JPH0549883B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a boiler load control device, and more particularly to a boiler load control device that optimizes load followability while monitoring the stress generated in each part of the boiler.

Thermal power plants are often operated at intermediate loads depending on power demand, and as a result of this intermediate load operation, startup, shutdown, or rapid load changes are required within a short period of time.

When intermediate load operation is performed to meet this demand, in a turbine plant, the turbine rotor section,
In boiler plants, stress generated in thick-walled pipes in pressure-resistant parts such as superheaters and the resulting shortened lifespan are a problem.

BACKGROUND ART Conventionally, in turbine plants, turbine control systems have been developed that predict thermal stress in a rotor section with the aim of ensuring earlier turbine startup, shutdown, and rapid load changes while ensuring the longevity of the turbine.

The life consumption of the turbine rotor is calculated from the expected operating pattern at the time of planning, and stress limit values for each operating mode are determined based on this. In actual operation, the generated stress is calculated by predicting the steam state at an arbitrary time point, depending on whether the turbine is in load control mode or speed control mode, and this is combined with a pre-planned stress limit value. While comparing, search for the maximum load change rate and maximum speed increase rate within a range that does not exceed the stress limit value. The load change rate obtained in this way is the ALR (Automatic Load Regulator) setting value, and the speed increase rate is the ASR (Automatic Speed Regulator) setting value.
A control system with a set value (Regulator) was used.

As described above, the control systems used in conventional turbine plants have the following drawbacks.

The stress limit value of a turbine plant is determined by the life consumption calculated from the planned number of operations and the expected operation pattern, so if the actual operation pattern differs from the planned one, the stress limit value may become stricter. It can be either hot or mild.

Furthermore, in actual operation, since the total life consumption up to the present time is not integrated, the operation results up to that point cannot be reflected in subsequent operations.

In order to predict the stress generated in the turbine rotor, since the rotor is a rotating body, the metal temperature cannot be directly measured, so the temperature distribution is calculated by estimating the heat transfer coefficient from the steam to the metal based on the steam state. The generated stress was predicted. Therefore, there is a natural limit to the improvement in accuracy of generated stress.

However, in boiler plants, boiler thick-walled tubes such as superheaters are high-temperature, high-pressure internal pressure equipment, so the temperature of the inner and outer surfaces of the metal can be directly measured, and the generated stress can be calculated. The accuracy of estimating generated stress can be further improved.

Furthermore, in the life consumption of the rotor of a turbine, fatigue damage due to repeated thermal stress occurring in this rotor is dominant, so monitoring and control of thermal stress has become a subject, but in boiler plants, Because it is a high-temperature, high-pressure internal pressure device, in addition to fatigue due to repeated thermal stress, a large proportion of its life is consumed by creep. For this reason, it is essential to accumulate creep damage data on various boiler materials and establish online creep damage evaluation technology in order to complete an online boiler life management system.

In addition, in order to optimize the lifetime consumption of the entire thermal power plant, it is necessary to manage the stress generated and lifetime consumption in both the boiler and turbine plants, and at the same time, coordinate control between them. There was no coordinated control.

In other words, in the turbine thermal stress management system, when thermal stress becomes excessive during operation, the turbine's load change rate is reduced, but even if the load change rate is zero (load maintenance), the turbine's thermal stress is sufficiently reduced. This happens especially during plant start-up. In such a case, it is necessary to reduce the temperature increase rate (fuel change rate) of the boiler, and cooperative control of the boiler and turbine is required.

The present invention attempts to eliminate such conventional drawbacks, and its purpose is to provide a boiler load control device that can perform load control based on stress limit setting values.

In order to achieve the above-mentioned object, the present invention includes a current stress value calculation means for predicting a current stress value based on a measured value from a boiler pressure-resistant part, and a stress limit value setting means for outputting a stress limit setting value planned in advance. , a comparison means for comparing the current stress value and the stress limit setting value, and comparing the current stress value and the stress limit setting value by the comparison means, and reducing the load within a range that does not exceed the stress limit setting value. The control is performed to obtain the rate of change.

Embodiments of the present invention will be described below with reference to the drawings.
Fig. 1 is a schematic system diagram of the boiler load control device of the present invention, Fig. 2 is a characteristic diagram showing time on the horizontal axis and load change rate on the vertical axis, and Fig. 3 is a characteristic diagram showing time on the horizontal axis and stress prediction on the vertical axis. FIG. 2 is a characteristic diagram showing values, and FIG. 3 shows changes in predicted stress values due to changes in load change rate. Figure 4 is a vertical cross-sectional view of the boiler thick-walled tube where stress is monitored.
Figure 5 is an enlarged view of the cylindrical model for calculating the temperature distribution in the - line cross section of Figure 4, Figure 6 is a temperature distribution curve diagram in the metal plate thickness direction, and Figure 7 shows the allowable number of repetitions on the horizontal axis. Figure 8 is a schematic diagram of principal stress difference generation with time on the horizontal axis and principal stress difference on the vertical axis, and Figure 9 is a design fatigue diagram with stress strip amplitude on the vertical axis. , a schematic diagram of stress generated during boiler operation with stress on the vertical axis, and time in Figure 10 on the horizontal axis,
Stress relaxation diagram showing relaxation stress values on the vertical axis, No. 11
The figure is a creep rupture diagram in which the horizontal axis shows the rupture time and the vertical axis shows the relaxation stress value, and FIG. 12 is a characteristic diagram in which the horizontal axis shows the stress limit value and the vertical axis shows the life consumption.

Hereinafter, the outline of the boiler load control device will be explained using FIG. 1.

As a typical example of a boiler thermal stress monitoring point, the nozzle corner part 2 of the high-temperature, high-pressure steam header (header) 1 of a superheater will be explained as an example, but such thermal stress monitoring points may be provided at multiple locations in the entire boiler. .

In order to determine the thermal stress at the monitoring points of the boiler, first, measure the metal temperature values 3 and 4 on the inner and outer surfaces of header 1.
The metal inner surface temperature 6 is detected by the metal inner surface temperature detector 5, and the metal outer surface temperature 8 is detected by the metal outer surface temperature detector 7, and these two signals 6, 8 are added to the temperature distribution calculator 9, and these two signals 6, 8, a metal thickness direction temperature distribution calculation value 10 is calculated.

On the other hand, a pressure measurement value 11 is detected from inside the header 1 by a pressure detector 12, and a steam pressure signal 13 obtained by detecting this pressure measurement value 11 by the pressure detector 12 is applied to a current stress value calculator 14.

This current stress value calculator 14 calculates the current stress by adding the thermal stress calculated based on the metal thickness direction temperature distribution calculation value 10 and the internal pressure calculated based on the steam pressure signal 13 detected by the pressure detector 12. Calculate the value 15.

The life consumption calculator 16 calculates a life consumption value 17 due to fatigue and creep based on the current stress value 15.
The stress limit value setter 18 subtracts the life consumption calculation value 17 in actual operation from the life distribution determined for each plan for each monitoring point and operation mode.
Calculate the remaining life, determine the allowable life consumption per operation mode in the future from the remaining life and the number of remaining operations, and set the stress that is predicted to cause this life consumption as the stress limit setting value 19 of header 1. do.

The stress limit set value 19 in this stress limit value setter 18 can be updated every arbitrary number of times of starting and stopping.

Next, the current stress value 15 and the stress limit set value 19 are compared by the comparator 20, and if the current stress value 15 exceeds the stress limit set value 19 as a result, the load hold signal 22 is generated from the load hold signal generator 21. , and sent to the boiler load controller 23.

On the other hand, if the current stress value 15 is less than or equal to the stress limit set value 19, one of multiple load change rates set in advance to determine the optimum load change rate (fuel change rate, pressure change rate) for the boiler. The maximum value is set as the load change rate setting value 25 by the load change rate setter 24 from the load change rate set value 25.

Next, the stress predicted value calculator 26 calculates the generated stress predicted value 27 after Δτ ₁ minute when the engine is operated at the load change rate setting value 25, as shown in FIG. Note that Δτ ₁ is given as the target predicted time 29 of the target predicted time setter 28.

The predicted generated stress value 27 and the stress limit value 19 are compared with the comparator 30, and if the predicted generated stress value 27 is less than the stress limit value 19, the set load change rate setting value 25 is determined.
is determined as the final load change rate 31 in terms of the generated stress. However, at the stage of calculating the predicted stress value 27, the predicted stress value 27 becomes the stress limit value 19.
If it exceeds the load change rate set value 25, use the load change rate setter 24 to select the load change rate set value 25 that is lowered by one step, calculate the generated stress prediction value 27 again, and use the load change rate setter 24 to set the load change rate as described above. rate setting value 25, stress prediction value calculator 26, generated stress prediction value 27, comparator 3
Repeat step 0.

The final load change rate 31 determined in this way is compared with the turbine load change rate limit set value 33 of the turbine thermal stress monitoring device 32 by the comparator 34, and the smaller value of the two is set as the final load change rate 35. The optimum load change rate set value 37 is set in the optimum load change rate setting device 36 and sent to the boiler load controller 23.

In this way, the desired period Δτ ₂ (<
Since the process is repeated at intervals of Δτ ₁ ), the optimum load change rate can be determined while constantly monitoring the generated stress.

Next, an example of determining the boiler load change rate will be explained using FIGS. 2 and 3.

Assuming that the boiler was operated at the load change rate A up to the present time T ₁ in Figure 2, and that it was operated at the same load change rate B until time T ₂ after Δτ ₁ , the predicted stress value would be the curve shown in Figure 3. D, and the stress limit value 19 is exceeded at time _T2 .

In this case, as described above, by selecting the load change rate set value 25 which is one level lower with the load change rate setter 24 shown in FIG. The boiler load change rate C as shown by curve E is determined.

On the other hand, recently, thermal stress monitoring and control equipment 32 has been installed in turbine plants, and this turbine thermal stress monitoring equipment 32 originally monitors the turbine so that the thermal stress in each part of the turbine is within the allowable value. It controls the load change rate (steam amount change rate), especially when starting the boiler (temperature rising process).
Now, even if the turbine load increase rate is set to zero (load maintenance), the turbine thermal stress may not be able to be reduced to a permissible value or less unless the steam temperature increase rate is lowered. In such a case, a turbine load change rate limit setting value 33 is sent from the turbine thermal stress monitoring device 32 as shown in FIG. If this turbine load change rate limit set value 33 is equal to or less than the load change rate C described above, the turbine load change rate set value 33 is determined as the optimum boiler load change rate, and coordinated load control between the boiler and the turbine is possible. Become.

Figure 4 shows an enlarged cross section of header 1 of the boiler.
Focusing on the nozzle corner part 2, which is the stress concentration part of the header 1, the generated stress and life consumption are monitored.

This stress generated in the nozzle corner part 2 is
It can be determined by multiplying the stress generated in the cylindrical part by the stress concentration factor.

The thermal stress of the cylindrical part is determined from the temperature distribution in the thickness direction of the cylindrical part, and the temperature distribution is determined by the heat conduction method of the cylindrical part (1) Equation 1/ααT/αt=α ² T/αr ² +1/ααT/αr... ...(1) α: Metal temperature conductivity T: Metal temperature t: Time r: Divide the distance from the cylinder center into N equal parts into concentric cylinders as shown in Figure 5, take contact points at equal intervals, and differentiate. solve. The difference formula is
(2) Formula 1/αT _i,j+1 −T _i,j /δt=1/2 {T _i+1,j+1
−2T _i,j+1 +T _i-1,j+1 /(δr) ² +T _i+1,j −2T _i,j +T _i-1,j /(δr) ² }+1/2r{T _{i+ 1,
j+1} −T _i-1,j+1 ／2δr＋T _i+1,j −T _i-1,j ／2δr}……(2)
i: Nodal parameter j: Time parameter expressed as (N-1) unknowns T _N,j+1 (N=
1, 2, ..., n-1), (N-1) equations are obtained, and the boundary values T _p , _j+1 , T _o,j+1 are the current internal and external temperatures of the metal. The measured values 3 and 4 are given by the metal inner surface temperature signal 6 and the metal outer surface temperature signal 8, respectively.

Therefore, the calculated value of the temperature distribution in the thickness direction inside the metal is 1
0, that is, T _N,j+1 (N=1 to n-1) can be obtained from equation (2).

An example of the calculated temperature distribution value 10 is shown in FIG. Here, the thermal stress in the three axial directions, that is, the radial thermal stress σrt, the circumferential thermal stress σθt, and the axial thermal stress σzt
can be calculated using the following equations.

σ _rt =Eα′/1−ν{1/b ² −a ² (1−a ²
/r ² )∫ ^b _a Trdr−1/r ² ∫ ^b Trdr……(3) σ〓 _t =Eα′/1−ν{1/b ² −a ² (1+a ²
/r ² )∫ ^b _a Trdr+1/r ² ∫ ^b _a Trdr−T}……(4) σ _zt =Eα′/1−ν{2/b ² −a ² ∫ ^b _a Trdr−T}……(
5) Here, σ _rt : Radial thermal stress E: Young's modulus σ _t : Circumferential thermal stress α′ : Linear expansion coefficient σ _zt : Axial thermal stress ν : Poisson's ratio Next, the three-directional stress due to internal pressure is Obtained from equations (6), (7), and (8).

σ _rp = −p ...(6) σ〓 _p =P・D _i /2t+p/2 ...(7) σ _zp =P・D _i /2t+p/2 ...(8) Here, σ _rp : radius Directional internal pressure stress, P: Internal pressure σ〓 _p : Circumferential internal pressure stress D _i : Internal diameter σ _zp : Axial internal pressure stress t: Plate thickness Equations (3) to (8) above are the stresses occurring in the general part of the cylinder, The stress generated in stress concentration areas such as the nozzle is determined by multiplying the stress generated in the general area by the stress concentration coefficient.
Therefore, the three-directional stress of the current stress value 15 generated in the nozzle corner portion 2 can be obtained from (9) to 〓.

σ _r =K _rt・σ _rt +K _rp・σ _rp ...(9) σ〓=K〓 _t・σ〓 _t +K〓 _p・σ〓 _p ...(10) σ _z =K _zt・σ _zt +K _zp・σ _zp ……〓 Where, K _rt : Radial thermal stress concentration coefficient K〓 _t : Circumferential thermal stress concentration coefficient K _zt : Axial thermal stress concentration coefficient K _rp : Radial internal pressure stress concentration coefficient K〓 _p : Circumferential internal pressure stress concentration coefficient K _zp : Axial internal pressure stress concentration coefficient σ _r : Total radial stress σ: Total circumferential stress σ _z : Total axial stress Next, the life consumption calculator 16 calculates the current stress value 1
5, determine fatigue life consumption and creep life consumption.

First, the fatigue life consumption is determined by calculating the principal stress difference from the equations (9) to .

S ₁ = σθ−σ _z …〓 S ₂ = σ _z −σ _r …〓 S ₃ =σ _r −σ〓 …〓 Next, the stress amplitudes for the three principal stress differences are determined and shown in Figure 7. Calculate fatigue life consumption from the design fatigue diagram. FIG. 8 shows the fluctuation of the principal stress difference S ₁ and the stress amplitude Z〓 ₁ Z〓 ₂ is determined. For S ₂ and S ₃ , Z〓 ₁ , Z〓 _{2 .} . . , Z〓 ₁ , Z〓 ₂ are found in the same way.

That is, Z〓 ₁ , Z〓 ₂ for S ₁ (Z〓 ₁ > Z〓 ₂ ) Z〓 ₁ , Z〓 ₂ for S ₂ … (Z〓 ₁ > Z〓 ₂ >…) Z〓 ₁ , Z for S ₃ 〓 ₂ ... (Z〓 ₁ > Z〓 ₂ > ...), and the stress strip amplitudes H ₁ , H ₂ , ... are obtained from equations (15) and (16).

H ₁ = max{Z〓 ₁ , Z〓 ₁ , Z〓 ₁ }÷2 ...(15) H ₂ =max{Z〓 ₂ , Z〓 ₂ , Z〓 ₂ }÷2 ...(16) Stress piece The amplitudes H ₁ , H ₂ . . . and the allowable number of repetitions N ₁ , N _{2 .} . . are obtained from the design fatigue diagram shown in _FIG .

φ _f =1/N ₁ -1/N ₂ +... (17) On the other hand, creep life consumption is determined by the following procedure. FIG. 9 shows a schematic diagram of generated stress. The stress generated changes to the compressive stress side due to fluctuations in thermal stress during cold start and hot start, and then as steady operation begins, the thermal stress decreases and becomes mainly tensile stress due to internal pressure stress. The stress that has reached the initial stress σ _i is relaxed over time and fluctuates as σ(t). Therefore, the time is divided into Δt, and the relaxation stress value σ(t) in each section is determined from the stress relaxation diagram shown in FIG. Next, the rupture time t _r (σ(t)) corresponding to the relaxed stress value σ(t) is determined from the creep rupture diagram shown in FIG.

The creep life consumption during time Δt is obtained by Δt/t _r , and the creep life consumption φ _c between one start and stop is
It is obtained using equation (14).

φ _C =〓 _T 〓 ^t=0 Δt/t _r (18) Then, the life consumption calculation value 25 is obtained by equation (19) using the generally used linear law equation.

φ=φ _f +φ _c ...(19) Here, the life consumption calculation is performed with the start-stop period as one cycle.

The stress limit value setting device 18 sets the number of operations N ^O _C (cold start number) and N ^O _W for each operation mode at the time of planning.
(number of warm starts), N ^O _H (number of hot starts), and the corresponding life distributions are φ ^O _C , φ ^O _W ,
Assuming φ ^O _H , the allowable life consumption per cycle for each mode during planning is φ ^O / _C / N ^O / _C , φ ^O / _W / N ^O / _W , and φ ^O / _H / N ^O /, respectively. Given as _H , the number of operations for each mode up to now is N ⁱ _C , N ⁱ _W , N ⁱ _H , and the total life consumption for each mode is φ ⁱ _tC , φ ⁱ _tW , φ ⁱ _tH . The allowable life consumption per start/stop for each mode is (21), respectively.
It can be obtained from equations (22) and (23).

φ ⁱ _C =φ ^O / _C −φ ⁱ / _tC /N ^O / _C −N ^o / _c ……(21) φ ⁱ _w =φ ^o / _w −φ ⁱ / _tW /N ^O / _W −N ⁱ / _w ......(22) φ ⁱ _H = φ ^O / _H φ ⁱ / _tH /N ^O / _H N ⁱ / _H ...... (23) Next, from the stress limit value-life consumption diagram shown in Fig. 12, Let the stress corresponding to the allowable life consumption be the stress limit value 19.

In addition, in the embodiment shown in FIG. 1, the current stress value is 15
As an estimation calculation method, a method is shown in which it is obtained from the measured values of metal temperature on the inner and outer surfaces of the pipe3 and 4. However, as an applied example, it is also possible to directly measure the generated stress by attaching a strain gauge directly to the stress monitoring point.The same effect can be obtained. can be obtained.

As described above, according to the boiler load control device of the present invention, life consumption associated with plant operation can be carried out in a planned manner, and the planned life can be used effectively. In other words, it becomes possible to operate with high reliability and safety during the scheduled period, which also leads to improved economic efficiency. Furthermore, since load control is performed based on stress values, there is no need to perform gentle load control that takes into account an excessive margin for limiting stress, and load followability is also improved.

When newly performing boiler load control as in the present invention, even if the boiler is already installed, it is economical to install a metal thermocouple or the like at the stress monitoring location.

The present invention has the above-described configuration, and can perform load control based on the stress limit set value.
Load followability can be improved.

[Brief explanation of the drawing]

Fig. 1 is a schematic system diagram of the boiler load control device of the present invention, Fig. 2 is a characteristic diagram showing time on the horizontal axis and load change rate on the vertical axis, and Fig. 3 is a characteristic diagram showing time on the horizontal axis and stress prediction on the vertical axis. Figure 4 is a longitudinal cross-sectional view of the boiler thick-walled tube (header) where the stress is monitored, Figure 5 is an enlargement of the cylindrical model for calculating the temperature distribution in the - line cross section of Figure 4. Figure 6 is a temperature distribution curve diagram in the metal plate thickness direction, Figure 7 is a design fatigue diagram with the horizontal axis showing the allowable number of repetitions and the vertical axis showing the stress fragment amplitude, and Figure 8 is the horizontal axis. Figure 9 is a schematic diagram of the generation of principal stress differences, with time on the horizontal axis and stress on the vertical axis. Figure 9 is a schematic diagram of the stress generated during boiler operation, with time on the horizontal axis and stress on the vertical axis. A stress relaxation diagram with time on the axis and relaxation stress value on the vertical axis, Figure 11 is a creep rupture diagram with time to rupture on the horizontal axis and relaxation stress value on the vertical axis, and Figure 12 shows stress on the horizontal axis. It is a characteristic diagram showing the limit value and the life consumption on the vertical axis. 1... Header, 2... Nozzle corner part, 3, 4
... Temperature measurement value, 11 ... Pressure measurement value, 14 ...
Current stress value calculator, 15...Current stress value, 18...
...Stress limit value setter, 19...Stress limit value, 20
... Comparator, 35 ... Optimal load change rate.

Claims (1)

[Scope of Claims] 1. Current stress value calculation means for predicting the current stress value based on the measured value from the boiler pressure-resistant part; Stress limit value setting means for outputting a stress limit setting value planned in advance; and a comparison means for comparing the current stress value and the stress limit set value, and the current stress value and the stress limit set value are compared by the comparison means, and the boiler load change rate is determined within a range that does not exceed the stress limit set value. A boiler load control device characterized by controlling the load so as to obtain the following.