufacture, bio-pharmaceuticals, the main reason is that
Air compressor group group scheduling it is unreasonable, not in time etc., result in that pneumatic system energy consumption is high, resource utilization is low.Therefore, reasonably
On the one hand air compressor machine machine unit scheduling can realize the structural optimization of resources of production in pneumatic system short-term operation, meet different
Produce user uses gas demand, provides basis for the safety in production and economic load dispatching of enterprise；On the other hand air compressor machine can also be promoted
The operational efficiency of unit makes it operate in optimal operating point as far as possible, to promote the economy and energy conservation of pneumatic system operation
Property.
Common air compressor group scheduling is that the large-scale nonlinear mixed integer optimization comprising multiple constraint conditions is asked
Topic, the solution type according to algorithm are generally divided into two classes: (1) classic algorithm.At present very to the research of such scheduling method for solving
It is more, mainly opening for unit is solved by dynamic programming, MIXED INTEGER method, Lagrangian Relaxation, branch definition method etc.
Open combined strategy.But such algorithm increases with unit number, calculates time-consuming exponentially and increases, such as based on Lagrangian pine
The distributed method of relaxation solves air transportation scheduling problem (Zhang, Y., Su, R., Li, Q., et al. (2017)
.Distributed flight routing and scheduling for air traffic flow
management.IEEE Transactions on Intelligent Transportation Systems,10,2681-
2692.), mixed integer nonlinear programming algorithm solves scheduling (Kim, J.S., Edgar, the T.F. of cogeneration of heat and power steam power plant
(2014).Optimal scheduling of combined heat and power plants using mixed-
integer nonlinear programming.Energy,77,675-690.).(2) biological intelligence method is imitated.Mainly by intelligence
The unlatching combination side of the search unit such as energy optimization algorithm such as genetic algorithm, simulated annealing, particle swarm algorithm, ant group algorithm
Case.Such algorithm, which generally requires, to be run multiple times solution and could obtain optimal solution, such as mixing of combining of population and genetic algorithm
Close model solution semiconductor manufacturing plant combined schedule problem (Jamrus, T., Chien, C.F., et al. (2017) .Hybrid
particle swarm optimization combined with genetic operators for flexible job-
shop scheduling under uncertain processing time for semiconductor
Manufacturing.IEEE Transactions on Semiconductor Manufacturing, 99,1-1.), particle
Real-Time Scheduling (Hu, W., Wang, H., Yan, L., Du, B. (2016) .A swarm of colony optimization algorithm solution traffic lights
intelligent method for traffic light scheduling:application to real urban
traffic networks.Applied Intelligence,1,208-231.)。
In addition to problem above and limitation, how will combine, and realizes fast with machine learning algorithm the characteristics of device model
Fast accurate acquisition globally optimal solution, is the key points and difficulties of presently relevant research.There are multiple air compressor machines in energy supply for metallurgy system
How group, realize the reasonable disposition of the different hollow press of group, belongs to intelligent scheduling problem, difficult point in artificial intelligence field
Be: air compressor machine of the present invention towards different brands, different type, different model, operation mechanism and model parameter respectively have difference
It is different, therefore the air compressor machine mechanism model for establishing a universality is difficult to realize；The consume of equipment during compressor operation,
The drift etc. of parameter requires air compressor machine energy consumption model that should have real-time, just can accurately reflect the actual production energy consumption shape of air compressor machine
Condition；To air compressor machine that may be present open assembled scheme carry out exhaustion be it is very difficult, need one kind that can quick and precisely ask
The method for solving air compressor machine group Model for Multi-Objective Optimization；Consider from practical application, air compressor group assembled scheme is commented there are multiple
Valence index chooses an optimal scheduling scheme so should comprehensively consider each evaluation index from numerous feasible schemes.Mesh
It is preceding also to lack a kind of effective method and be systematic while solving the above problems.
Summary of the invention
The present invention is to solve air compressor group group scheduling problem.The present invention first by the real-time sample of industry spot online more
Newly into model training sample set, and then air compressor machine energy consumption model relevant parameter is learned online using least-squares algorithm
It practises.Then application simulation emulation technology and layering tree search algorithm solve air compressor machine group Multiobjective Optimal Operation model, utilize
Entropy assessment and superiority and inferiority Furthest Neighbor evaluate the air compressor group group scheduling scheme of acquisition.The invention can effectively improve pneumatics
The resource utilization of group of planes group, to provide decision branch for the air compressor group group scheduling that field personnel carries out safety and economy
It holds.
Technical solution of the present invention:
A kind of air compressor machine group Optimization Scheduling based on mixed model, as shown in Fig. 1, the specific steps are as follows:
(1) air compressor machine energy consumption model sample set constructs: the air inlet interior for a period of time of every air compressor machine is obtained from database
Flow, current of electric, exiting exhaust gas pressure, rate of discharge construct the sample set of every air compressor machine energy consumption model.
(2) air compressor machine energy consumption model sample set online updating: air compressor machine energy consumption model sample set online updating can pass through reality
When collecting sample update and timers trigger dynamic updates the dynamic adjustment that two processes realize energy consumption model sample sets.
(3) on-line study of air compressor machine energy consumption model relevant parameter: the pass between air compressor air intake flow and current of electric
System, the relationship between air compressor air intake flow and exiting exhaust gas pressure, the relationship between air compressor air intake flow and rate of discharge
The online updating that least square method realizes air compressor machine energy consumption model relevant parameter can be utilized respectively.
(4) modeling of air compressor machine group Multiobjective Optimal Operation and solution based on analog simulation and layering tree search algorithm:
The air compressor machine Optimal Operation Model established includes multiple objective functions such as economic cost, air compressor machine specific power, using hierarchical tree
Searching algorithm solves air compressor machine Optimal Operation Model, and carries out analogue simulation using air compressor machine energy consumption model.
(5) the scheduling scheme Optimal Decision-making based on entropy assessment and superiority and inferiority Furthest Neighbor: to multiple feasible schedule schemes of acquisition,
Analysis of Policy Making is carried out to the scheduling scheme under different evaluation index using entropy assessment and superiority and inferiority Furthest Neighbor.
Effect and benefit of the invention: the present invention can be realized the rapid solving of air compressor group assembled scheme, feasible schedule
The functions such as the Optimal Decision-making of the analog simulation of scheme and scheduling scheme, when dramatically shortening the formulation of scheduling scheme
Between, the cost of power consumption in production process is reduced, the economic benefit of enterprise's production is improved.The present invention is without the concern for sky
Brand, type and the model of press, but a kind of air compressor machine energy consumption model of universality is established by the method that data are fitted；With
Optimized Operation module afterwards has fully considered the actual motion state of equipment, proposes one kind and is searched based on analog simulation and hierarchical tree
The air compressor machine group combined schedule model of rope algorithm quick and precisely method for solving；In addition, the present invention has also comprehensively considered selection sky
The different evaluation index of press scheme realizes the Optimal Decision-making of feasible schedule scheme using entropy assessment and superiority and inferiority Furthest Neighbor.
Detailed description of the invention
Fig. 1 is the relational graph in the present invention between modules.
Fig. 2 is the composition and plant area's distribution map of air compressor machine group of the present invention.
Fig. 3 is specific implementation flow chart of the invention.
Specific embodiment
Technical solution for a better understanding of the present invention, the present invention is by taking smelter air compressor group group scheduling as an example, knot
Attached drawing 2 is closed to be described in detail embodiments of the present invention.
A kind of air compressor machine group Optimization Scheduling based on mixed model, steps are as follows:
Step 1: the building of air compressor machine energy consumption model sample set
I-th of air compression station jth platform air compressor machine charge flow rate interior for a period of time is obtained from database, goes out throughput, electricity
Electromechanics stream and exiting exhaust gas pressure；It empirically selects to construct the air compressor machine energy consumption mould from the part sample in the above-mentioned period
The original training set of type；Successively initialize the sample set of every air compressor machine in different air compression stations；
Step 2: air compressor machine energy consumption model sample set online updating
Air compressor machine energy consumption model sample set online updating includes that real-time collecting sample updates air compressor machine energy consumption model sample set
Two processes of air compressor machine energy consumption model sample set are updated with timers trigger dynamic；
Real-time collecting sample updates air compressor machine energy consumption model sample set process: a newest acquisition sample is obtained from database
This, calculates in the newest collecting sample and i-th of air compression station jth platform air compressor machine energy consumption model sample set the firstA sample
Euclidean distanceWhenWhen, which only updatesThe time of a sample；WhenWhen,
The air compressor machine energy consumption model increases this newest collecting sample into air compressor machine energy consumption model sample set；σijValue empirically into
Row setting；The energy consumption model sample set of every air compressor machine is successively updated in this way in different air compression stations；
Timers trigger dynamic updates air compressor machine energy consumption model sample set process: empirically setting every air compressor machine energy consumption
The renewal time δ of model sample collectionij；When the of i-th of air compression station jth platform air compressor machineA sampleWhen, pneumatics function
Consumption model forgets the sample, and otherwise, air compressor machine energy consumption model then retains the sample；Every air compressor machine in different air compression stations
Energy consumption model sample set successively in this way carry out dynamic adjustment；
Step 3: the building of air compressor machine energy consumption model and its parameter update
Within an alternate run period, if the charge flow rate of i-th of air compression station jth platform air compressor machine is υij, air compressor machine
Energy consumption is different with unloading three phases power consumption in unlatching, the load of air compressor machine, therefore is expressed as follows using a piecewise function:
Wherein, i-th of air compression station jth platform air compressor machine is in startup stageAnd unloading phaseElectric power energy consumption it is logical
It crosses the integral of the energy consumption to the corresponding period to obtain, is a fixed numerical value；UijWithRespectively indicate i-th of air compression station jth platform
The voltage of air compressor machine and the power factor of driving motor；Ψ(υij) indicate i-th of air compression station jth platform air compressor machine charge flow rate with
Relationship between current of electric, the undated parameter of relationship is fitted updated air compression station energy by least-squares algorithm between the two
Model sample collection is consumed to realize；
Relationship between air compressor machine exiting exhaust gas pressure and charge flow rate is expressed as follows:
Pij=φij(υij) (2)
Wherein, PijAnd υijRespectively indicate i-th of air compression station jth platform air compressor machine load phase exiting exhaust gas pressure and
Charge flow rate, the parameter of relationship, which is updated, between the two is fitted updated air compression station energy consumption model sample by least-squares algorithm
Collection is realized；
Consider the leakage of residual high pressure gas expansion in air compressor machine clearance volume, air cylinder structure volumetric loss and inner passage
Etc. reasons, initially set up in γ period the mapping relations set between i-th of air compression station jth platform air compressor machine inlet and outlet flowUsing its mean value in γ period come characterize air compressor machine inlet and outlet flow between relationship；Air compressor machine stream
It is as follows to measure inlet and outlet relationship calculation formula:
Wherein, υij, Qij、λijRespectively indicate charge flow rate, outlet of i-th of air compression station jth platform air compressor machine in load phase
Flow, charge flow rate and mapping relations between throughput out；
Step 4: the modeling of air compressor machine Optimal Scheduling
1) objective function
1. specific power maximizes
Wherein,Indicate the objective function of air compressor machine specific power in m air compression station, SiIndicate that i-th of air compression station is hollow
The unlatching strategy of press,Indicate that the i-th air compression station jth air compressor machine is opening strategy SiUnder load power, υi′jIndicate i-th
A air compression station jth air compressor machine is opening strategy SiUnder charge flow rate；
2. economic operation cost is minimum
Wherein,Indicate the objective function of air compressor machine economic cost in m air compression station,Indicate the unit price (kw/ of electric energy
Member),It indicates in (t0,t1) strategy S is opening in m air compressor plant in the periodiUnder electricity
Power consumes energy (kw),And εijRespectively indicate the start-up cost of j-th of air compressor machine in i-th of air compression station, unloading at
Sheet and depreciable cost；
2) bound for objective function
Objective function constraint condition is described as follows:
1. the aperture of air compressor gas valve flow constrains
WithRespectively indicate the minimax constraint of i-th of air compression station jth platform air compressor air intake flow；
2. the gas production of air compressor machine constrains
WithRespectively indicate i-th of minimum and maximum constraint that air compression station jth platform air compressor machine goes out throughput；
3. gas production and gas consumption matching constraint
Indicate m air compression stationThe gas production of platform air compressor machine, QneedIndicate air requirements user
Air demand；
4. compressor operation time-constrain
WithRespectively indicate the most short and maximum duration constraint of i-th of air compression station jth platform compressor operation, the constraint
The long-time of the frequent start-stop and air compressor machine that are intended to avoid air compressor machine uses；
5. ductwork pressure variation constraint
HL、HHIt is the pressure bound of pipe network, H respectively0Indicate the original state of air pressure, Δ H is then corresponding variation
Amount；
Step 5: analog simulation and layering tree search algorithm solve pneumatics scheduling model
Originally a kind of layering tree search algorithm is researched and proposed, with the optimum combination of quick obtaining air compressor machine unit；The algorithm is asked
Solution preocess can be divided into three layers from top to bottom, i.e. human-computer interaction layer, decision search layer and equipment selects layer, specific steps are as follows:
1) human-computer interaction layer: Air Compressor Equipment state is that Field Force sets according to production status or production plan；State
State=1 indicates that air compressor machine is in normal condition；State State=0 indicates that air compressor machine is in maintenance or malfunction, is in
The air compressor machine of the state cannot function as the optional equipment of combined schedule solution；
2) decision search layer: the air compressor machine state of human-computer interaction layer setting is to construct the unlatching plan of different air compression station air compressor machines
The slightly basis of set S；Depth-first tree search algorithm traverses each and opens strategy set, first judges in the unlatching strategy set
Whether the unlatching number of units of air compressor machine meets the aggregate demand of user's gas, selects needs into equipment selection layer choosing if meeting and opens
The air compressor machine combination opened, otherwise enters next air compressor machine decision search process；
3) equipment selects layer: depth-first tree search algorithm traverses the combination side under each air compression station air compressor machine unlatching strategy
Case set T, and calculate using Monte Carlo simulation and air compressor machine energy consumption model the economy and specific power of each assembled scheme；?
After all assembled schemes opened under strategy all traverse, starts repetition strategies search layer step and carry out searching for next unlatching strategy
Rope；
Step 6: the scheduling scheme Analysis of Policy Making based on entropy assessment and superiority and inferiority Furthest Neighbor
Device group scheduling scheme often possesses different evaluation indexes, often obtains because a certain evaluation index is only considered
Deviate the scheduling scheme of global optimum；Multiple target entropy assessment is introduced into scheduling scheme formulation process by the present invention, to establish
The decision analytic model of scheduling scheme.Specific calculation process is as follows:
1) the appraisement system building of air compressor machine group operation
The assessment indicator system of air compressor machine group operation includes specific power and operating cost, the former section as air compressor machine
Energy index is the bigger the better, and the latter is the smaller the better as the economical operation index of air compressor machine；
2) standardization of evaluation index numerical value
K scheduling scheme, the characteristic value of a-th of index of b-th of evaluation object are evaluated equipped with θ scheduling evaluation index
For xab, the eigenmatrix for obtaining scheduling scheme is X=(xab)θ×k；Obtained eigenmatrix is standardized, elimination refers to
Due to dimension difference bring difference between mark；Standardization post-processes obtained matrixX is respectively worth standardized method are as follows:
Wherein, I1For air compressor machine specific power index, I2For cost-effectivenes index；
3) calculating of entropy and entropy weight
Determine the weight of each index, the entropy H of i-th of indexaAre as follows:
Wherein, HaIndicate comentropy, this is worth, and the smaller degree of variation for showing the index is bigger, the information content that is capable of providing and
Its effect played in overall merit is also bigger, thus weight is also bigger；ωaIndicate the weight coefficient of a-th of index；
4) definition of ideal disaggregation
The maximum value and minimum value of each evaluation of programme relative lndex are calculated, and then constitutes most ideal solution and negative ideal
Solution, is expressed as follows:
Wherein, X+And X-Ideal disaggregation and minus ideal result collection respectively；
5) each alternative is calculated at a distance from ideal disaggregation
Wherein, D+And D-Object to be evaluated is respectively indicated at a distance from optimal and most inferior solution；
6) comprehensive evaluation index
Wherein, O*∈ (0,1), O*Value it is bigger characterize closer to optimal solution, evaluation object is more excellent.
By taking certain smelter air compressor group system as an example, it is assumed that compressed gas is transmitted as perfect condition, pneumatics in the duct
The total demand of machine user is artificial setting, does not consider that the difference of the power consumption price different time of air compressor machine, i.e. electricity price are pressed without exception
0.458 yuan/kilowatt hour calculates.Table 1 gives the Contrast on effect of the method for the present invention Yu manual dispatching method
Table 1 gives the Contrast on effect of the method for the present invention Yu manual dispatching method
Claims (1)
1. a kind of air compressor machine group Optimization Scheduling based on mixed model, which is characterized in that steps are as follows:
Step 1: the building of air compressor machine energy consumption model sample set
I-th of air compression station jth platform air compressor machine charge flow rate interior for a period of time is obtained from database, goes out throughput, motor electricity
Stream and exiting exhaust gas pressure；It empirically selects to construct the air compressor machine energy consumption model from the part sample in the above-mentioned period
Original training set；Successively initialize the sample set of every air compressor machine in different air compression stations；
Step 2: air compressor machine energy consumption model sample set online updating
Air compressor machine energy consumption model sample set online updating includes that real-time collecting sample updates air compressor machine energy consumption model sample set and determines
When device triggering dynamic update two processes of air compressor machine energy consumption model sample set；
Real-time collecting sample updates air compressor machine energy consumption model sample set process: a newest collecting sample is obtained from database,
It is calculated in the newest collecting sample and i-th of air compression station jth platform air compressor machine energy consumption model sample set firstThe Europe of a sample
Formula distanceWhenWhen, which only updatesThe time of a sample；WhenWhen, it should
Air compressor machine energy consumption model increases this newest collecting sample into air compressor machine energy consumption model sample set；σijValue empirically carry out
Setting；The energy consumption model sample set of every air compressor machine is successively updated in this way in different air compression stations；
Timers trigger dynamic updates air compressor machine energy consumption model sample set process: empirically setting every air compressor machine energy consumption model
The renewal time δ of sample setij；When the of i-th of air compression station jth platform air compressor machineA sampleWhen, air compressor machine energy consumption mould
Type forgets the sample, and otherwise, air compressor machine energy consumption model then retains the sample；The energy of every air compressor machine in different air compression stations
It consumes model sample collection and successively carries out dynamic adjustment in this way；
Step 3: the building of air compressor machine energy consumption model and its parameter update
Within an alternate run period, if the charge flow rate of i-th of air compression station jth platform air compressor machine is υij, the energy consumption of air compressor machine
It is different with unloading three phases power consumption in unlatching, the load of air compressor machine, therefore be expressed as follows using a piecewise function:
Wherein, i-th of air compression station jth platform air compressor machine is in startup stageAnd unloading phaseElectric power energy consumption by pair
The energy consumption of corresponding period, which integrates, to be obtained, and is a fixed numerical value；UijWithRespectively indicate i-th of air compression station jth platform pneumatics
The voltage of machine and the power factor of driving motor；Ψ(υij) indicate i-th of air compression station jth platform air compressor machine charge flow rate and motor
Relationship between electric current, the undated parameter of relationship is fitted updated air compression station energy consumption mould by least-squares algorithm between the two
Type sample set is realized；
Relationship between air compressor machine exiting exhaust gas pressure and charge flow rate is expressed as follows:
Pij=φij(υij) (2)
Wherein, PijAnd υijI-th of air compression station jth platform air compressor machine is respectively indicated in the exiting exhaust gas pressure and inlet air flow of load phase
Amount, the parameter of relationship, which is updated, between the two is fitted updated air compression station energy consumption model sample set reality by least-squares algorithm
It is existing；
Consider the leakage of residual high pressure gas expansion in air compressor machine clearance volume, air cylinder structure volumetric loss and inner passage, it is first
First establish the mapping relations set in γ period between i-th of air compression station jth platform air compressor machine inlet and outlet flowUsing its mean value in γ period come characterize air compressor machine inlet and outlet flow between relationship；Air compressor machine stream
It is as follows to measure inlet and outlet relationship calculation formula:
Wherein, υij, Qij、λijI-th of air compression station jth platform air compressor machine is respectively indicated in the charge flow rate of load phase, out air-flow
Amount, charge flow rate and mapping relations between throughput out；
Step 4: the modeling of air compressor machine Optimal Scheduling
1) objective function
1. specific power maximizes
Wherein,Indicate the objective function of air compressor machine specific power in m air compression station, SiIndicate air compressor machine in i-th of air compression station
Strategy is opened,Indicate that the i-th air compression station jth air compressor machine is opening strategy SiUnder load power, υi′jIndicate i-th of pneumatics
Jth of standing air compressor machine is opening strategy SiUnder charge flow rate；
2. economic operation cost is minimum
Wherein,Indicate the objective function of air compressor machine economic cost in m air compression station,Indicating the unit price of electric energy, kw/ is first,It indicates in (t0,t1) strategy S is opening in m air compressor plant in the periodiUnder electric power consumption
Energy (kw),And εijRespectively indicate the start-up cost of j-th of air compressor machine in i-th of air compression station, unloading cost
And depreciable cost；
2) bound for objective function
Objective function constraint condition is described as follows:
1. the aperture of air compressor gas valve flow constrains
WithRespectively indicate the minimax constraint of i-th of air compression station jth platform air compressor air intake flow；
2. the gas production of air compressor machine constrains
WithRespectively indicate i-th of minimum and maximum constraint that air compression station jth platform air compressor machine goes out throughput；
3. gas production and gas consumption matching constraint
Indicate m air compression stationThe gas production of platform air compressor machine, QneedIndicate the sky of air requirements user
Gas demand；
4. compressor operation time-constrain
WithThe most short and maximum duration constraint of i-th of air compression station jth platform compressor operation is respectively indicated, which is intended to
The long-time of the frequent start-stop and air compressor machine that avoid air compressor machine uses；
5. ductwork pressure variation constraint
HL、HHIt is the pressure bound of pipe network, H respectively0Indicate the original state of air pressure, Δ H is then corresponding variable quantity；
Step 5: analog simulation and layering tree search algorithm solve pneumatics scheduling model
Originally a kind of layering tree search algorithm is researched and proposed, with the optimum combination of quick obtaining air compressor machine unit；The algorithm solved
Journey can be divided into three layers from top to bottom, i.e. human-computer interaction layer, decision search layer and equipment selects layer, specific steps are as follows:
1) human-computer interaction layer: Air Compressor Equipment state is that Field Force sets according to production status or production plan；State State
=1 expression air compressor machine is in normal condition；State State=0 indicates that air compressor machine is in maintenance or malfunction, is in the state
Air compressor machine cannot function as combined schedule solution optional equipment；
2) decision search layer: the air compressor machine state of human-computer interaction layer setting is the unlatching set of strategies for constructing different air compression station air compressor machines
Close the basis of S；Depth-first tree search algorithm traverses each and opens strategy set, first judges pneumatics in the unlatching strategy set
Whether the unlatching number of units of machine meets the aggregate demand of user's gas, and equipment selection layer choosing is entered if meeting and selects what needs were opened
Air compressor machine combination, otherwise enters next air compressor machine decision search process；
3) equipment selects layer: depth-first tree search algorithm traverses the assembled scheme collection under each air compression station air compressor machine unlatching strategy
T is closed, and calculates the economy and specific power of each assembled scheme using Monte Carlo simulation and air compressor machine energy consumption model；All
After assembled scheme under unlatching strategy all traverses, starts repetition strategies search layer step and carry out the tactful search of next unlatching；
Step 6: the scheduling scheme Analysis of Policy Making based on entropy assessment and superiority and inferiority Furthest Neighbor
Device group scheduling scheme often possesses different evaluation indexes, is often deviateed because a certain evaluation index is only considered
The scheduling scheme of global optimum；Multiple target entropy assessment is introduced into scheduling scheme formulation process by the present invention, to establish scheduling
The decision analytic model of scheme；Specific calculation process is as follows:
1) the appraisement system building of air compressor machine group operation
The assessment indicator system of air compressor machine group operation includes specific power and operating cost, the former refers to as the energy conservation of air compressor machine
Mark is the bigger the better, and the latter is the smaller the better as the economical operation index of air compressor machine；
2) standardization of evaluation index numerical value
K scheduling scheme is evaluated equipped with θ scheduling evaluation index, the characteristic value of a-th of index of b-th of evaluation object is
xab, the eigenmatrix for obtaining scheduling scheme is X=(xab)θ×k；Obtained eigenmatrix is standardized, index is eliminated
Between due to dimension difference bring difference；Standardization post-processes obtained matrixX is respectively worth standardized method are as follows:
Wherein, I1For air compressor machine specific power index, I2For cost-effectivenes index；
3) calculating of entropy and entropy weight
Determine the weight of each index, the entropy H of i-th of indexaAre as follows:
Wherein, HaIndicate comentropy, this is worth, and the smaller degree of variation for showing the index is bigger, the information content that is capable of providing and its
Effect played in overall merit is also bigger, thus weight is also bigger；ωaIndicate the weight coefficient of a-th of index；
4) definition of ideal disaggregation
The maximum value and minimum value of each evaluation of programme relative lndex are calculated, and then constitutes most ideal solution and minus ideal result,
It is expressed as follows:
Wherein, X+And X-Ideal disaggregation and minus ideal result collection respectively；
5) each alternative is calculated at a distance from ideal disaggregation
Wherein, D+And D-Object to be evaluated is respectively indicated at a distance from optimal and most inferior solution；
6) comprehensive evaluation index
Wherein, O*∈ (0,1), O*Value it is bigger characterize closer to optimal solution, evaluation object is more excellent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810888070.3A CN109164704B (en) | 2018-08-07 | 2018-08-07 | Air compressor group optimal scheduling method based on hybrid model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810888070.3A CN109164704B (en) | 2018-08-07 | 2018-08-07 | Air compressor group optimal scheduling method based on hybrid model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109164704A true CN109164704A (en) | 2019-01-08 |
| CN109164704B CN109164704B (en) | 2021-03-26 |
Family
ID=64895111
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810888070.3A Active CN109164704B (en) | 2018-08-07 | 2018-08-07 | Air compressor group optimal scheduling method based on hybrid model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109164704B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886493A (en) * | 2019-02-26 | 2019-06-14 | 江南大学 | A kind of Logistics System Design method based on improvement multi-objective particle swarm algorithm |
| WO2020140564A1 (en) * | 2019-01-02 | 2020-07-09 | 大连理工大学 | Optimization scheduling decision method employing simulation technique for air compressor assembly |
| CN111736466A (en) * | 2020-06-08 | 2020-10-02 | 武汉理工大学 | Optimal control method and system for quick load shedding system of semi-submersible platform |
| CN111831963A (en) * | 2020-07-14 | 2020-10-27 | 中国南方电网有限责任公司 | Method for evaluating controllability of comprehensive energy service provider under power market background |
| CN113537644A (en) * | 2021-08-23 | 2021-10-22 | 中冶赛迪技术研究中心有限公司 | Multi-air compression station dynamic collaborative optimization regulation and control system and method |
| CN113606111A (en) * | 2021-09-09 | 2021-11-05 | 广东鑫钻节能科技股份有限公司 | Energy-saving protection system based on air compression station and implementation method thereof |
| CN113669242A (en) * | 2021-08-03 | 2021-11-19 | 新奥数能科技有限公司 | Power control method and device of air compressor system and computer equipment |
| CN115343967A (en) * | 2022-10-19 | 2022-11-15 | 蘑菇物联技术（深圳）有限公司 | Method, apparatus, and medium for simulation control of air compression station |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101718270A (en) * | 2009-11-20 | 2010-06-02 | 上海应用技术学院 | Prediction and pressure regulation method for control system of air compressor |
| CN104635684A (en) * | 2014-12-25 | 2015-05-20 | 冶金自动化研究设计院 | Cluster optimization control system for air compressor |
| CN106651182A (en) * | 2016-12-25 | 2017-05-10 | 北京工业大学 | Rail passenger flow congestion risk evaluation method based on entropy weight |
| CN206190525U (en) * | 2016-11-04 | 2017-05-24 | 大同煤矿集团有限责任公司 | Energy -conserving centralized control system of air compressor machine frequency conversion |
| US20170261264A1 (en) * | 2017-05-25 | 2017-09-14 | Northeastern University | Fault diagnosis device based on common information and special information of running video information for electric-arc furnace and method thereof |
| CN107358332A (en) * | 2017-05-18 | 2017-11-17 | 国网浙江省电力公司 | A kind of dispatching of power netwoks runs lean evaluation method |
-
2018
- 2018-08-07 CN CN201810888070.3A patent/CN109164704B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101718270A (en) * | 2009-11-20 | 2010-06-02 | 上海应用技术学院 | Prediction and pressure regulation method for control system of air compressor |
| CN104635684A (en) * | 2014-12-25 | 2015-05-20 | 冶金自动化研究设计院 | Cluster optimization control system for air compressor |
| CN206190525U (en) * | 2016-11-04 | 2017-05-24 | 大同煤矿集团有限责任公司 | Energy -conserving centralized control system of air compressor machine frequency conversion |
| CN106651182A (en) * | 2016-12-25 | 2017-05-10 | 北京工业大学 | Rail passenger flow congestion risk evaluation method based on entropy weight |
| CN107358332A (en) * | 2017-05-18 | 2017-11-17 | 国网浙江省电力公司 | A kind of dispatching of power netwoks runs lean evaluation method |
| US20170261264A1 (en) * | 2017-05-25 | 2017-09-14 | Northeastern University | Fault diagnosis device based on common information and special information of running video information for electric-arc furnace and method thereof |
Non-Patent Citations (4)
| Title |
|---|
| JONG SUKKIM，ET AL.: "Optimal scheduling of combined heat and power plants using mixed-integer nonlinear programming", 《ENERGY》 * |
| THITIPONG JAMRUS,ET AL.: "Hybrid Particle Swarm Optimization Combined With Genetic Operators for Flexible Job-Shop Scheduling Under Uncertain Processing Time for Semiconductor Manufacturing", 《IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING》 * |
| WENBIN HU,ET AL.: "A swarm intelligent method for traffic light scheduling: application to real urban traffic networks", 《APPLIED INTELLIGENCE》 * |
| YICHENG ZHANG，ET AL.: "Distributed flight routing and scheduling for air traffic flow management", 《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020140564A1 (en) * | 2019-01-02 | 2020-07-09 | 大连理工大学 | Optimization scheduling decision method employing simulation technique for air compressor assembly |
| US11126765B2 (en) | 2019-01-02 | 2021-09-21 | Dalian University Of Technology | Method for optimal scheduling decision of air compressor group based on simulation technology |
| CN109886493A (en) * | 2019-02-26 | 2019-06-14 | 江南大学 | A kind of Logistics System Design method based on improvement multi-objective particle swarm algorithm |
| CN109886493B (en) * | 2019-02-26 | 2022-09-02 | 江南大学 | Logistics system design method based on improved multi-target particle swarm algorithm |
| CN111736466A (en) * | 2020-06-08 | 2020-10-02 | 武汉理工大学 | Optimal control method and system for quick load shedding system of semi-submersible platform |
| CN111736466B (en) * | 2020-06-08 | 2021-09-10 | 武汉理工大学 | Optimal control method and system for quick load shedding system of semi-submersible platform |
| CN111831963A (en) * | 2020-07-14 | 2020-10-27 | 中国南方电网有限责任公司 | Method for evaluating controllability of comprehensive energy service provider under power market background |
| CN113669242A (en) * | 2021-08-03 | 2021-11-19 | 新奥数能科技有限公司 | Power control method and device of air compressor system and computer equipment |
| CN113537644A (en) * | 2021-08-23 | 2021-10-22 | 中冶赛迪技术研究中心有限公司 | Multi-air compression station dynamic collaborative optimization regulation and control system and method |
| CN113537644B (en) * | 2021-08-23 | 2022-07-08 | 中冶赛迪技术研究中心有限公司 | Multi-air compression station dynamic collaborative optimization regulation and control system and method |
| CN113606111A (en) * | 2021-09-09 | 2021-11-05 | 广东鑫钻节能科技股份有限公司 | Energy-saving protection system based on air compression station and implementation method thereof |
| CN113606111B (en) * | 2021-09-09 | 2023-02-24 | 广东鑫钻节能科技股份有限公司 | Energy-saving protection system based on air compression station and implementation method thereof |
| CN115343967A (en) * | 2022-10-19 | 2022-11-15 | 蘑菇物联技术（深圳）有限公司 | Method, apparatus, and medium for simulation control of air compression station |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109164704B (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109164704A (en) | A kind of air compressor machine group Optimization Scheduling based on mixed model | |
| CN109783916B (en) | Air compressor group optimal scheduling decision method based on simulation technology | |
| CN109270842B (en) | Bayesian network-based regional heat supply model prediction control system and method | |
| Lei et al. | Multi-stage stochastic planning of regional integrated energy system based on scenario tree path optimization under long-term multiple uncertainties | |
| CN102289590B (en) | Method for estimating operating state of SF6 high-voltage circuit breaker and intelligent system | |
| CN107784397B (en) | Power grid material demand prediction system and prediction method thereof | |
| CN103591637B (en) | A kind of central heating secondary network runing adjustment method | |
| CN106651200A (en) | Electrical load management method and system for industrial enterprise aggregate user | |
| CN102705957A (en) | Method and system for predicting hourly cooling load of central air-conditioner in office building on line | |
| CN105529700A (en) | Non-invasive online load decomposition device | |
| CN105069521A (en) | Photovoltaic power plant output power prediction method based on weighted FCM clustering algorithm | |
| CN114692265A (en) | Zero-carbon building optimization design method based on deep reinforcement learning | |
| Moradzadeh et al. | A secure federated deep learning-based approach for heating load demand forecasting in building environment | |
| Song et al. | An indoor temperature prediction framework based on hierarchical attention gated recurrent unit model for energy efficient buildings | |
| CN108984830A (en) | A kind of building efficiency evaluation method and device based on FUZZY NETWORK analysis | |
| Huang et al. | Smart energy management system based on reconfigurable AI chip and electrical vehicles | |
| CN109946968B (en) | Distributed energy matching system and method in building | |
| Zhang et al. | The power big data-based energy analysis for intelligent community in smart grid | |
| CN112950001B (en) | Intelligent energy management and control system and method based on cloud edge closed-loop architecture | |
| Cen et al. | Multi-Task Learning of the PatchTCN-TST Model for Short-Term Multi-Load Energy Forecasting Considering Indoor Environments in a Smart Building | |
| CN104657789B (en) | A kind of operation operating method of steam power system | |
| Li et al. | Optimizing combination of aircraft maintenance tasks by adaptive genetic algorithm based on cluster search | |
| CN110458358A (en) | The sea of meter and production-process systems constraint is micro- can system optimization dispatching method | |
| CN112821456B (en) | Distributed source-storage-load matching method and device based on transfer learning | |
| CN113959071A (en) | Machine learning assistance-based operation control optimization method for centralized water chilling unit air conditioning 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 |