CN105259758A - Thermal power unit operating parameter intelligent online optimization method based on massive historical data - Google Patents

Abstract

The invention discloses a thermal power unit operating parameter intelligent online optimization method based on massive historical data. The method comprises the steps of 1) setting optimization parameters, consisting of uncontrollable parameters which are the unit load, circulating water inlet temperature, ash content as received basis, moisture as received basis and the like, and controllable parameters which are the main steam temperature, main steam pressure, reheat steam temperature, vacuum, feed-water temperature, overheating attemperating water amount and the like; 2) extracting six uncontrollable parameters in 15 minutes, searching a case library for similar operation conditions according to a matching algorithm, if similar operation conditions can be found in the case library, going a step 4), and otherwise going to a step 3); 3) searching historical data for one operation condition with the lowest rate of coal consumption in a year according to the six uncontrollable parameter based on the matching algorithm; 4) finding out or nine controllable parameter optimization values in the corresponding case under a timestamp; and 5) storing the optimized result in the case library if the result is not in the case library.

Description

Based on mass historical data thermal power unit operation parameter intelligent online optimizing method

Technical field:

The invention belongs to thermal power generating technology field, be specifically related to a kind of based on mass historical data thermal power unit operation parameter intelligent online optimizing method.

Background technology:

At present, the adoptable optimization target values of fired power generating unit has design load, calculated value, trial value and empirical value, these methods are all also existing drawback in varying degrees, because some unit thermodynamic system sets up accurate mathematical model, obtain economical operation calculating desired value and also have difficulties; Get design parameter is generally applicable to be with basic load unit as desired value, for the unit of long-term variable load operation, uncomfortable conjunction design load is as desired value; Test method is then tested by set optimization, and by carrying out repetition test and adjustment to multiple typical load operating mode, the problem of test method is that experimentation cost is high, and the target operating condition point obtained is limited.

Summary of the invention:

Desired value is basis and the key problem of diagnosis of energy saving optimization, due to the drawback of prior art, the object of the invention is to determine unit optimal objective value accurately in real time, provide based on mass historical data thermal power unit operation parameter intelligent online optimizing method, the method passes through history steady state data optimizing value as Calculation Basis and basis for estimation, the operation characteristic change of the unit of tracking in time own, determines unit optimal objective value online.

For achieving the above object, the present invention adopts following technical scheme to realize:

Based on mass historical data thermal power unit operation parameter intelligent online optimizing method, comprise the steps:

1) set the parameter of optimizing, comprise uncontrollable parameter and controllable parameter, wherein, uncontrollable parameter is: unit load, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value; Controllable parameter is: main steam temperature, main steam pressure, reheat steam temperature, vacuum, feed temperature, overheated spray water flux, reheating spray water flux, flue gas oxygen content and exhaust gas temperature;

2) unit load value, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value 6 the uncontrollable parameters extracted in 15 minutes are pressed matching algorithm and are found operating condition close in case library, if this operating mode can be found in case library, then enter step 4), if can not find in case library, enter step 3);

3) 1 minimum operating mode of coa consumption rate in a year is found according to unit load value, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value 6 uncontrollable parameters by matching algorithm in the historical data;

4) find out the main steam temperature under corresponding case or time stamp, main steam pressure, reheat steam temperature, vacuum, feed temperature, cross diminishing flow, again diminishing flow, flue gas oxygen content and exhaust gas temperature 9 controllable parameter optimizing values;

5) if the result that this time optimizing obtains does not have in case library, be then saved in case library.

The present invention further improves and is, step 2) and 3) in, matching algorithm is as follows:

The similarity utilizing the geometric model method based on Distance geometry directional information to carry out unit operation operating mode case describes, current steady state operating condition data x _qwith case x in unit case library _isimilarity function can be expressed as:

$S_i=w_1\sqrt{e^{-D^2(x_q,x_i)}}+w_{2}cos\left({\mathrm{\δ}}_i\right)---\left(1\right)$

$D(x_q,x_i)=\sqrt{\underset{j=1}{\overset{6}{\Σ}}{\mathrm{\γ}}_j{(x_{qj}-x_{ij})}^2}---\left(2\right)$

$\cos \left({\mathrm{\δ}}_i\right)=\frac{\underset{j=1}{\overset{6}{\Σ}}{x}_{qj}{x}_{ij}}{\sqrt{\underset{j=1}{\overset{6}{\Σ}}{x}_{qj}^2}\sqrt{\underset{j=1}{\overset{6}{\Σ}}{x}_{ij}^2}}---\left(3\right)$

In formula, i=1 to 6, j=1 to 6, x _qjbe respectively the value of lower 6 the real-time measuring points of uncontrollable parameter of current steady state operating condition, x _ijrepresent the case value of 6 uncontrollable parameters in unit case library respectively, w ₁, w ₂for weight factor, be taken as 0.75 and 0.25 respectively; D (x _q, x _i) represent range information, γ _jrepresent the weighting coefficient of 6 uncontrollable parameters respectively, be taken as 0.3,0.2,0.1,0.1,0.1,0.2, cos (δ respectively _i) represent case directional information;

Formula (1) is utilized to calculate current steady state operating condition data and unit history operating mode similarity S _i, be greater than similarity threshold S by all _vhistory operating mode all as coupling operating mode.

The present invention further improves and is, similarity threshold S _vvalue is 0.8.

Relative to prior art, the present invention has following beneficial effect:

The present invention adopts based on history steady state data optimizing value as the Calculation Basis of the functional modules such as Optimized Diagnosis and basis for estimation, and the operation characteristic change of the unit of tracking in time own, determines unit target operating condition online.The present invention can meet the demand of producing the functional modules such as Examination of Small Indicators, power consumption analysis, running optimizatin, diagnosis of energy saving in real-time live, and then improves and improve unit performance.

Accompanying drawing illustrates:

Fig. 1 is the process flow diagram that the present invention is based on mass historical data thermal power unit operation parameter intelligent online optimizing method.

Fig. 2 is Present Thermal Power unit operation parameter intelligent online optimizing figure.

Embodiment:

Below in conjunction with concrete enforcement, the present invention will be further described.

Based on mass historical data thermal power unit operation parameter intelligent online optimizing method, as shown in Figure 1, detailed step comprises following content to its calculation process:

1) set the parameter of optimizing, uncontrollable parameter is: unit load, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter, as received basis low heat value; Controllable parameter is: main steam temperature, main steam pressure, reheat steam temperature, vacuum, feed temperature, overheated spray water flux, reheating spray water flux, flue gas oxygen content, exhaust gas temperature;

In actual motion, as shown in table 1 by the data gathering the above-mentioned uncontrollable parameter drawn:

Table 1:

Parameter name	Unit	Steady-state operation value
			Unit load	MW	310.19
Inlet Temperature of Circulating Water	℃	22.6
			As received basis ash content	％	18.06
Moisture as received coal	％	9.5
			Dry ash-free basis volatile matter	％	34.12
As received basis low heat value	kJ/kg	22640

2) 1 minimum operating mode of coa consumption rate in a year is found according to controllable parameters such as unit load value, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter, as received basis low heat values by matching algorithm in the historical data, the controllable parameter optimizing values such as main steam temperature corresponding under finding out this operating mode, main steam pressure, reheat steam temperature, vacuum, feed temperature, excessively diminishing flow, again diminishing flow, flue gas oxygen content, exhaust gas temperature, the optimizing result obtained is as shown in Figure 2;

Each optimizing parameter value corresponding to the operating mode that the coal consumption calculated under all coupling operating modes by matching algorithm is minimum is as follows:

Table 2:

Power plant operations staff can be instructed to adjust current operating parameter in time by above-mentioned optimizing parameter value, under making the unit moment be in optimized operation condition, and meet the function needs such as Examination of Small Indicators, power consumption analysis, running optimizatin, diagnosis of energy saving of production scene further, improve and improve unit performance.

Claims (3)

1., based on mass historical data thermal power unit operation parameter intelligent online optimizing method, it is characterized in that, comprise the steps:

1) set the parameter of optimizing, comprise uncontrollable parameter and controllable parameter, wherein, uncontrollable parameter is: unit load, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value; Controllable parameter is: main steam temperature, main steam pressure, reheat steam temperature, vacuum, feed temperature, overheated spray water flux, reheating spray water flux, flue gas oxygen content and exhaust gas temperature;

2) unit load value, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value 6 the uncontrollable parameters extracted in 15 minutes are pressed matching algorithm and are found operating condition close in case library, if this operating mode can be found in case library, then enter step 4), if can not find in case library, enter step 3);

3) 1 minimum operating mode of coa consumption rate in a year is found according to unit load value, Inlet Temperature of Circulating Water, as received basis ash content, moisture as received coal, dry ash-free basis volatile matter and as received basis low heat value 6 uncontrollable parameters by matching algorithm in the historical data;

4) find out the main steam temperature under corresponding case or time stamp, main steam pressure, reheat steam temperature, vacuum, feed temperature, cross diminishing flow, again diminishing flow, flue gas oxygen content and exhaust gas temperature 9 controllable parameter optimizing values;

5) if the result that this time optimizing obtains does not have in case library, be then saved in case library.

2. according to claim 1 based on mass historical data thermal power unit operation parameter intelligent online optimizing method, it is characterized in that, step 2) and 3) in, matching algorithm is as follows:

The similarity utilizing the geometric model method based on Distance geometry directional information to carry out unit operation operating mode case describes, current steady state operating condition data x _qwith case x in unit case library _isimilarity function can be expressed as:

$S_i=w_1\sqrt{e^{-D^2(x_q,x_i)}}+w_{2}cos\left({\mathrm{\δ}}_i\right)---\left(1\right)$

$D(x_q,x_i)=\sqrt{\underset{j=1}{\overset{6}{\Σ}}{\mathrm{\γ}}_j{(x_{qj}-x_{ij})}^2}---\left(2\right)$

$\cos \left({\mathrm{\δ}}_i\right)=\frac{\underset{j=1}{\overset{6}{\Σ}}{x}_{qj}{x}_{ij}}{\sqrt{\underset{j=1}{\overset{6}{\Σ}}{x}_{qj}^2}\sqrt{\underset{j=1}{\overset{6}{\Σ}}{x}_{ij}^2}}---\left(3\right)$

In formula, i=1 to 6, j=1 to 6, x _qjbe respectively the value of lower 6 the real-time measuring points of uncontrollable parameter of current steady state operating condition, x _ijrepresent the case value of 6 uncontrollable parameters in unit case library respectively, w ₁, w ₂for weight factor, be taken as 0.75 and 0.25 respectively; D (x _q, x _i) represent range information, γ _jrepresent the weighting coefficient of 6 uncontrollable parameters respectively, be taken as 0.3,0.2,0.1,0.1,0.1,0.2, cos (δ respectively _i) represent case directional information;

Formula (1) is utilized to calculate current steady state operating condition data and unit history operating mode similarity S _i, be greater than similarity threshold S by all _vhistory operating mode all as coupling operating mode.

3. according to claim 2 based on mass historical data thermal power unit operation parameter intelligent online optimizing method, it is characterized in that, similarity threshold S _vvalue is 0.8.