JPS5915000B2 - Computer-based load runback control method for power generation plants - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a load runback control method using a computer, and more particularly to a method for determining an optimal control value for a load runback proceeding process and determining the appropriateness of the control result.

In a generator stator winding direct water cooling system, if the cooling system fails, the cooling water flowing through the conductors may be cut off or reduced, or the cooling water may be supplied to the conductors before it is sufficiently cooled.

In this state, cooling is no longer possible, and the stator winding conductor becomes abnormally high in temperature, putting the generator in a dangerous state, so it is necessary to quickly reduce the load current.
This operation is commonly called load runback operation. With conventional load runback, when a so-called cooling water cutoff occurs due to low cooling water pressure or high cooling water temperature, a load reduction command is issued to the load limiter of the prime mover (for example, a turbine), and the generator is The load is controlled so that the load is reduced within the allowable time to the specified load that can be operated with cooling water cut off.

FIG. 1 is an explanatory diagram of the progress of load runback, where solid line 1 indicates an ideal state and line 23 indicates a failed state. T in FIG. After the cooling water is cut off at point T, the load runback progresses at a constant rate of change as shown by solid line 1, and reaches the specified load value for cooling water cutoff for that generator at point T, after the allowable time T has elapsed. For example, it is determined that the load runback is successful, and operation can be continued for a certain period of time determined by the conductivity of the cooling water at the time the failure occurs. If the load has not fallen to the specified load value at time T, it is determined that the load runback has failed and the generator is immediately tripped. However, this conventional method had the following problems. First, dashed line 2 in Figure 1
As shown in Figure 1, even though the load current has indeed decreased to the specified value at time T1, the load hardly decreases for some reason until time T2 in the process of reduction. As shown by chain line 3,
This is a case where load runback does not work properly for some reason, and operation continues with full load, and failure is determined for the first time at time T1, causing the generator to trip.

In either case, the cooling water in the coils begins to boil, causing problems such as local overheating and water delamination, which puts the generator in an extremely dangerous condition. Of course, it is also possible to verify the progress of load runback by detecting the load with a load detector at an intermediate point from TO to T during load runback, but if we consider the progress of load runback before and after that point, The concerns mentioned above remain, and it cannot be said to be a sufficient measure.

Furthermore, even if the number of load detection points in the middle is increased, the conventional load detection relay for load detection has a single general specification and must be installed at each load detection point, so the number of load detection points in the middle is limited in terms of application. In addition, conventional load runbacks require only the first load runback command regardless of the operating status of the plant.
Load runback operations were performed uniformly.

However, generators are not always operating at full load.
The details will be explained later, but the load runback when operating at a load smaller than the full load can be done more slowly than the load runback during full load operation, and there is less strain on the prime mover, so the plant The possibility of tripping can be reduced. However, in circuits using conventional load detection relays, control is only possible by detecting a single fixed load current after a certain period of time, so it is necessary to calculate the optimal setting value according to the operating condition and issue a load runback control command. It was not possible to achieve this function. In view of the above points, the purpose of the present invention is to calculate the optimal control value according to the plant operating state, continuously monitor whether the load runback is progressing as set during the entire load runback process, and It is an object of the present invention to provide a load runback control method using a computer that can issue load runback control commands and trip commands for equipment protection.

First, before describing the embodiments of the present invention, an example of a method for calculating an ideal optimum load reduction rate will be described. If the starting point of load runback in FIG. 1 and the prescribed load value for cooling water cutoff are taken as the origin, the result will be as shown in FIG. 2. Here, the load reduction rate, that is, the state in which the load current of the generator decreases, is a function of time, and when a constant is determined as shown in FIG. 2, it becomes.

When this load current 1 flows through the generator after the start of load runback, the amount of heat H added to the water in the stator winding is determined by R, the resistance of the stator winding, and k1, the coefficient for converting electric power into heat amount.
, if the loss due to the heat radiation and structure of the generator is K2, then
From the following generator design requirements, the full load current L
The load runback completion time limit when 1 is flowing is T.
, so the value of H in the two equations is constant and unique to that generator. In other words, if a heat amount of H or more is applied after the cooling water is cut off, the cooling water in the stator winding starts to boil.

Therefore, if it is predicted that a heat amount of H or more is likely to be added during load runback, it is necessary to operate the load limiter further or, if that does not work, to trip the generator. Furthermore, Equation 2 means that the load runback completion time limit T changes depending on the magnitude of the load current 1 flowing through the generator immediately before the cooling water is cut off.

In other words, if the load current immediately before the cooling water is cut off is smaller than the full load current, it will take a longer time to run back the load. Load runback completion time limit T
is obtained from the following formula. 11 ↓ Moreover, the ideal optimum load reduction rate K according to the operating condition can be obtained as follows.

H Next, in order to obtain the actual reduction rate k of the load current, ΔI/Δt can be calculated by setting Δ to the current change amount during the minute time Δt.

That is, when the load current at time t is n, and the load current after Δt is 11+1, the actual load current reduction rate k is as follows.

Therefore, by comparing the value of K in equation 4 and the value of k in equation 5 every Δt, it is possible to monitor whether the actual load reduction rate is progressing at an ideal rate.

When α is the margin value, when K-α≦k≦K+α, it is determined that the load runback is operating normally, and k>K+
In the case of α, the load runback is too strong, so the load limiting operation is stopped, and if k<K-α, the load runback must be made into a force, so the load runback is further strengthened. If k<K-α still exists after a certain period of time, it means that the load runbatter is not operating normally for some reason, so a trip signal is issued to safely protect the equipment. Next, an apparatus for implementing this method will be described.

Figure 3 shows the outline. That is, when the generator 1 is driven and operated by the prime mover 2 and a failure detection device 3 such as a pressure switch or a temperature switch detects a failure detection device 3 such as a pressure switch or a temperature switch, the load runback control device 4 activates the CT5. Based on the current load current obtained through
The main steam amount to be reduced by a regulating valve 7 or a trip signal is issued. The load runback control device 4 can be obtained by programming and operating part of the functions of a computer such as an electronic computer used for controlling a power plant or a so-called microcomputer according to the flowchart shown in FIG. .

Next, the contents of the flowchart shown in FIG. 4 will be explained.

(1) Start the program, define n as 1, read the load current, and store the value as 1n.

(2) Based on the read In, calculate the optimum load reduction rate K using the 4 formula and store the value, and calculate the load runback completion time limit T using the 3 formula and store it.

(3) Sampling time Ts if there is no cooling water cutoff failure
The calculations to obtain the values of K and T are repeated at a cycle of , and the memory is updated each time. As a result, the optimum load reduction rate and load runback completion time limit T corresponding to the operating state immediately before the failure are calculated and stored. (4) In the event of a cooling water cutoff failure, a load runback signal is sent to the first stage load limiter.

(5) Count the time Δt to find the actual rate of decrease in load current.

(6) In the subsequent calculations, it is determined whether the load runback completion time limit T has passed.

If it is within T, the program proceeds to the next program for comparing the reduction rates; if it exceeds T, it means that the load runback has ended, and the program is stopped. (7)
The load current after nΔt is read and set as In+1.

(8) The load current value before Δt is stored as In. The actual load reduction rate k is calculated using equation 5. (9)
The actual load reduction rate k and the optimal load reduction rate K are compared.

If k>K+α, there is no need to further operate the load limiter, so the operation of the load limiter is stopped, n is redefined as n+1, and further monitoring is performed. If 0Ik>K+α, it is determined whether k<K−α.

If k<K-α, then K-α≦k>K+α, and the load runback is operating normally.
The variable m is reset to 1 and n is redefined as n+1, and the process of calculating and comparing the load reduction rate after Δt is repeated. If k<K-α, it is necessary to further operate the load limiter to make the load reduction stronger, so a signal is sent to the second or third load limiter.

Compare m△t to see if the state of ADk<K-α is within the normal response time TR of the plant for load runback, and calculate T・8
If it is within, redefine m as m+1 and n as n+
1 and verify the load reduction rate after the next Δt.

Q]) When k<Kb, it is determined that the rate of change is when the load runback is not operating normally for some reason, and a trip signal is immediately issued.

′ (v.) If there exists a state where k<K-α, K>K-α,
Even if TR is exceeded, the load runback is not operating normally for some reason, so a plant trip signal is immediately issued and the program is also stopped.

Q3) When K - α, there is almost no load runback operation, so a trip command is immediately issued to the power generation plant and the program is also stopped.

According to the present invention described above, by calculating the optimal load reduction rate K according to the plant operating state at the time of occurrence of the refrigerant cutoff failure and activating the load limiter 6 in multiple stages,
Compared to the conventional method, it is no longer necessary to force the plant more than necessary, so the success rate of load runback can be increased.

In addition, since it is possible to constantly monitor whether or not the load runback is progressing ideally, the monitoring ability is greatly improved compared to conventional technology.・
Predictive control such as issuing trip commands becomes possible, allowing equipment to be protected more quickly and safely. Although the embodiments described above are for cooling the stator windings with water, they can also be applied to generators that use other cooling media, such as hydrogen gas, instead of water cooling. .

In addition, in the above embodiment, the load runback completion time limit T optimal load reduction rate K and k are obtained by sampling the load current.
The power of the generator may also be sampled using a current transformer (CT) to determine T, K, and k.

[Brief explanation of the drawing]

Figure 1 is a diagram explaining the progress of load runback, Figure 2
The figure shows the change in time and load current obtained from Figure 1, Figure 3 is a conceptual diagram of a system to which the method of the present invention is applied, and Figure 4 is the flow of a program used in an embodiment of the method of the present invention. It's a chat.

Claims (1)

[Scope of Claims] 1. A load runback control method using a computer for a power generation blunt, which includes the following steps, in which the generator output is runback when the cooling device of the generator breaks down. (b) Calculate the optimal load reduction rate K and load runback completion time limit T according to the magnitude of the output at that time from the heat capacity specific to the generator and the sampled value of the output electricity amount of the generator; ) Give a load runback command to the prime mover due to a failure of the generator cooling system, (c) Calculate the actual load reduction rate from the sampled value of the generator output electricity obtained during the load runback operation, and (d) A computer-based load runback control method for a power generation plant, which detects whether or not the optimum load reduction rate K and the actual load reduction rate K have a scheduled relationship, and continues or stops load runback operation. . 2. In the method according to claim 1, the optimum load reduction rate K and the actual load reduction rate k are such that α>IK−
A computer-based load runback control method for a power generation plant that continues load runback when kI (α: margin value). 3. A power generation plant calculator that temporarily suspends load run rack operation when the optimum load reduction rate K and the actual load reduction rate k are α<k-K in the method set forth in claim 1. load runback control method. 4. In the method set forth in claim 1, when the optimum load reduction rate K and the actual load reduction rate k satisfy k<K-α, a stronger load runback operation is performed. Computer-based load runback control method for power generation plants. 5. In the method according to claim 1, when the optimum load reduction rate K and the actual load reduction rate k are equal to or longer than the scheduled time k<K-α, a trip command is issued to the power generation plant. A computer-based load runback control method for power generation plants. 6. A load runback control method using a computer for a power generation plant, in which a trip command is immediately issued to the power generation plant when the optimum load reduction rate k is k≪K−α in the method set forth in claim 1. .