CN104102211B

CN104102211B - Scheduling method, server and system of combined heat and power generation system of fire coal thermal power plant

Info

Publication number: CN104102211B
Application number: CN201410367650.XA
Authority: CN
Inventors: 苏宏业; 张扬; 侯卫锋; 吴玉成
Original assignee: ZHEJIANG SUPCON SOFTWARE CO Ltd
Current assignee: Zhejiang SUPCON Software Co., Ltd.; Zhejiang University ZJU
Priority date: 2014-07-29
Filing date: 2014-07-29
Publication date: 2017-02-22
Anticipated expiration: 2034-07-29
Also published as: CN104102211A

Abstract

The application provides a scheduling method, a scheduling server and a scheduling system of a combined heat and power generation system of a fire coal thermal power plant. The scheduling method of the combined heat and power generation system of the fire coal thermal power plant is deployed on the scheduling application server in the scheduling system of the combined heat and power generation system of the fire coal thermal power plant, and includes: obtaining initial data needed by building working condition models of all devices in the combined heat and power generation system from a synthetic data integration platform, and building the working condition models of all the devices in the combined heat and power generation system according to the initial data; judging whether preset scheduling conditions are met or not, and if yes, confirming a scheduling policy of the combined heat and power generation system of the fire coal thermal power plant under a current load according to real-time power and heat load data output by the working condition models, and controllable variables and auxiliary variables, which influence outputting of the working condition models. According to the scheduling method, the scheduling server and the scheduling system of the combined heat and power generation system of the fire coal thermal power plant, scheduling problems are considered by integrating blending coal, a boiler, a turbine generator and a temperature and pressure reducer together, and a scheduling scheme can be globally optimized.

Description

A kind of dispatching method of Hazards in Power Plant co-generation unit, server and system

Technical field

The application is related to areas of information technology, particularly to a kind of dispatching method of Hazards in Power Plant co-generation unit, Server and system.

Background technology

With scientific and technological development, electricity is in people's life in occupation of particularly important position.General power plant only produces electricity Can be to customer power supply, commercial production and life heat (steam) are then individually supplied by other Industrial Boiler and heating boiler, this Plant the mode of production referred to as thermoelectricity and divide product.And then adopt heat-supply type unit in steam power plant, in addition to supply electric energy, simultaneously also sharp It was used as the extracted steam from turbine of work((generating electricity) or steam discharge produced and life institute calorific requirement to meet, this mode of production was referred to as Cogeneration of heat and power.

The structure chart of typical Hazards in Power Plant co-generation unit is as shown in figure 1, this co-generation unit can include： Several part such as coal yard, coal-burning boiler, steam turbine generator, temperature-decreased pressure reducer.Wherein, raw coal is delivered to after proportioning by coal yard Each boiler, boiler utilizes heat boiler feedwater during fired coal combustion, makes water be heated as the superheated steam of High Temperature High Pressure. Superheated steam enters back into the rotation generating of steam turbine generator drive motor and is supplied to electrical network, is done part from steam turbine generator simultaneously The steam (temperature and pressure reduces) of work(is extracted out and is supplied to steam pipe system.Due to downstream user for steam grade requirement not With steam turbine generator may have the outer of several grades to supply to draw gas.When steam turbine generator draws gas deficiency, temperature-decreased pressure reducer is direct Attract steam from upper level steam main, make steam become the relatively low steam of temperature and pressure after desuperheat, decompression and be supplied to next Level steam pipe system.

From figure 1 it appears that how the load that co-generation unit operationally, has electrically and thermally (steam) divides Join the problem with raw coal how proportioning.For steam turbine generator, under meeting electrical network and the total workload demand of steam pipe system, respectively The electric load (i.e. generated energy) of steam turbine generator and thermic load (i.e. the amount of drawing gas) can have multiple combination, and the steamer of every kind of combination is sent out Total throttle flow of motor may be different.And for temperature-decreased pressure reducer, after high temperature and high pressure steam enters temperature-decreased pressure reducer, energy has Very big loss, just opens temperature-decreased pressure reducer for general in the case that steam turbine generator draws gas and cannot meet demand.If opened Temperature-decreased pressure reducer, the thermic load (outlet vapor amount) of temperature-decreased pressure reducer can have multiple combination, the temperature-decreased pressure reducer of every kind of combination Total throttle flow be likely to difference.And for boiler, meeting downstream steam turbine generator and temperature-decreased pressure reducer vapour aggregate demand In the case of, the steam production of each boiler also has multiple combination, and the coal-fired total amount that every kind of combination is consumed is likely to difference.And it is right For coal yard, if the raw coal price of each coal yard, ature of coal are different, meeting steam coal may have multiple Coal Blending Schemes, different Use coal cost under scheme is again possibly different.

It can be seen that, co-generation unit is that a coal yard, boiler, steam turbine generator, temperature-decreased pressure reducer and downstream user are mutual The complication system of association.In order to meet the requirement of customer charge, dispatcher needs the links producing are scheduling：Coal Need how proportioning raw coal, how much coal-fired to boiler conveying；Every boiler should produce how much steam；Every steam turbine generator should This generates how many electricity, extracts how much steam out；Temperature-decreased pressure reducer, the need of unlatching, if opened, needs opened which platform, flow is such as How etc., this just constitutes a scheduling scheme.And under meeting electric, hot (steam) burden requirement of downstream user, it is understood that there may be Multiple scheduling schemes are selective, but which kind of scheme can energy saving be the problem needing dispatcher to consider.

Inventor finds that prior art has problems with：Dispatcher, when selection scheduling scheme, will not Coal blending in co-generation unit, boiler, steam turbine generator and temperature-decreased pressure reducer combine consideration scheduling problem, often only It is scheduling project study for the part (as only for Turbo-generator Set) in co-generation unit, this may make The energy consumption obtaining thermoelectricity production is higher, and, purchase coal high expensive.

Content of the invention

Based on problem present in the current cogeneration of heat and power that inventor finds, this application provides a kind of Hazards in Power Plant is warm The dispatching method of cogeneration system, mainly adjusts to co-generation unit by rule of thumb in order to solve dispatcher in prior art Degree, the high energy consumption that the thermoelectricity leading to produces, and the problem of high cost, also production can be made on the premise of meeting production requirement to become This minimum.And, the coal blending in co-generation unit, boiler, steam turbine generator and temperature-decreased pressure reducer are incorporated into one by the application Rise and consider scheduling problem, scheduling scheme can be made to accomplish " global optimum ".Further, the equipment in co-generation unit runs one In the case that the section time deviates its original operating mode, the application also can make up the impact that this difference is brought from scheduling scheme.

Present invention also provides the scheduling application server of Hazards in Power Plant co-generation unit and system, on ensureing State method realization in practice and application.

In order to solve the above problems, this application discloses a kind of dispatching method of Hazards in Power Plant co-generation unit, institute The method of stating is deployed on the scheduling application server in the scheduling system of Hazards in Power Plant co-generation unit, described scheduling system Also include：Synthetic data integrated platform server；Methods described includes：

Obtain from synthetic data integrated platform server and set up needed for the condition model of individual device co-generation unit Primary data, described primary data includes：Steam turbine generator throttle flow, generated energy and the amount of drawing gas, the steam production of boiler, desuperheat Decompressor imports and exports real time data and historical data, the real time data of vapor (steam) temperature and the historical data of quantity of steam, steam pressure Real time data and historical data, raw coal physico-chemical analysis data, and, the price of raw coal；

Set up the condition model of individual device in described co-generation unit according to described primary data；

Judge whether to meet the schedulable condition pre-setting, if it is, exporting electrically and thermally according to described condition model Real-time load data, and the controlled variable of impact described condition model output and auxiliary variable, determine combustion under current loads The scheduling strategy of coal steam power plant co-generation unit.

Preferably, described judge whether to meet the schedulable condition pre-setting, including：

The change judging the total load electrically and thermally of described Hazards in Power Plant co-generation unit within preset time period is No exceed preset first threshold value, or, judge in described Hazards in Power Plant co-generation unit scheduling fire coal value and the reality of prediction Whether the difference of the coal-fired value in border is more than default Second Threshold.

Preferably, the described real-time load data electrically and thermally according to the output of described condition model, and affect described work The controlled variable of condition model output and auxiliary variable, determine the scheduling plan of Hazards in Power Plant co-generation unit under current loads Omit, including：

By the real-time load data electrically and thermally of described condition model output, and from synthetic data integrated platform server The controlled variable of impact described condition model output obtaining and the value of auxiliary variable, as the Optimal Operation Model pre-building Input；Described Optimal Operation Model includes object function and constraints, and described object function is：

MinC=∑ C_F+∑PE_n·G_n,in,

Wherein, C is the total production cost of described co-generation unit, C_FFor various raw coal costs, G_n,inFor n-th grade of desuperheat The flow that decompressor is opened, PE_nPenalty value after opening for n-th grade of temperature-decreased pressure reducer；Described constraints includes：Material balance Constraints, power balance constraints, steam turbine generator constraints, boiler constraints, temperature-decreased pressure reducer constraints With coal blending constraints；

Calculate the knot of described Optimal Operation Model under current load electrically and thermally using MILP algorithm Really, to obtain scheduling strategy under current load electrically and thermally for the described Hazards in Power Plant co-generation unit.

Preferably, also include：

The scheduling strategy of described Hazards in Power Plant co-generation unit is stored.

Preferably, also include：

In response to user's request, the described scheduling strategy of storage is showed user, so that user is according to described scheduling plan Slightly trigger the Optimized Operation of described Hazards in Power Plant co-generation unit.

Preferably, also include：

Primary data needed for described condition model is carried out with pretreatment, described preprocess method includes but is not limited to：Office Exterior point detection algorithm, linear smoothing algorithm or standardized algorithm.

Preferably, also include：

Judge whether the number of times that the output valve of described condition model is more than default 3rd threshold value with the difference of actual value exceedes Default difference frequency threshold value, if it is, determine the regulated quantity of the output valve for compensating condition model according to this difference.

Preferably, also include：

Record input and the output of described condition model, and the sample that the equipment operating condition of reality is constituted, and judge Record sample number whether exceed preset recording sample threshold, if it is, using described record the input of condition model and Condition model described in output re -training.

This application discloses a kind of scheduling application server of Hazards in Power Plant co-generation unit, including：

Obtain initial data unit, set up in co-generation unit respectively for obtaining from synthetic data integrated platform server Primary data needed for the condition model of platform equipment, described primary data includes：Steam turbine generator throttle flow, generated energy and draw gas Amount, the steam production of boiler, temperature-decreased pressure reducer imports and exports real time data and the historical data of quantity of steam, the real time data of vapor (steam) temperature And historical data, the real time data of steam pressure and historical data, raw coal physico-chemical analysis data, and, the price of raw coal；

Set up condition model unit, for setting up individual device in described co-generation unit according to described primary data Condition model；

First judging unit, for judging whether to meet the schedulable condition pre-setting；

Determine scheduling strategy unit, for described judging unit result be in the case of, according to described operating mode mould The real-time load data electrically and thermally of type output, and affect controlled variable and the auxiliary variable of described condition model output, really The scheduling strategy of Hazards in Power Plant co-generation unit under settled preload.

Preferably, described first judging unit includes：

First judge module, for judging that described Hazards in Power Plant co-generation unit is electrically and thermally within preset time period The change of total load whether exceed preset first threshold value；

Second judge module, for judging scheduling fire coal value and the reality of prediction in described Hazards in Power Plant co-generation unit Whether the difference of the coal-fired value in border is more than default Second Threshold.

Preferably, described determination scheduling strategy unit includes：

Determine input module, for the real-time load data electrically and thermally exporting described condition model, and from synthesis The controlled variable of impact described condition model output and the value of auxiliary variable that data integration platform server obtains, as in advance The input of the Optimal Operation Model set up；Described Optimal Operation Model includes object function and constraints, described object function For：

MinC=∑ C_F+∑PE_n·G_n,in,

Computing module, for calculating described optimization under current load electrically and thermally using MILP algorithm The result of scheduling model, to obtain scheduling plan under current load electrically and thermally for the described Hazards in Power Plant co-generation unit Slightly.

Preferably, also include：

Memory element, for being stored the scheduling strategy of described Hazards in Power Plant co-generation unit.

Preferably, also include：

Show scheduling strategy unit, in response to user's request, the described scheduling strategy of storage being showed user, with Just user triggers the Optimized Operation of described Hazards in Power Plant co-generation unit according to described scheduling strategy.

Preferably, also include：

Data pre-processing unit, for carrying out pretreatment, described pretreatment to the primary data needed for described condition model Method includes but is not limited to：Outlier detection algorithm, linear smoothing algorithm or standardized algorithm.

Preferably, also include：

Judging unit, presets the 3rd threshold value for judging that the output valve of described condition model is more than with the difference of actual value Whether number of times exceedes default difference frequency threshold value；

Determine regulated quantity unit, for described second judging unit result be in the case of, true according to this difference The regulated quantity of the fixed output valve for compensating condition model.

Preferably, also include：

Recording unit, for recording input and the output of described condition model；

3rd judging unit, whether the number of times for judging to record exceedes preset recording frequency threshold value；

Training unit, for described 3rd judging unit result be in the case of, using the operating mode of described record The input of model and condition model described in output re -training.

The embodiment of the present application additionally provides a kind of scheduling system of Hazards in Power Plant co-generation unit, and this system includes： Scheduling application server and synthetic data integrated platform server, wherein, described scheduling application server includes：

Compared with prior art, the application includes advantages below：

In the embodiment of the present application using the historical data obtaining from scheduling system, real-time equipment operating data, and Set up each equipment working condition model and global optimization scheduling model in co-generation unit based on these data, to having coal-fired thermal power Factory's co-generation unit implements global optimization scheduling.Therefore, have been directed to the concrete feature of Hazards in Power Plant co-generation unit, from Fundamentally solve the problems, such as the deficiency that enterprise dispatcher is scheduling by rule of thumb, thus improving Hazards in Power Plant cogeneration of heat and power system The accuracy of the scheduling of system, improves the economy of co-generation unit scheduling, saves coal consumption, reduces purchase coal cost, thus can To reach the purpose of energy-saving and emission-reduction, cost efficiency.

Brief description

For the technical scheme being illustrated more clearly that in the embodiment of the present application, will make to required in embodiment description below Accompanying drawing be briefly described it should be apparent that, drawings in the following description are only some embodiments of the present application, for For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings Accompanying drawing.

Fig. 1 is the system configuration example figure during scheduling carrying out Hazards in Power Plant co-generation unit in prior art；

Fig. 2 is a kind of flow chart of the dispatching method embodiment of Hazards in Power Plant co-generation unit of the application；

Fig. 3 is a kind of application scenario diagram of the dispatching method embodiment of Hazards in Power Plant co-generation unit of the application；

Fig. 4 be the application a kind of dispatching method embodiment of Hazards in Power Plant co-generation unit in condition model is entered The flow chart of row on-line correction；

Fig. 5 is a kind of structured flowchart of the scheduling application server of Hazards in Power Plant co-generation unit of the application；

Fig. 6 be the application a kind of Hazards in Power Plant co-generation unit in carry out the structured flowchart of on-line correction.

Accompanying drawing described herein is only some examples.In the case of without departing from the application spirit, figure described here Can there are different changes.All above-mentioned changes are considered as the part of claimed the application.

Specific embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is carried out clear, complete Site preparation describes it is clear that described embodiment is only some embodiments of the present application, rather than whole embodiments.It is based on Embodiment in the application, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work Embodiment, broadly falls into the scope of the application protection.

With reference to Fig. 2, show a kind of flow process of the dispatching method embodiment of the application Hazards in Power Plant co-generation unit Figure, methods described is deployed on the scheduling application server in the scheduling system of Hazards in Power Plant co-generation unit, described tune Degree system can also include：Synthetic data integrated platform server；The present embodiment may comprise steps of：

Step 201：Scheduling application server obtains from synthetic data integrated platform server to be set up co-generation unit Primary data needed for the condition model of individual device.

In the embodiment of the present application, scheduling application server is firstly the need of the work setting up individual device in co-generation unit Condition model, setting up the primary data needed for condition model can obtain from synthetic data integrated platform server, this primary data Specifically can include：Steam turbine generator throttle flow, generated energy and the amount of drawing gas, the steam production of boiler, temperature-decreased pressure reducer is imported and exported and is steamed The real time data of vapour amount and historical data；Run closely related auxiliary variable, for example, the real time data of vapor (steam) temperature with equipment And historical data, the real time data of steam pressure and historical data；The data closely related with raw coal consumption, processing cost, example As raw coal physico-chemical analysis data, and, the price of raw coal.Wherein, real time data and historical data can be from DCS, real time datas Obtain in storehouse, and be stored in synthetic data integrated platform server and called with being scheduled for application server；Production-Plan and scheduling Data, laboratory analysis of data, raw coal price data or from Third party system, such as the database server in MES, LIMS obtains, Or inputted by client by user by way of being manually entered, and be stored in synthetic data integrated platform server for adjusting Degree application server calls.

With reference to shown in Fig. 3, it is the instance graph of the global optimization scheduling system of the co-generation unit of the application, wherein wraps Include：Measuring instrumentss at the scene, sensor 307, DCS306, synthetic data integrated platform server 305, scheduling application clothes are installed The hardware devices such as business device 302, client 301, fire wall, antivirus server 303, Third party system database server 304 And link the computer network composition of each computer equipment, controller and sensor.

In figure 3, the measuring instrumentss at scene can be used for data necessary to on-line checking scheduling system, and sensor will be The data that line detects is sent to DCS system, realizes data acquisition and control.Data acquisition is as follows：Co-generation unit Scene be dispersed with several measure individual device service datas measuring instrumentss, by these measuring instrumentss according to each self-metering not The pretreatment such as filter, buffer, amplifying with index signal, then by signal after Phototube Coupling, sending into the corresponding control of DCS In system point label.

In figure 3, client can propose to read or the related letter of write to scheduling application server according to user's request The requirement of breath, and according to user's request, scheduling scheme is shown on the client.Fire wall antivirus server main It is comprehensive that task is that monitoring thermoelectricity global optimization dispatches the directly related client of system, scheduling system application server, scheduling system Close the working environment of data integration platform server etc..

In figure 3, scheduling application server is the core component entirely dispatching system, is also the application embodiment of the method Executive agent, it mainly runs co-generation unit equipment modeling method (corresponding step 202), co-generation unit global optimization The modules such as scheduling modeling method (corresponding step 204).Scheduling application server needs to call synthetic data integrated platform server The data of middle storage, and will be real to co-generation unit equipment modeling method, co-generation unit global optimization scheduling modeling method The model result obtaining after applying writes the data base of synthetic data integrated platform server.Calculating simultaneously for some models must The model parameter wanted, scheduling application server meeting automatic decision is the need of adjustment, and passes through application service when needing adjustment Device is updated to the related data of synthetic data integrated platform server processing automatically.

In figure 3, synthetic data integrated platform server is based on professional real-time data base and RDBM Relational Data Base Management system System, by the measurement data of the production scene implemented required for the scheduling of co-generation unit global optimization, cost of material data, chemical examination Analytical data, Production-Plan and scheduling data etc. are stored in data base.Wherein, cost of material data, laboratory analysis of data, life Produce planning and scheduling data or come from Third party system database server, or write synthesis by way of manual entry In the data base of data integration platform server.

In various embodiments, before step 201, can also include：

Step 200：Primary data needed for condition model is carried out with pretreatment, described preprocess method includes but do not limit In：Outlier detection algorithm, linear smoothing algorithm or standardized algorithm.

It is understood that the accuracy of the data in order to ensure to be stored in synthetic data integrated platform server, step These data in rapid 201 are required to by pretreatment, to ensure correctness and the reliability of gathered data, it is to avoid because losing Lead to gathered data to occur abnormal, wherein error refers to due to control, environment, the unstable and artificial mistake of measuring instrument by mistake The error caused by factor such as mistake.Conventional data preprocessing method has a lot, such as outlier detection, linear smoothing and standardization Deng.These are all existing preprocess methods, will not be described here.

After step 201, the execution step that continues 202：Set up in described co-generation unit according to described primary data The condition model of individual device.

In this step, then can be according to the initial number getting from synthetic data integrated platform server in step 201 According to setting up the condition model of individual device in co-generation unit.

Equipment working condition model in the present embodiment is as follows：

Y=f (X, X^*)

Wherein, Y is the output vector of current device condition model；X is the controlled variable of impact equipment working condition model output, It is the variable that artificially can control, such as supplies consumption amount；X* is the auxiliary variable of impact equipment working condition model output, but artificially not Controlled, such as ambient temperature etc.；F () is selected model structure.

When co-generation unit equipment working condition model is set up, obtain equipment input with the variable being closely related with equipment Functional relationship and output between.When obtaining these functional relations it is also desirable to introduce other auxiliary variables, to improve equipment The precision of condition model.When setting up co-generation unit equipment working condition model, mainly model structure f in formula (1) The selection of (), and, the selection of model controlled variable X and auxiliary variable X*.Wherein, co-generation unit equipment working condition model Model structure storehouse can be by linear regression model (LRM), sectional linear fitting model, principal component model, offset minimum binary mould Type, artificial nerve network model or fuzzy neural network model etc. form, but not limited to this.Above-mentioned model structure is all current Highly developed model, the application process of these models is known to those skilled in the art, repeats no more in this.

In the specific implementation, the model structure select permeability of co-generation unit equipment working condition model can be changed into 0-1 Planning problem, and solved using genetic algorithm, select the best model structure of simulation effect from model structure storehouse.Its In, simulation effect preferably refers to equipment working condition data and this time period being gathered in following a period of time that model provides Difference between the service data of physical device is minimum.Above-mentioned one-zero programming method is a kind of integer programming of specific form.This The decision variable only value 0 or 1 of planning, 0-1 variable can with quantification Push And Release is described, take and abandon, being and not being etc. existing As the constraints of logical relation, ordering relation and mutual exclusion between the discrete variable that reflected, the integer of every bounded variable Planning can be converted into one-zero programming to process.Therefore, it can turn the model structure select permeability of thermal power unit condition model It is melted into one-zero programming problem to be processed.Wherein, because genetic algorithm algorithm known to those skilled in the art, in this Also repeat no more.

Similarly, the select permeability of controlled variable and auxiliary variable is also translated into one-zero programming problem, and adopts genetic algorithm Solved, select best controlled variable X of simulation effect and auxiliary to become from initial controlled variable and auxiliary variable vector The vectorial X* of amount.Similarly, simulation effect preferably refers in following a period of time that model provides equipment operating data and is gathered The physical device service data of this time period between difference minimum.

In the other embodiment of the application, can also be after step 202 establishes condition model, for operating mode Model is detected and on suitable opportunity, it is adjusted, then with reference to shown in Fig. 4, after step 202, can also wrap Include：

Step 401：Judge that the output valve of described condition model and the difference of actual value are more than the number of times of default 3rd threshold value Whether exceed default difference frequency threshold value, if it is, entering step 402.

In the present embodiment, co-generation unit is a coal yard, boiler, steam turbine generator, temperature-decreased pressure reducer and downstream User is mutually related complication system, is not generally possible to set up accurate equipment working condition model and becomes accurately reflecting industrial process Change, and need very big dependence real-time process model on-line correction is followed the tracks of with change in process, with accurately anti- Reflect the change of process major trend.Therefore, to the on-line correction of co-generation unit equipment working condition model it is ensured that it is long-term Reliablely and stablely run.In actual applications, can be by being in addition added independently of an on-line correction in equipment working condition model Module realizing the flow process of Fig. 4, with the output of charge of overseeing equipment working condition model with the operating mode actual value that feeds back.

Need in this step to compare the output valve of equipment working condition model and the difference of actual value, when difference between the two During more than predetermined threshold, then record the production status in this difference and corresponding section, and if this species diversity consecutive numbers time produces And corresponding production status are when being in steady operational status all the time, number of times has been over default difference frequency threshold value, then just Need subsequently adjusted.Additionally, without exceeding default difference frequency threshold value, being then left intact.

Step 402：Determine the regulated quantity of the output valve for compensating condition model according to this difference.

Provide the regulated quantity of an output valve for compensating condition model in this step by calculating deviation, and will This regulated quantity is added on the output node layer of equipment working condition model, thus compensating the output of equipment working condition model, so that itself and reality The equipment operating condition data on border is close.

Step 403：Record input and the output of described condition model, and the sample that the equipment operating condition of reality is constituted This.

In the present embodiment, joined by thermoelectricity in addition it is also necessary to record when finding that difference between the two is more than predetermined threshold Produce input (i.e. various variables) and the output of system equipment condition model, and the sample that the equipment operating condition of reality is constituted.

Step 404：Judge whether the sample number recording exceedes preset recording sample threshold, if it is, entering step 405.

Likewise, when the sample number being recorded reaches preset recording sample threshold, then execution step 405.If not yet Have and reach, then do not make any process.

Step 405：Using the input of the condition model of described record and condition model described in output re -training.

In this step using the sample re -training equipment working condition model being recorded, the more parameter of new model f () so that The precision of prediction of model output result reaches pre-provisioning request.Wherein, described predetermined threshold and predetermined quantity all can be according to reality Required precision of prediction, to set, is not construed as limiting in this.

Because in actual motion, the equipment of co-generation unit is after running a period of time or normal after maintenance, transformation Its original operating mode often can be deviateed, dispatcher is sometimes difficult to discover this difference, if dispatcher practises according to original scheduling It is used to be scheduling, often result in the unreasonable of scheduling.And the flow process passing through Fig. 4 carries out on-line correction to it is possible to ensure to set The accuracy of standby condition model, thus also can guarantee that the accuracy of follow-up Optimized Operation.

It is then returned to Fig. 2, enter step 203：Judge whether to meet the schedulable condition pre-setting, if it is, entering Step 204.

In this step, need to judge whether current co-generation unit has met the scheduling bar pre-setting Part, specifically, can be by judging the total load electrically and thermally of Hazards in Power Plant co-generation unit within preset time period Whether change exceedes preset first threshold value, or, judge in Hazards in Power Plant co-generation unit the scheduling fire coal value of prediction with Whether the difference of actual coal-fired value is more than default Second Threshold.Wherein, if the total load electrically and thermally of co-generation unit Whether change exceedes preset first threshold value, then the total load of explanation co-generation unit has the larger adjustment of ratio, and it is right now to need Co-generation unit is scheduling.And if the difference of the coal-fired value of the scheduling of prediction and actual coal-fired value is also greater than default second threshold Value, then illustrate in the case of current total load electrically and thermally, and the also larger saving space of ratio is therefore also required to thermoelectricity Co-generation system is scheduling.

Step 204：According to the real-time load data electrically and thermally of described condition model output, and affect described operating mode mould The controlled variable of type output and auxiliary variable, determine the scheduling strategy of Hazards in Power Plant co-generation unit under current loads.

When determining scheduling strategy in this step, can be by the co-generation unit global optimization scheduling model pre-building To realize, the core concept of this step is with the minimum target of the total production cost of co-generation unit, by optimizing boiler, vapour Turbine generator and temperature-decreased pressure reducer load, optimization raw coal proportion strategy, in the case of ensureing that produce load requires, reduce desuperheat Decompressor opening times and flow, the coal-fired consumption of minimizing, minimizing purchase coal cost, such that it is able to realize energy-saving and emission-reduction, drop this increasing The target of effect.

In the specific implementation, this step specifically can include：

Step A1：By the real-time load data electrically and thermally of described condition model output and integrated flat from synthetic data The controlled variable of impact described condition model output and the value of auxiliary variable that platform server obtains, as the optimization pre-building The input of scheduling model.

In the present embodiment, Optimal Operation Model includes object function and constraints, wherein object function such as formula (1) Shown：

MinC=∑ C_F+∑PE_n·G_n,in(1)

Wherein, C is the total production cost of described co-generation unit, C_FFor various raw coal costs, G_n,inFor n-th grade of desuperheat The flow that decompressor is opened, PE_nPenalty value after opening for n-th grade of temperature-decreased pressure reducer.On the right of equation, Section 1 guarantees that fire coal disappears Consumption, processing cost are minimum, and Section 2 guarantees that the unlatching of temperature-decreased pressure reducer is as far as possible few.Whole object function represents the mesh of Optimized Operation Be in order that total production cost is minimum when co-generation unit runs.

Wherein, constraints specifically can include：Material balance constraints, power balance constraints, turbine generator Machine constraints, boiler constraints, temperature-decreased pressure reducer constraints and coal blending constraints.

Wherein, thermoelectricity dispatches the material that is related to of system raw coal and steam, therefore material balance constraint as formula (2), (3) and shown in (4)：

∑F_R=∑ F_B+∑F_L(2)

∑S_B=∑ D_in+∑G_n,in+∑S_L(3)

∑V_n,T+∑V_n,R=∑ V_n,U+∑V_n,L(4)

In formula (2)～(4), F_RRepresent former the consumption of coal；F_BRepresent boiler fired coal consumption；F_LFor raw coal defeated Loss during sending.S_BRepresent boiler duty；D_inRepresent steam turbine generator throttle flow, G_n,inRepresent n-th grade of pressure and temperature reducing Device throttle flow；S_LRepresent loss in course of conveying for the steam.V_n,TRepresent n-th grade of steam turbine generator draw gas (supplying outward) amount；V_n,R Represent that n-th grade of temperature-decreased pressure reducer exports quantity of steam；V_n,URepresent n-th grade of steam demand amount of user；V_n,LRepresent n-th grade of steam defeated Loss during sending.

Wherein, the generated energy of steam turbine generator will meet the demand of downstream user, therefore power balance constraint such as formula (5) Shown：

∑ E=∑ E_U+∑E_L(5)

In formula (5), E represents steam turbine generator generated energy, E_URepresent user power utilization demand, E_LRepresent electricity transmission process In power consumption.

Wherein, the confidential requirement meeting steam turbine generator equipment working condition model structure and restriction of production of turbine generator, therefore Shown in equipment (steam turbine generator) constraints such as formula (6), (7), (8) and (9)：

D_in=f (E, D_n,out,P_in,P_n,out,T_in,T_n,out,X^*) (6)

E^min·y_T≤E·y_T≤E^max·y_T(8)

In formula (6)～(9), D_inFor steam turbine generator throttle flow, E is the generated energy of steam turbine generator, D_{N, out}For vapour (supplying outward) amount of drawing gas of n-th grade of turbine generator, P_inFor steam turbine generator initial steam pressure, P_{N, out}For n-th grade of (supplying outward) pressure of drawing gas Power, T_inFor steam turbine generator throttle (steam) temperature, T_{N, out}For n-th grade of (supplying outward) temperature of drawing gas, X^*For other auxiliary variables.It is respectively lower limit and the higher limit of steam turbine generator throttle flow；E^min、E^maxIt is respectively steam turbine generator generated energy Lower limit and higher limit；Be respectively n-th grade of steam turbine generator draw gas lower limit and higher limit；y_TBecome for 0-1 Amount, is the switching on and shutting down control variable as each steam turbine generator, if steam turbine generator start, y_TValue is 1, shutdown then y_T For 0.

Wherein, boiler will meet the requirement of boiler plant condition model structure and restriction of production, therefore equipment (boiler) about Shown in bundle condition such as formula (10) and (11)：

F_B=f (S_B,P,T,P₀,T₀,W_B,Pw,Tw,η,Q,X^*) (10)

In formula (10) and (11), F_BFor boiler fired coal consumption, S_BFor the evaporation capacity of boiler, P steams pressure for boiler Power, T steams temperature for boiler, P₀For boiler feedwater pressure, T₀For boiler feed temperature, W_BFor boiler blow-off discharge, Pw is pot Fire grate sewage pressure, Tw is boiler blow-off coolant-temperature gage, and η is boiler efficiency, and Q is the low heat valve of fuel, X^*For other auxiliary Variable.It is respectively lower limit and the higher limit of boiler duty.y_BFor 0-1 variable, it is the switch as each boiler Machine control variable, if boiler start, y_BValue is 1, shutdown then y_BFor 0.

Wherein, temperature-decreased pressure reducer will meet the requirement of desuperheat prelum equipment working condition model structure and restriction of production, therefore Shown in temperature-decreased pressure reducer constraints such as formula (12) and (13)：

G_n,out=f (G_n,in,P_n,in,P_n,out,T_n,in,T_n,out,Pw_n,Tw_n,X^*) (12)

In formula (12) and (13), G_n,outFor n-th grade of temperature-decreased pressure reducer outlet vapor amount, G_n,inSubtract for n-th grade of desuperheat Depressor inlet steam amount, P_n,inFor n-th grade of temperature-decreased pressure reducer inlet steam pressure, P_n,outSteam for n-th grade of temperature-decreased pressure reducer outlet Steam pressure, T_n,inFor n-th grade of temperature-decreased pressure reducer inlet steam temperature, T_n,outFor n-th grade of temperature-decreased pressure reducer outlet steam temperature, Pw For the pressure of desuperheating water of n-th grade of temperature-decreased pressure reducer, Tw is the desuperheat coolant-temperature gage of n-th grade of temperature-decreased pressure reducer, X^*Become for other auxiliary Amount.It is respectively lower limit and the higher limit of n-th grade of temperature-decreased pressure reducer inlet steam.y_G,nFor 0-1 variable, it is conduct The switching on and shutting down control variable of n-th grade of temperature-decreased pressure reducer, if n-th grade of temperature-decreased pressure reducer start, y_G,nValue is 1, shutdown then y_G,n For 0.

Wherein, due to the requirement of safety in production and process conditions, the physical and chemical index of the fire coal to use for the boiler is (as low level is sent out Heat, moisture, ash grade) have certain limitations, therefore shown in the constraints of coal blending such as formula (14), (15) and (16)：

Zm_i=∑ f_i(Z_i) (14)

C_F=f (F_R,Z,L,X^*) (16)

In formula (14)～(16), Zm_iRepresent i-th physical and chemical index of the boiler fired coal after raw coal proportion, Z_iFor former I-th physical and chemical index of coal, ∑ f_i() is the model structure of i-th coal-fired physical and chemical index of different raw coal proportion gained,It is respectively lower limit and the higher limit of coal-fired i-th physical and chemical index.C_FFor raw coal buying, processing cost；F_RFor Various former the consumption of coals；Z is the physico-chemical analysis data of raw coal；L is the unit price of raw coal；X^*For the buying of other influences raw coal, place The auxiliary variable of reason cost；F () is the model structure obtaining selected by raw coal buying, processing cost.

It is understood that in formula (1)～(16), having a lot of constants, damage including ambient temperature, penalty value, material Consumption, power consumption, various upper lower limit value, the unit price of raw coal, physico-chemical analysis data of raw coal etc., are required to according to each application enterprise Practical situation manually to determine, inputted by client by user by way of being manually entered, and it be comprehensive to be stored in scheduling system Close data integration platform server to call to be scheduled for application server.

It is understood that in formula (1)～(16), defeated except above-mentioned constant and co-generation unit equipment working condition model Outside going out, the data of other specification can obtain from DCS, real-time data base, or from Third party system, in MES, LIMS Database server obtain, and be stored in scheduling system synthesis data integration platform server called with being scheduled for application server.

After step A1, then execution step A2：Calculated currently electrically and thermally using MILP algorithm Load under described Optimal Operation Model result, with obtain described Hazards in Power Plant co-generation unit current electrically and thermally Scheduling strategy under load.

It should be noted that be made up of co-generation unit target function type (1) and constraint equation (2)～(16) overall The most basic scheduling model of Optimized Operation, scheduling model, due to there is 0-1 variable, needs using MILP method Solved, MILP algorithm is relatively common, repeats no more in this.Especially, due to different fire coals There are different requirements in steam power plant to the scheduling of co-generation unit, so the foundation of this Optimal Operation Model is not limited to formula (1)～(16), this global optimization scheduling system can be extended to model according to the concrete condition of each steam power plant, to adapt to not Demand with scene.

It can be seen that, the embodiment of the present application by the way of multi-layer framework, by essential core model and algorithm in the way of assembly It is deployed in the scheduling application server in intermediate layer (executive agent is scheduling application server), thus providing a kind of general , expansible transplantable, disclosure satisfy that the requirement of different Hazards in Power Plant co-generation unit global optimization scheduling.With existing Technology is compared, and the application sets up co-generation unit global optimization scheduling model based on equipment working condition model, to meet electricity, heat Premised on (steam) total load, the minimum target of the total production cost of co-generation unit, automatically provide the scheduling scheme of optimization, Achieve and coal blending, boiler, steam turbine generator, temperature-decreased pressure reducer are combined the scheduling scheme making " global optimization ", from And improve the economy that co-generation unit is dispatched, realize the target of energy-saving and emission-reduction, cost efficiency, and keep away to a certain extent The artificial subjectivity of dispatcher and the problem brought because lacking experience are exempted from.

In various embodiments, after step 204, can also include：

Step 205：The scheduling strategy of described Hazards in Power Plant co-generation unit is stored.

After obtaining scheduling strategy, this scheduling strategy can also be stored in synthetic data integrated platform server.

Step 206：In response to user's request, the described scheduling strategy of storage is showed user, so that user is according to institute State the Optimized Operation that scheduling strategy triggers described Hazards in Power Plant co-generation unit.

If receiving the request that user is triggered by client, can adjust from synthetic data integrated platform server Take out scheduling strategy and show user on the client, and the scheduling strategy triggering coal fuel heating that user then can see according to it The Optimized Operation of power plant's co-generation unit.

For aforesaid embodiment of the method, in order to be briefly described, therefore it is all expressed as a series of combination of actions, but Those skilled in the art should know, the application is not limited by described sequence of movement, because according to the application, some Step can be carried out using other orders or simultaneously.Secondly, those skilled in the art also should know, described in the specification Embodiment belong to preferred embodiment, necessary to involved action and module not necessarily the application.

Corresponding with a kind of dispatching method embodiment of above-mentioned the application Hazards in Power Plant co-generation unit, referring to Fig. 5, Present invention also provides a kind of scheduling application server embodiment of Hazards in Power Plant co-generation unit, in the present embodiment, This server can include：

Obtain initial data unit 501, set up co-generation unit for obtaining from synthetic data integrated platform server Primary data needed for the condition model of middle individual device, described primary data includes：Steam turbine generator throttle flow, generated energy and The amount of drawing gas, the steam production of boiler, temperature-decreased pressure reducer imports and exports the real time data of quantity of steam and historical data, vapor (steam) temperature real-time Data and historical data, the real time data of steam pressure and historical data, raw coal physico-chemical analysis data, and, the price of raw coal.

Set up condition model unit 502, set for setting up in described co-generation unit each according to described primary data Standby condition model.

Wherein, described device can also carry out on-line correction to the condition model set up, then with reference to shown in Fig. 6, this device Can also include：

Judging unit 601, presets the 3rd threshold for judging that the output valve of described condition model is more than with the difference of actual value Whether the number of times of value exceedes default difference frequency threshold value.

Determine regulated quantity unit 602, for described second judging unit result be in the case of, according to this difference Determine the regulated quantity of the output valve for compensating condition model.

Recording unit 603, for recording input and the output of described condition model, and the equipment operating condition of reality The sample constituting.

3rd judging unit 604, whether the sample number for judging to record exceedes preset recording sample threshold.

Training unit 605, for described 3rd judging unit result be in the case of, using the work of described record The input of condition model and condition model described in output re -training.

First judging unit 503, for judging whether to meet the schedulable condition pre-setting.

Wherein, described first judging unit specifically can include：First judge module, for judging within preset time period Whether the change of the total load electrically and thermally of described Hazards in Power Plant co-generation unit exceedes preset first threshold value；Second judgement Module, the difference of the coal-fired value of the scheduling for judging in described Hazards in Power Plant co-generation unit prediction and actual coal-fired value is No more than default Second Threshold.

Determine scheduling strategy unit 504, for described judging unit result be in the case of, according to described operating mode The real-time load data electrically and thermally of model output, and affect controlled variable and the auxiliary variable of described condition model output, Determine the scheduling strategy of Hazards in Power Plant co-generation unit under current loads.

Wherein, described determination scheduling strategy unit 504 specifically can include：

Determine input module, for the real-time load data electrically and thermally exporting described condition model, and from synthesis The controlled variable of impact described condition model output and the value of auxiliary variable that data integration platform server obtains, as in advance The input of the Optimal Operation Model set up；Described Optimal Operation Model includes object function and constraints, described object function Can be：

MinC=∑ C_F+∑PE_n·G_n,in,

Wherein, C is the total production cost of described co-generation unit, C_FFor various raw coal costs, G_n,inFor n-th grade of desuperheat The flow that decompressor is opened, PE_nPenalty value after opening for n-th grade of temperature-decreased pressure reducer；Described constraints includes：Material balance Constraints, power balance constraints, steam turbine generator constraints, boiler constraints, temperature-decreased pressure reducer constraints With coal blending constraints；Computing module, for being calculated under current load electrically and thermally using MILP algorithm The result of described Optimal Operation Model, to obtain described Hazards in Power Plant co-generation unit under current load electrically and thermally Scheduling strategy.

In various embodiments, this scheduling application server can also include：

Data pre-processing unit 500, for carrying out pretreatment, described pre- place to the primary data needed for described condition model Reason method includes but is not limited to：Outlier detection algorithm, linear smoothing algorithm or standardized algorithm.

In various embodiments, this scheduling application server can also include：

Memory element 505, for being stored the scheduling strategy of described Hazards in Power Plant co-generation unit.And,

Show scheduling strategy unit 506, in response to user's request, the described scheduling strategy of storage being showed use Family, so that user triggers the Optimized Operation of described Hazards in Power Plant co-generation unit according to described scheduling strategy.

The application sets up co-generation unit global optimization scheduling model based on equipment working condition model, and to meet electricity, heat (is steamed Vapour) premised on total load, the minimum target of the total production cost of co-generation unit, automatically provide the scheduling scheme of optimization, real Show and coal blending, boiler, steam turbine generator, temperature-decreased pressure reducer have been combined the scheduling scheme making " global optimization ", thus Improve the economy of co-generation unit scheduling, realize the target of energy-saving and emission-reduction, cost efficiency, and avoid to a certain extent The artificial subjectivity of dispatcher and the problem brought because lacking experience.

The embodiment of the present application additionally provides a kind of scheduling system, and this scheduling system specifically can include dispatching application server With synthetic data integrated platform server, wherein, described scheduling application server includes：

It should be noted that each embodiment in this specification is all described by the way of going forward one by one, each embodiment weight Point explanation is all difference with other embodiment, between each embodiment identical similar partly mutually referring to. For system class embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, related part ginseng See that the part of embodiment of the method illustrates.

Last in addition it is also necessary to explanation, herein, such as first and second or the like relational terms be used merely to by One entity or operation are made a distinction with another entity or operation, and not necessarily require or imply these entities or operation Between there is any this actual relation or order.And, term " inclusion ", "comprising" or its any other variant meaning Covering comprising of nonexcludability, so that including a series of process of key elements, method, article or equipment not only include that A little key elements, but also include other key elements being not expressly set out, or also include for this process, method, article or The intrinsic key element of equipment.In the absence of more restrictions, the key element being limited by sentence "including a ...", does not arrange Remove and also there is other identical element in the process including described key element, method, article or equipment.

But described above is only example.Many different changes can be had in the case of without departing from the application spirit. For example, although in the present embodiment only for the purpose of network performance measurement, many large-scale networks also can be from the essence of the application Obtain an advantage in god, this includes but is not limited to, network monitoring system, network service cooperative operation system；These systems can be from this Moving on to certain class method consuming resource in the thought run on the high-performance server of concentration in application obtains useful enlightenment.

Although only carrying out result data matching with numerical computation method in the preferred embodiment of the application, for In specific embodiment, the matching of result data can be produced a desired effect using multiple mutation methods and model, and these mutation are calculated Method is the part of claimed the application.

Although in embodiments herein, scheduling application server employs a host process, in specific embodiment In this host process be not necessary the part that the application is implemented, on the premise of spiritual without departing from the application, to this host process The technology mutation modify, removing is the part of claimed the application.

Although in embodiments herein, can be carried out using UDP (User Datagram Protoco) between each part Communication it is clear that for those skilled in the art, can be by this communication in the case of without departing from the application spirit Agreement is replaced or changes.

Although in the preferred embodiment of the application, simple network ping-pong protocol is used in scheduling application server and network is visited Communicated between pin.Obviously for those skilled in the art, can in the case of without departing from the application spirit This communication protocol will be replaced or change.

Although all parts described in this application are computer program, any part in the application all can be soft Realize in part, firmware, hardware or its combination.

Although describing preferred embodiment here in detail, for those skilled in the art, very aobvious Different changes, increase so can be made in the case of without departing from the application spirit, replace or similar operations, and these It is regarded as within the protection domain of the application defined in claim.

Above to a kind of dispatching method of Hazards in Power Plant co-generation unit provided herein, server and system It is described in detail, specific case used herein is set forth to the principle of the application and embodiment, above reality The explanation applying example is only intended to help and understands the present processes and its core concept；General technology simultaneously for this area Personnel, according to the thought of the application, all will change in specific embodiments and applications, in sum, this theory Bright book content should not be construed as the restriction to the application.

Claims

1. a kind of dispatching method of Hazards in Power Plant co-generation unit is it is characterised in that methods described is deployed in coal-fired thermal power On scheduling application server in the scheduling system of factory's co-generation unit, described scheduling system also includes：Synthetic data is integrated Platform Server；Methods described includes：

Set up first needed for the condition model of individual device co-generation unit from the acquisition of synthetic data integrated platform server Beginning data, described primary data includes：Steam turbine generator throttle flow, generated energy and the amount of drawing gas, the steam production of boiler, pressure and temperature reducing Device imports and exports real time data and historical data, the real time data of vapor (steam) temperature and the historical data of quantity of steam, the reality of steam pressure When data and historical data, raw coal physico-chemical analysis data, and, the price of raw coal；

Judge whether to meet the schedulable condition pre-setting, if it is, the reality electrically and thermally according to the output of described condition model When load data, and the controlled variable of impact described condition model output and auxiliary variable, determine coal fuel heating under current loads The scheduling strategy of power plant's co-generation unit；

Wherein, the described real-time load data electrically and thermally according to the output of described condition model, and affect described condition model The controlled variable of output and auxiliary variable, determine the scheduling strategy of Hazards in Power Plant co-generation unit under current loads, including：

By the real-time load data electrically and thermally of described condition model output, and obtain from synthetic data integrated platform server The controlled variable of impact described condition model output and auxiliary variable value, defeated as the Optimal Operation Model pre-building Enter；Described Optimal Operation Model includes object function and constraints, and described object function is：

MinC=∑ C_F+∑PE_n·G_n,in,

Wherein, C is the total production cost of described co-generation unit, C_FFor various raw coal costs, G_n,inFor n-th grade of pressure and temperature reducing The flow that device is opened, PE_nPenalty value after opening for n-th grade of temperature-decreased pressure reducer；Described constraints includes：Material balance constrains Condition, power balance constraints, steam turbine generator constraints, boiler constraints, temperature-decreased pressure reducer constraints and join Coal constraints；

Calculate the result of described Optimal Operation Model under current load electrically and thermally using MILP algorithm, with Obtain scheduling strategy under current load electrically and thermally for the described Hazards in Power Plant co-generation unit.

2. method according to claim 1 is it is characterised in that described judge whether to meet the schedulable condition pre-setting, Including：

Judge whether the change of the total load electrically and thermally of described Hazards in Power Plant co-generation unit within preset time period surpasses Cross preset first threshold value, or, judge the scheduling fire coal value of prediction in described Hazards in Power Plant co-generation unit and actual combustion Whether the difference of coal value is more than default Second Threshold.

3. method according to claim 1 is it is characterised in that also include：

4. method according to claim 3 is it is characterised in that also include：

In response to user's request, the described scheduling strategy of storage is showed user, so that user touches according to described scheduling strategy The Optimized Operation sending out Hazards in Power Plant co-generation unit described.

5. method according to claim 1 is it is characterised in that also include：

Primary data needed for described condition model is carried out with pretreatment, described preprocess method includes：Outlier detection algorithm, Linear smoothing algorithm or standardized algorithm.

6. method according to claim 1 is it is characterised in that also include：

7. method according to claim 6 is it is characterised in that also include：

Record input and the output of described condition model, and the sample that the equipment operating condition of reality is constituted, and judge to record Sample number whether exceed preset recording sample threshold, if it is, using the input of condition model of described record and output Condition model described in re -training.

8. a kind of scheduling application server of Hazards in Power Plant co-generation unit is it is characterised in that this server includes：

Obtain initial data unit, set up in co-generation unit each and set for obtaining from synthetic data integrated platform server Primary data needed for standby condition model, described primary data includes：Steam turbine generator throttle flow, generated energy and the amount of drawing gas, The steam production of boiler, temperature-decreased pressure reducer imports and exports the real time data of quantity of steam and historical data, the real time data of vapor (steam) temperature and Historical data, the real time data of steam pressure and historical data, raw coal physico-chemical analysis data, and, the price of raw coal；

Set up condition model unit, for setting up the operating mode of individual device in described co-generation unit according to described primary data Model；

Determine scheduling strategy unit, for described first judging unit result be in the case of, according to described operating mode mould The real-time load data electrically and thermally of type output, and affect controlled variable and the auxiliary variable of described condition model output, really The scheduling strategy of Hazards in Power Plant co-generation unit under settled preload；

Wherein, described determination scheduling strategy unit includes：

Determine input module, for the real-time load data electrically and thermally exporting described condition model, and from synthetic data The controlled variable of impact described condition model output and the value of auxiliary variable that integrated platform server obtains, as pre-building Optimal Operation Model input；Described Optimal Operation Model includes object function and constraints, and described object function is：

MinC=∑ C_F+∑PE_n·G_n,in,

With,

Computing module, for calculating described Optimized Operation under current load electrically and thermally using MILP algorithm The result of model, to obtain scheduling strategy under current load electrically and thermally for the described Hazards in Power Plant co-generation unit.

9. scheduling application server according to claim 8 is it is characterised in that described first judging unit includes：

First judge module, for judging electrically and thermally total of within preset time period described Hazards in Power Plant co-generation unit Whether the change of load exceedes preset first threshold value；

Second judge module, for judging the scheduling fire coal value of prediction in described Hazards in Power Plant co-generation unit and actual combustion Whether the difference of coal value is more than default Second Threshold.

10. scheduling application server according to claim 8 is it is characterised in that also include：

11. scheduling application servers according to claim 10 are it is characterised in that also include：

Show scheduling strategy unit, in response to user's request, the described scheduling strategy of storage being showed user, to use The Optimized Operation of described Hazards in Power Plant co-generation unit is triggered according to described scheduling strategy in family.

12. scheduling application servers according to claim 8 are it is characterised in that also include：

Data pre-processing unit, for carrying out pretreatment, described preprocess method to the primary data needed for described condition model Including：Outlier detection algorithm, linear smoothing algorithm or standardized algorithm.

13. scheduling application servers according to claim 8 are it is characterised in that also include：

Second judging unit, presets the 3rd threshold value for judging that the output valve of described condition model is more than with the difference of actual value Whether number of times exceedes default difference frequency threshold value；

Determine regulated quantity unit, for the result in described second judging unit for, in the case of being, determining according to this difference and using Regulated quantity in the output valve compensating condition model.

14. scheduling application servers according to claim 13 are it is characterised in that also include：

Training unit, for described 3rd judging unit result be in the case of, using the condition model of described record Input and output re -training described in condition model.

A kind of 15. scheduling systems of Hazards in Power Plant co-generation unit are it is characterised in that this system includes：Scheduling application clothes Business device and synthetic data integrated platform server, wherein, described scheduling application server includes：

Wherein, described determination scheduling strategy unit includes：

MinC=∑ C_F+∑PE_n·G_n,in,

With,