JPS6333370B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a system for collectively controlling the load (output) of a plant including a plurality of gas turbines and generators. More specifically, the present invention includes:
Considering economics such as start-stop loss and partial load efficiency loss when the load fluctuates, create a load pattern that minimizes loss (number of units switching mode)
The present invention relates to a method for collectively controlling loads of multiple generators, which determines and starts and stops gas turbines and generators (cycles) based on the determination.

A power generation plant consisting of multiple cycles consisting of a turbine and a generator, or a combined cycle power generation consisting of multiple waste heat recovery combined cycles consisting of a gas turbine, a waste heat recovery boiler, a steam turbine and a generator. In plants, etc., when controlling the load (output) of multiple generators at once in response to fluctuations in load (demand), conventional control was controlled based on the operator's judgment or feeling, without any particular standards. , the number of units was changed as appropriate.

The purpose of the present invention is to determine a load pattern (number switching mode) that minimizes the total loss, taking into account total losses such as cycle start/stop losses and partial load efficiency losses, and to determine the load pattern (number switching mode) that minimizes the total loss. Stop·
The object of the present invention is to provide a method for collectively controlling loads of multiple generators that can perform start-up control.

The present invention was made based on the following considerations.

(1) Turbine generators become less efficient at partial loads. In particular, in a combined cycle, efficiency decreases significantly at partial load compared to rated load.

(2) Therefore, in a plant consisting of multiple turbine generators or combined cycle generators, it is desirable to operate with the minimum number of generators when operating under low load for long periods of time.

(3) On the other hand, when switching the number of units, there is a loss when starting and stopping, so if the load drops for a short time, switching the number of units will increase the loss.

(4) Therefore, it is best to calculate the total loss for each possible number switching mode for the drop load, its duration, and the subsequent final load, and then determine the number of switches.

In order to achieve the above-mentioned object, in the first method of the present invention, (a) the minimum number of machines for the descending load and the final load is determined, and (b) the number of stopped machines and the number of started machines are maximized, respectively. For each load pattern (number switching mode) where the load is lower than that, calculate the total loss (at least the sum of start-stop loss and partial load efficiency loss) with respect to the scheduled or predicted descending load duration, and (c) Compare the total loss of the load pattern of
A load pattern with a minimum total loss is determined, and (d) each cycle is stopped and started based on the determination.

In addition, in the second method of the present invention, (a) find the minimum number of machines for the descending load and final load, and (b) find the maximum number of machines to stop and the number of machines to start, and the cases where the numbers of machines to start are respectively maximized and below. For each load pattern (number of units switching mode), determine the total loss (at least the sum of start-stop loss and partial load efficiency loss) as a function of the falling load duration time, and (c) determine whether the total loss in any two load patterns is equal. (d) Determine the total loss for the falling load duration by comparing the actual or predicted falling load duration with said critical duration. By keeping the smaller load pattern and (e) repeating the above comparison operation one after another, one load pattern with the minimum total loss is determined.

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of a normal combined cycle, which is configured with a gas turbine 1, a waste heat recovery boiler 2, a steam turbine 3, a condenser 4, a feed water pump 5, and a generator 6 as main equipment.

In FIG. 1, 7 is a fuel control valve, 8 is a control device, 9 is a generator output detector, 10 is a fuel flow rate detector, 11 is a plant output demand, and 12 is a generator breaker.

The control device 8 has a plant output demand 11,
It compares the signal with the signal from the generator output detector 9 and outputs the control output to the fuel control valve 7. That is, in a normal combined cycle, the generator load (output) is controlled only by the gas turbine fuel amount.

FIG. 2 shows an example of an electrical system of a power generation plant consisting of a plurality of gas turbines or a combined cycle. In the figure, TR ₁ to _{TR o} are the main transformers of each generator G ₁ to _{G o} .

Generally, the load demands of these plants are not given from a central power dispatch center for each individual cycle, but one demand is given for a plurality of plants.

On the other hand, a characteristic of gas turbines or combined cycles is that each cycle has high efficiency at its rated load, but the efficiency decreases significantly at partial loads.

FIG. 3 shows the relationship between load (output) and efficiency for n combined cycles.
Efficiency is expressed on the vertical axis, and the higher it goes, the better the efficiency is.

When the load decreases from L _a to L _b , if n units continue to operate, the operating point moves from point A to point B, resulting in inefficient operation. However, if the number of units is reduced to two for the load L _b and operation is performed at point B', efficient operation will be achieved.

Figure 4 shows that the total plant output (demand) is
An example of a case where L _a decreases to L _b between time t ₁ and t ₂ , and then increases again toward L _c between time t ₃ and t ₄ after a further time T ₂ has elapsed. It is.

In such a case, various methods of operating the combined cycle can be considered, but three typical cases will be illustrated and explained.

The first case is a method in which all the machines are operated without switching the number of machines, and the load of each individual cycle is the solid line L1.
It is indicated by.

The second case is a method of switching the number of devices.
In this case, the stop cycle is operated at L0 (dotted line), and the remaining operating cycles are performed at high output load L2.
(1-dot chain line). The third case is also a method of switching the number of devices. However, unlike the second case, in this case, the minimum number necessary for the final load L _c is left, each is operated at load L3 (double-dashed line), and the others are stopped.

The total loss L _ps in each case can be expressed using the symbols shown in FIG. 4 as shown in equation (1) below.

L _ps =L _Lps 〓L _pss =∫ ^t=t4 _t=t1 L・K(1/η−1/η _A
)dt＋∫ ^t=t4 _t=t1 N(q _F +q _EH −q _L +q _A +q _Z )dt…(1) Here, L _ps 〓 is partial load efficiency loss L _pss is start-stop loss L is, Plant load (excluding the load during the start-stop operation period of start-stop machines) η is plant efficiency (efficiency excluding start-stop machines) η _A is plant efficiency at rated load K is proportionality constant N is Number of combined cycle units to be stopped and started q _F is the calorific value of the supplied fuel q _E is the value equivalent to the calorific value of the in-house electric power during the start and stop operations q _L is the value equivalent to the calorific value of the electric power generated during the start and stop operations q _A is the calorific value of the auxiliary steam amount q _Z is the value equivalent to the calorific value of other start-stop losses Note that the partial load efficiency loss L _ps 〓 is the partial load efficiency fuel heat generation compared to the fuel calorific value at rated load efficiency. means an increase in quantity.

In addition, the start-stop loss L _pss means the loss from the start of the stop operation (start of load drop) to the completion of the stop, during the stop period, and from the start of the start-up operation to the completion of the start (until the target load is reached) in the start-stop cycle.

The number of stopped and started units is typically as follows for the first to third cases described above.

(1) First case (no switching of the number of units) Number of stopped units N _SD = 0 Number of activated units N _ST = N _M 〓−N _p … (When N _M 〓 > N _p ) = 0… (N _M 〓≦N _p ) (2) Second case (maximum number of units switched) Number of stopped units N _SD = N _p −N _M 〓 Number of started units N _ST = N _M 〓−N _M 〓 (3) Third case (corresponding to the final load Number of stopped units N _SD = N _p −N _M 〓 (However, if N _M 〓≧N _M 〓) Number of starting units N _ST = 0… (If N _M 〓≦N _p ) = N _M 〓− N _p ...(When N _M 〓＞N _P ) Here, N _P is the current number of operating machines N _M 〓 is the minimum number of machines required to output the final load L _c N _M 〓 is the falling load Minimum number of units required to output L _b Also, if the total losses corresponding to the first to third cases mentioned above are respectively L _ps1 , L _ps2 , and L _ps3 , then the minimum one among them corresponds to By selecting the load pattern (number of units switching mode), the optimum operation method corresponding to the required load demand can be obtained.

FIG. 5 shows a control device 8 that specifically realizes this.
0, and the current load L _a , the falling load L _b , the re-change load L _c , the duration T ₂ of the falling load, and the current number of operating vehicles N _P are given from the outside.

The outputs of the control device 80 include load commands, start and stop commands for each cycle, and the like. Note that the control device 80 may be any device that can perform the calculation of equation (1) above, and any calculation device with any configuration can be employed.

The efficiency characteristics of each combined cycle can be stored in advance in the control device 80 as characteristics corresponding to the load, or can be calculated by taking into account changes over time and environmental conditions.

In addition, the fuel amount calorific value q _F associated with starting and stopping, the calorific value equivalent value of the in-house electric energy during starting and stopping operations q _E , the calorific value equivalent value of the generated electricity during starting and stopping operations q _L , and the auxiliary The amount of steam q _A can be stored in advance in the control device 80, or can be calculated according to the start and stop modes.

In addition, losses due to startup and shutdown include gas turbine life loss, chemical injection loss, make-up water loss, etc.
It goes without saying that if you take price into account and convert it into a value equivalent to calorific value, you can select the optimal operating method that takes all of these losses into account.

FIG. 6 shows gas turbines 1-1 to 1-n, waste heat recovery boilers 2-1 to 2-n, and the steam generated by each waste heat recovery boiler collected in a steam header 16 to provide a common steam to the whole. 2 is a schematic diagram of a combined plant that supplies the steam to a turbine 14 and drives a steam turbine generator 15. FIG.

In this case, as is clear from the figure, the generator 1 corresponding to each gas turbine 1-1 to 1-n, respectively.
3-1 to 13-n and a steam turbine generator 15 are provided.

Even in the case of FIG. 6, the loss calculation according to the above-mentioned equation (1) is valid. Therefore, the economic efficiency of not switching the number of gas turbines and when switching the number of gas turbines is expressed by equation (1).
It is possible to compare and select the most suitable driving method based on the calculation results.

Next, a specific method (calculation method in the control device 80) for determining the number of units to start and stop the combined cycle when the load is expected to decrease once from the current value and then increase again as shown in FIG. 4 will be explained. .

In addition, as in the previous explanation, the current load
The number of operating units at L _a is N _P , the minimum number of units required to output the falling load L _b is N _M 〓, and the minimum number of units required to output the final (re-change) load L _c is Assume N _M 〓.

(1) First, calculate the minimum numbers N _M 〓 and _NM 〓 for the drop L _b and the final load L _c , respectively. This can be easily determined by dividing each load (demand) by the rated load of one combined cycle.

(2) When the number of operating units is N _P →N _M 〓→N _M 〓, that is, when the number of stopped units and the number of started units are set to their respective maximum values, total loss (at least starting and stopping) is calculated based on equation (1). (sum of loss and partial load efficiency loss).

(3) Next, increase N _M 〓 and N _M 〓 by 1 each, and calculate the total loss based on formula (1) for all load patterns (number switching mode) when the number of stopped or started machines is reduced. calculate.

(4) Among the total losses determined as above, the load pattern (number of units switching mode) corresponding to the minimum one is adopted, and starting and stopping of the combined cycle is controlled based on this.

In the above explanation, there are 6 N _P units, 2 N _M 〓 units,
Assuming that there are four N _M 〓, the above-mentioned method will be explained in more detail as follows.

That is, after calculating the total loss for each of the eight cases shown in Table 1, the total loss for each case is compared, and the smallest one is adopted as the actual load or number switching method.

Table 1 N _P →N _M 〓→N _M 〓 (A) 6 − 2 − 4 (B) 6 − 3 − 4 (C) 6 − 4 − 4 (D) 6 − 5 − 4 (E) 6 − 6-4 (F) 6-5-5 (G) 6-6-5 (H) 6-6-6 In addition to the methods described above, the following methods can be used to determine the actual load pattern (number of units switching mode). A method like this is also possible.

Given the falling load L _b and the final load L _c ,
If assumed, the total loss for each load pattern is the falling load, as seen from Eq.
It is a function of the duration T ₂ of L _b .

Therefore, when starting and stopping i unit, (i+j)
Assuming two load patterns (number switching mode) when starting and stopping machines, total loss for each
Let L _ps(i) and L _ps(i+j) be functions of time T ₂ , and L _ps(i) = F _i (T ₂ ) L _ps(i+j) = F _i+j (T ₂ ). Find the critical duration T ₂₀ of the falling load L _b such that F _i (T ₂ )=F _i+j (T ₂ ).

In the normal case, the start-stop losses are approximately constant, while the part-load efficiency losses increase with time T ₂ . Furthermore, the rate of increase increases as the number of combined cycles operated under partial load increases.

Therefore, the actual or predicted falling load duration T ₂ is compared with the critical duration T ₂₀ determined as above, and if T ₂ is greater than T ₂₀ then (i + j) orders of magnitude are determined. The total loss can be reduced by adopting a pattern of starting and stopping only the i unit when T ₂ is smaller than T ₂₀ .

Therefore, if we apply the above i and (i+j) to all cases and sequentially find the critical duration for each two cases, leaving a load pattern with a small total loss, finally, the total loss It is possible to find the load pattern that minimizes the

As is clear from the above description, the present invention (1) calculates the total loss for all expected load patterns (number of units switching mode),
(2) Find the total loss as a function of the falling load duration and find the actual critical duration that makes the total losses equal for two load patterns. Based on the comparison with the (or expected) falling load duration, leave the load pattern with the lower total loss for said falling load duration and do the same with respect to the remaining load pattern and the third load pattern. By repeating this process, the load pattern with the minimum total loss is found.

Therefore, according to the present invention, it is possible to realize a highly economical number switching method (load control method) that takes total loss into consideration.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing an example of the configuration of a combined cycle; Figure 2 is a schematic diagram showing the electrical system of multiple generators; Figure 3 is a diagram showing plant efficiency characteristics of a multiple generator; The figure shows an example of the temporal change in each cycle load when the load is changed, FIG. 5 shows the input and output of the control device embodying the control method of the present invention, and FIG. FIG. 2 is a block diagram showing an example of the configuration of a shaft type combined cycle. 1... Gas turbine, 2... Waste heat recovery boiler,
3... Steam turbine, 4... Condenser, 5... Water supply pump, 6... Generator, 7... Fuel control valve, 8,
80...control device.

Claims (1)

[Scope of Claims] 1. A plurality of cycles each consisting of a gas turbine and a generator driven by the gas turbine,
In a plant that applies output demand all at once and performs load control, when the load decreases from the current value to a drop load, continues for a certain period of time, and then increases again to the final (re-change) load, In the load batch control method, (a) find the minimum number of units for the descending load and final load, and (b) calculate each load pattern when the number of stopped units and the number of starting units is maximized and lower than that. (3) Find the total loss (at least the sum of start/stop loss and partial load efficiency loss) for each load pattern (number switching mode), and (c) compare the total loss for each load pattern.
A collective load control method for multiple generators, characterized by: determining a load pattern that minimizes total loss, and (d) executing stopping and starting of each cycle based on the determination. 2. In a plant that collectively applies output demand to multiple cycles, each consisting of a gas turbine and a generator driven by the gas turbine, and performs load control, the load once decreases from the current value to a drop load, or This is a collective load control method for multiple generators when the load increases again to the final (re-change) load after continuing for a certain period of time. For each load pattern (number switching mode) when the number of starting units is maximized or less than that, the total loss (at least the sum of starting/stopping loss and partial load efficiency loss) is calculated as a function of the falling load duration. (c) Find the critical duration of the falling load such that the total loss in any two load patterns is equal; and (d) Find the actual or predicted falling load duration and the critical duration. By comparing the above, the load pattern with the smaller total loss for the falling load duration is retained, and (e) the load pattern left in the above (d) and the third
By repeating the operations (c) and (d) above regarding the load pattern of A collective load control method for multiple generators, characterized by: executing startup.