CN104156781A - Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal - Google Patents
Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal Download PDFInfo
- Publication number
- CN104156781A CN104156781A CN201410337701.4A CN201410337701A CN104156781A CN 104156781 A CN104156781 A CN 104156781A CN 201410337701 A CN201410337701 A CN 201410337701A CN 104156781 A CN104156781 A CN 104156781A
- Authority
- CN
- China
- Prior art keywords
- coal
- minimum
- blending
- mixed
- ature
- 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.)
- Pending
Links
Abstract
The invention discloses a method for acquiring a coal parameter of thermal power generating unit boiler cost minimum mixed coal. The method adopts the objective functions that high-grade coal composition is minimum, poor coal composition is maximum and raw coal unit cost is minimum to establish a dynamic coal composition mathematical model, and predicts the model according to a general regression neural network, so that the acquired coal quality parameter value of the mixed coal is more reasonable, and rationality and economy of boiler operation are ensured. The method comprises the following steps: adopting the objective functions that high-grade coal composition is minimum, poor coal composition is maximum and raw coal unit cost is minimum, adopting the constraint conditions of single coal database, boiler design coal quality and historical coal data, establishing the dynamic coal composition mathematical model, and adopting the general regression neural network technology to establish coal quality characteristic prediction model for the mixed coal. The method is suitable for acquiring the coal parameter of thermal power generating unit boiler cost minimum mixed coal.
Description
Technical field
The present invention relates to the minimum mixed coal ature of coal of a kind of thermal power unit boiler cost parameter acquiring method, belong to thermal power unit operation mixed coal optimisation technique field.
Background technology
Coal resources in China distributes wide, wide in variety, and Coal rank property difference is large.Boiler of power plant when design be generally based on certain given design coal, and boiler is subject to many objective condition restrictions while using design coal, as coal is nervous, coal source is numerous, transportation cost etc.In the face of the various unsettled fire coals of fire coal and coal source with different qualities, need to adopt scientific and reasonable blending and advanced ature of coal forecasting techniques, reach and can reduce coal blending at power plant cost objective, can improve again into stove coal stability and reliability, make finally to enter coal-based of stove and can realize designing coal quality requirement, guarantee boiler safety stable operation.
Least cost mixed coal ature of coal parameter prediction is an important technology of thermal power unit operation.Least cost mixed coal ature of coal parameter is generally entered coal list coal database by recent power plant,, colm proportioning maximum minimum with fat coal proportioning and raw coal unit price is minimum joins target for mixing, adopt the method for exhaustion to calculate all coal blending combinations in single coal database, determine Rational mixed coal coal and best proportioning, the mixed coal ature of coal nonlinear parameter forecast model that utilizes nerual network technique to set up, mixed coal ature of coal parameter is predicted, offer operations staff, operations staff, according to the mixed coal ature of coal parameter of prediction, adjusts boiler operatiopn in time.Also optimum coal mixture scheme array mode is offered to Ran Yun administrative center, can realize online quick optimum coal blending simultaneously.Therefore, the key of least cost mixed coal ature of coal parameter prediction is Rational mixed coal coal and best proportioning.
Fired power generating unit mixed coal ature of coal parameter, relates to determining of boiler side least cost mixed coal ature of coal parameter at present, and Main Basis mixed coal ature of coal parameter and boiler design parameter, determine in conjunction with operations staff's experience.Due to diversity and the complicacy in power plant's coal source, this predicted value determines that method is difficult to take into full account into stove ature of coal diversity the impact on predicted value, lacks accuracy.
Summary of the invention
The object of the invention is, determine the problem existing according to existing apparatus of thermo-electric power boiler side least cost mixed coal ature of coal parameter, the present invention proposes a kind of thermal power unit boiler least cost mixed coal ature of coal parameter acquiring method.
Realizing technical scheme of the present invention is, the present invention adopts the method for exhaustion and General Neural Network technology, minimum according to fat coal proportioning minimum, colm proportioning maximum and raw coal unit price is objective function, set up dynamic blending coals mathematical model, and according to generalized regression nerve networks forecast model, make to obtain mixed coal ature of coal parameter value and have more rationality, to guarantee rationality, the economy of boiler operatiopn.
A kind of thermal power unit boiler least cost of the present invention mixed coal ature of coal parameter acquiring method step is:
(1) gather the conventional single coal data of station boiler, set up single coal information database;
(2) take that fat coal proportioning is minimum, colm proportioning maximum and raw coal unit price minimum be objective function, and using single coal database, boiler design ature of coal and in history coal data as constraint condition, set up dynamic blending coals mathematical model;
(3) the dynamic blending coals mathematical model of setting up based on step (2), adopts the method for exhaustion to seek all over all coal blendings combinations in single coal database, determines most effective Coal Blending and optimal proportion, and is labeled as " optimum mixed coal ";
(4) the single coal database based on step (3) " optimum mixed coal " and step (1), adopts generalized regression nerve networks technology to set up mixed coal coal characteristic forecast model;
(5) based on step (3) " optimum mixed coal " and the generalized regression nerve networks forecast model based on step (4), input coal and mixed coal ratio, technical analysis, ultimate analysis and the combustion characteristics of prediction mixed coal;
The constraint condition that the inventive method step (2) is set up is input variable, comprises single coal database, boiler design ature of coal and coal data in history.
Boiler dynamic blending coals computation model of the present invention comprises coal blending constraint condition and objective function:
The coal blending constraint condition of described coal blending computation model:
There is n kind list coal, prepare the steam coal with m technical indicator T, if j kind list coal (j=1,2,3 ... n) through chemical examination, draw i technical indicator (i=1,2,3 ... m) be T
ij, establish again the percent X of j kind list coal in coal blending
j, by i index of n kind list coal configuration, be so: f (T
ij, X
j, n);
A certain boiler, according to its single coal database, boiler design ature of coal and the reference of coal data etc. in history, is limited to A in its i technical indicator
i, under be limited to B
i,, just must be at A by i technical indicator of n kind list coal preparation
i～B
ibetween, that is: B
i≤ f (T
ij, X
j, n)≤A
i;
In addition, single coal blending in n, its proportioning sum must be 100%, and various single coal proportioning be on the occasion of.
The objective function of described coal blending computation model:
(1) pursue fat coal proportioning ratio minimum
In coal resources in China, fat coal proportion is less, and price is also relatively high, and general industry boiler should be used fat coal as far as possible less, with rationally with coal, reduce costs.Be provided with n kind coal and carry out blending, if j kind coal is fat coal, its proportioning is X
j, the objective function of fat coal proportioning minimum is P
min=X
j;
(2) pursue low-grade coal proportioning ratio maximum
In order to make full use of or some second-rate coals of local use, save fat coal, save transport power, reduce costs, with coal, should be based on utilizing low-grade coal more.Be provided with the blending of n kind list coal facies, wherein i kind list coal is low-grade coal or aboundresources, and convenient sources is supplied guaranteed coal, and its proportioning is X
j, for the more blending of vast scale, should pursue its proportioning for maximum, i.e. objective function P
max=X
j;
(3) pursuit cost is minimum
Reducing coal blending cost is the target that the present invention mainly pursues, and supposes to have the single coal of several differences to carry out coal blending, and the cost price of j kind list coal is C
j, its proportioning ratio is X
j, the cost price of coal blending is
determine that the objective function that coal blending cost is minimum is
General Neural Network forecast model of the present invention is:
As shown in Figure 3, it comprises four layers of neuron to mixed coal ature of coal forecast model network structure, i.e. input layer, mode layer, summation layer and output layer.Take input and output as a dimensional vector explanation.Wherein network is input as 1=[x
1, x
2... x
m]
t, be output as=[y1, y
2... y
m]
t.
Input layer number equals the dimension m of learning sample input layer, and each neuron directly passes to hidden layer by input variable.The neuron number of mode layer equals the number n of learning sample, the sample that each neuron is corresponding different, and in mode layer, neuronic transport function is Gaussian function,
At summation layer, have two types of neurons, wherein a kind of neuronic effect is that arithmetic summation is carried out in neuronic output to all mode layers, and each neuron of mode layer is 1 with these neuronic weights that are connected; Other neurons are weighted summation to the output of all mode layers, and the neuron number in output layer equals the dimension L of output vector in learning sample, and each neuron is divided by the output of summation layer.The ature of coal parameter that is input as three coals choosing from 15 kinds of single coals of forecast model and the wherein ratio of three kinds of single coals, be output as the ature of coal actual measurement parameter of corresponding mixed coal, by mixed coal, test the training of the basic ature of coal parameter of gained, obtain the forecast model that predicated error is lower.
The inventive method beneficial effect is compared with the prior art, the inventive method adopts method of exhaustion technology and General Neural Network technology, objective function and single coal database, the boiler design ature of coal minimum according to boiler fat coal proportioning, colm proportioning maximum, raw coal unit price is minimum, coal data are that constraint condition is set up dynamic blending coals mathematics computing model in history, and according to generalized regression nerve networks forecast model, make to obtain mixed coal ature of coal parameter value and have more rationality, to guarantee rationality, the economy of boiler operatiopn.
The inventive method is applicable to thermal power unit boiler least cost mixed coal ature of coal parameter acquiring.
Accompanying drawing explanation
Fig. 1 is boiler dynamic blending coals computation model;
Fig. 2 is boiler mixed coal ature of coal parameter neural network model;
Fig. 3 is General Neural Network forecast model of the present invention.
Embodiment
The present invention be directed to the minimum coal blending ature of coal of boiler cost difficult parameters with problem identificatioin, the minimum coal blending ature of coal of a kind of boiler cost parameter value acquisition methods of proposition.
Specific implementation method is as follows:
Step 1: gather the conventional single coal data of station boiler, set up single coal information database;
Step 2: the fat coal proportioning of take is minimum, colm proportioning maximum and raw coal unit price minimum be objective function, and using single coal database, boiler design ature of coal and in history coal data as constraint condition, set up dynamic blending coals mathematical model;
Step 3: the dynamic blending coals mathematical model of setting up based on step 2, adopt the method for exhaustion to seek all over all coal blendings combinations in single coal database, determine and meet constraint condition scope and minimum Coal Blending and the optimal proportion of cost, and be labeled as " optimum mixed coal ";
Step 4: the single coal database based on step 3 " optimum mixed coal " and step 1, adopts generalized regression nerve networks technology to set up mixed coal coal characteristic forecast model;
Step 5: based on step 3 " optimum mixed coal " and the generalized regression nerve networks forecast model based on step 4, input coal and mixed coal ratio, technical analysis, ultimate analysis and the burning performance parameter of prediction mixed coal.
The present embodiment adopts method of exhaustion technology and General Neural Network technology, the objective function minimum by boiler fat coal proportioning, colm proportioning maximum, raw coal unit price is minimum and single coal database, boiler design ature of coal, coal data are that constraint condition is set up dynamic blending coals calculated with mathematical model and obtained optimum mixed coal mode in history, and by generalized regression nerve networks forecast model, according to optimum mixed coal mode, predict least cost mixed coal ature of coal parameter, make to obtain mixed coal ature of coal parameter value and have more rationality, to guarantee rationality, the economy of boiler operatiopn.
Claims (4)
1. the minimum mixed coal ature of coal of a thermal power unit boiler cost parameter acquiring method, it is characterized in that, described method is minimum according to fat coal proportioning minimum, colm proportioning maximum and raw coal unit price is objective function, set up dynamic blending coals mathematical model, and according to generalized regression nerve networks forecast model, make to obtain mixed coal ature of coal parameter value and have more rationality, to guarantee rationality, the economy of boiler operatiopn;
The step of described method is:
(1) gather the conventional single coal data of station boiler, set up single coal information database;
(2) take that fat coal proportioning is minimum, colm proportioning maximum and raw coal unit price minimum be objective function, and using single coal database, boiler design ature of coal and in history coal data as constraint condition, set up dynamic blending coals mathematical model;
(3) the dynamic blending coals mathematical model of setting up based on step (2), adopts the method for exhaustion to seek all over all coal blendings combinations in single coal database, determines most effective Coal Blending and optimal proportion, and is labeled as " optimum mixed coal ";
(4) the single coal database based on step (3) " optimum mixed coal " and step (1), adopts generalized regression nerve networks technology to set up mixed coal coal characteristic forecast model;
(5) based on step (3) " optimum mixed coal " and the generalized regression nerve networks forecast model based on step (4), input coal and mixed coal ratio, technical analysis, ultimate analysis and the combustion characteristics of prediction mixed coal.
2. the minimum mixed coal ature of coal of a kind of thermal power unit boiler cost according to claim 1 parameter acquiring method, it is characterized in that, the constraint condition that described method step (2) is set up is input variable, comprises single coal database, boiler design ature of coal and coal data in history.
3. the minimum mixed coal ature of coal of a kind of thermal power unit boiler cost according to claim 1 parameter acquiring method, is characterized in that, described dynamic blending coals mathematical model comprises coal blending constraint condition and objective function;
The coal blending constraint condition of described coal blending computation model:
There is n kind list coal, prepare the steam coal with m technical indicator T, if j kind list coal (j=1,2,3 ... n) through chemical examination, draw i technical indicator (i=1,2,3 ... m) be T
ij, establish again the percent X of j kind list coal in coal blending
j, by i index of n kind list coal configuration, be so: f (T
ij, X
j, n);
A certain boiler, according to its single coal database, boiler design ature of coal and the reference of coal data in history, is limited to A in its i technical indicator
i, under be limited to B
i,, just must be at A by i technical indicator of n kind list coal preparation
i～B
ibetween, that is: B
i≤ f (T
ij, X
j, n)≤A
i;
In addition, single coal blending in n, its proportioning sum must be 100%, and various single coal proportioning be on the occasion of;
The objective function of described coal blending computation model:
(1) pursue fat coal proportioning ratio minimum:
Be provided with n kind coal and carry out blending, if j kind coal is fat coal, its proportioning is X
j, the objective function of fat coal proportioning minimum is P
min=X
j;
(2) pursue low-grade coal proportioning ratio maximum:
Be provided with the blending of n kind list coal facies, wherein i kind list coal is low-grade coal or aboundresources, and convenient sources is supplied guaranteed coal, and its proportioning is X
j, for the more blending of vast scale, should pursue its proportioning for maximum, i.e. objective function P
max=X
j;
(3) pursuit cost is minimum:
Reducing coal blending cost is the target that the present invention mainly pursues, and supposes to have the single coal of several differences to carry out coal blending, and the cost price of j kind list coal is C
j, its proportioning ratio is X
j, the cost price of coal blending is
determine that the objective function that coal blending cost is minimum is
4. the minimum mixed coal ature of coal of a kind of thermal power unit boiler cost according to claim 1 parameter acquiring method, is characterized in that, described General Neural Network forecast model is,
Mixed coal ature of coal forecast model network structure comprises four layers of neuron, i.e. input layer, mode layer, summation layer and output layer; When input and output are a dimensional vector, wherein network is input as 1=[x
1, x
2... x
m]
t, be output as=[y1, y
2... y
m]
t;
Input layer number equals the dimension m of learning sample input layer, and each neuron directly passes to hidden layer by input variable; The neuron number of mode layer equals the number n of learning sample, the sample that each neuron is corresponding different, and in mode layer, neuronic transport function is Gaussian function,
Summation layer has two types of neurons, and wherein a kind of neuronic effect is that arithmetic summation is carried out in neuronic output to all mode layers, and each neuron of mode layer is 1 with these neuronic weights that are connected; Other neurons are weighted summation to the output of all mode layers, and the neuron number in output layer equals the dimension L of output vector in learning sample, and each neuron is divided by the output of summation layer; The ature of coal parameter that is input as three coals choosing from 15 kinds of single coals of forecast model and the wherein ratio of three kinds of single coals, be output as the ature of coal actual measurement parameter of corresponding mixed coal, by mixed coal, test the training of the basic ature of coal parameter of gained, obtain the forecast model that predicated error is lower.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410337701.4A CN104156781A (en) | 2014-07-16 | 2014-07-16 | Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410337701.4A CN104156781A (en) | 2014-07-16 | 2014-07-16 | Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104156781A true CN104156781A (en) | 2014-11-19 |
Family
ID=51882277
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410337701.4A Pending CN104156781A (en) | 2014-07-16 | 2014-07-16 | Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104156781A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793580A (en) * | 2014-12-31 | 2015-07-22 | 远光共创智能科技股份有限公司 | System and method for data management of coaling plan information |
| CN105117808A (en) * | 2015-09-22 | 2015-12-02 | 华润电力登封有限公司 | Coal blending combustion optimizing method |
| CN105160499A (en) * | 2015-10-21 | 2015-12-16 | 华中科技大学 | Thermal power plant fuel analysis system and method |
| CN105320116A (en) * | 2015-11-19 | 2016-02-10 | 华润电力登封有限公司 | A thermal power plant fuel total value optimizing method and system |
| CN106054836A (en) * | 2016-06-21 | 2016-10-26 | 重庆科技学院 | Converter steelmaking process cost control method and system based on GRNN |
| CN106247395A (en) * | 2016-06-30 | 2016-12-21 | 华润电力登封有限公司 | A kind of feeder control method |
| CN110319455A (en) * | 2019-07-18 | 2019-10-11 | 国网山东省电力公司电力科学研究院 | A kind of boiler mixed coal blending |
| CN110728073A (en) * | 2019-10-23 | 2020-01-24 | 北京黑色智慧科技有限公司 | Multi-objective optimization method for coal washing and blending |
| CN112731868A (en) * | 2020-11-27 | 2021-04-30 | 山西焦化股份有限公司 | Refined intelligent coal blending system |
| CN116307513A (en) * | 2023-02-01 | 2023-06-23 | 华能国际电力股份有限公司上海石洞口第二电厂 | Thermal power plant coal blending scheme optimization method based on improved MOEA/D algorithm |
-
2014
- 2014-07-16 CN CN201410337701.4A patent/CN104156781A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 朱再兴: "锅炉动力配煤优化模型和专家系统研发及炉内燃烧仿真优化研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793580A (en) * | 2014-12-31 | 2015-07-22 | 远光共创智能科技股份有限公司 | System and method for data management of coaling plan information |
| CN105117808A (en) * | 2015-09-22 | 2015-12-02 | 华润电力登封有限公司 | Coal blending combustion optimizing method |
| CN105117808B (en) * | 2015-09-22 | 2018-08-10 | 华润电力登封有限公司 | A kind of coal mixing combustion optimization method |
| CN105160499A (en) * | 2015-10-21 | 2015-12-16 | 华中科技大学 | Thermal power plant fuel analysis system and method |
| CN105160499B (en) * | 2015-10-21 | 2018-09-07 | 华中科技大学 | A kind of fuel analysis System and method in thermal power plant |
| CN105320116B (en) * | 2015-11-19 | 2018-02-06 | 华润电力登封有限公司 | A kind of thermal power plant's fuel full price value optimization method and system |
| CN105320116A (en) * | 2015-11-19 | 2016-02-10 | 华润电力登封有限公司 | A thermal power plant fuel total value optimizing method and system |
| CN106054836A (en) * | 2016-06-21 | 2016-10-26 | 重庆科技学院 | Converter steelmaking process cost control method and system based on GRNN |
| CN106247395A (en) * | 2016-06-30 | 2016-12-21 | 华润电力登封有限公司 | A kind of feeder control method |
| CN110319455A (en) * | 2019-07-18 | 2019-10-11 | 国网山东省电力公司电力科学研究院 | A kind of boiler mixed coal blending |
| CN110319455B (en) * | 2019-07-18 | 2020-10-16 | 国网山东省电力公司电力科学研究院 | Boiler coal blending method |
| CN110728073A (en) * | 2019-10-23 | 2020-01-24 | 北京黑色智慧科技有限公司 | Multi-objective optimization method for coal washing and blending |
| CN112731868A (en) * | 2020-11-27 | 2021-04-30 | 山西焦化股份有限公司 | Refined intelligent coal blending system |
| CN116307513A (en) * | 2023-02-01 | 2023-06-23 | 华能国际电力股份有限公司上海石洞口第二电厂 | Thermal power plant coal blending scheme optimization method based on improved MOEA/D algorithm |
| CN116307513B (en) * | 2023-02-01 | 2023-12-22 | 华能国际电力股份有限公司上海石洞口第二电厂 | Thermal power plant coal blending scheme optimization method based on improved MOEA/D algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104156781A (en) | Method for acquiring coal parameter of thermal power generating unit boiler cost minimum mixed coal | |
| Ren et al. | Sustainability decision support framework for industrial system prioritization | |
| CN104123680A (en) | Post-evaluation method for comprehensively evaluating power grid science and technology projects | |
| Wang et al. | The impact and economic cost of environmental regulation on energy utilization in China | |
| Qu et al. | Using hybrid model to evaluate performance of innovation and technology professionals in marine logistics industry | |
| CN102902882A (en) | Method for evaluating operation quality of information systems | |
| Gatto et al. | An extensive data set on energy, economy, environmental pollution and institutional quality in the petroleum-reliant developing and transition economies | |
| CN104837184A (en) | Heterogeneous wireless network selection method based on interval triangular fuzzy number | |
| Musango et al. | Technology sustainability assessment of biodiesel development in South Africa: A system dynamics approach | |
| Yang et al. | Assessment and selection of smart agriculture solutions using an information error‐based Pythagorean fuzzy cloud algorithm | |
| CN105160422A (en) | Sewage treatment effluent total phosphorus prediction method based on self-organization cascade neural network | |
| CN107545321A (en) | A kind of ARMA RBF by-product gas generating capacity combination forecasting methods based on wavelet transformation | |
| CN104008429B (en) | Aero-engine assembly task priority ordering method | |
| CN105469180A (en) | TPS based industrial cluster enterprise production coordinated management evaluating system | |
| Yang et al. | The impact of quality of life on industrial and agricultural production and environmental efficiency in China's provinces | |
| Yang et al. | Comprehensive evaluation model of maritime industrial economic activities based on an AHP and BP neural network | |
| Liu et al. | Regional scale dynamic prediction of energy ecological footprint and its influencing factors | |
| Wang et al. | A simulation study of an innovation‐driven sustainable development model and its impacts and applications: A case study of the Yangtze River Economic Belt | |
| CN112488438A (en) | Industrial enterprise comprehensive benefit evaluation method and system combining credit information | |
| Li et al. | Regional unevenness in the construction of digital villages: A case study of China | |
| CN104679633A (en) | Effectiveness evaluation method of multi-aircraft parallel test process | |
| Bing et al. | Application of BP neural network in the evaluation of standard implementation | |
| Fang | The fuzzy mathematical evaluation of new energy power generation performance | |
| Li et al. | Research on GGDP accounting system based on GAN-BP neural network algorithm | |
| Li et al. | Application of BP Network based on GA in Student Evaluation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141119 |
|
| WD01 | Invention patent application deemed withdrawn after publication |