Adjusting method for combined operation mode of circulating pumps of coastal power plant

Abstract

The invention relates to the technical field of machine learning, in particular to an adjusting method for a coastal power plant circulating pump combined operation mode. One circulating water pump has two modes of high-speed operation and low-speed operation, if one unit is equipped with a plurality of circulating water pumps, a plurality of circulating pump combined operation modes are provided, and the amount of circulating water passing through the circulating water pump and the power consumption of the circulating water pump in each operation mode are fixed values; the circulating water flow and the circulating pump power consumption of the unit can be determined after the circulating pump operation mode of the unit is determined. After the seawater temperature is predicted and the load of the unit is determined, the output and net power of the unit can be calculated. And under the condition that the steam admission amount of the unit and the seawater temperature are fixed, each circulating pump combination operation mode corresponds to one unit net power, and the unit net powers under the multiple circulating pump combination operation modes are compared, so that the operation mode with the maximum net power is the optimal circulating pump combination operation mode under the working condition. And the optimal circulating pump combination operation mode at the future moment is determined according to the seawater temperature predicted value and the unit load.

Description

Technical Field

[0001] This invention relates to the field of machine learning technology, and in particular to a method for regulating the combined operation mode of circulating pumps in a coastal power plant. Background Technology

[0002] Circulating water pumps are crucial components of a power plant's cold-end system, and the circulating water flow rate is a key indicator directly impacting the economic efficiency of generator unit operation. For units cooled by seawater, seawater temperature also significantly affects the unit's economic operation. When determining the circulating water pump operating mode, both the unit's electrical and thermal load requirements and seawater temperature variations must be considered. Typically, a single unit is equipped with multiple circulating water pumps, resulting in various pump combination operating modes. To maintain economical unit operation, the circulating water pump operating mode and circulating water flow rate need to be adjusted according to specific operating conditions. Insufficient circulating water flow rate leads to high unit back pressure and increased unit heat rate, while excessively high flow rate results in higher power requirements for the circulating water pumps, leading to a decrease in net power output while maintaining unit output, thus worsening the unit's economic efficiency.

[0003] The experiential circulating water pump operation mode scheduling curves provided by manufacturers are usually not the most economical circulating pump operation mode for the unit at a certain seawater temperature, and they cannot provide an accurate time for switching operation modes. They usually do not take into account the unit load and environmental parameters at future moments, and the adjustment has a lag.

[0004] Therefore, there is an urgent need for a regulation method for the combined operation of circulating pumps in coastal power plants to solve the above-mentioned technical problems. Summary of the Invention

[0005] This invention provides a method for adjusting the combined operation mode of circulating pumps in coastal power plants. By predicting the seawater temperature and the circulating water pump operation mode that maximizes the net power of the unit under that temperature and load in advance, and adjusting it at an appropriate time, the unit can be kept in the most economical operating condition.

[0006] This invention provides a method for adjusting the combined operation mode of circulating pumps in a coastal power plant, comprising:

[0007] By predicting seawater temperature, the optimal circulating pump combination operation mode is determined at every moment through iterative optimization. By considering pipeline dynamics and condenser dynamic response time, various advance adjustment strategies are explored and compared to find the optimal operating condition point and obtain the optimal adjustment time. This allows the circulating water pumps to always operate in the most economical way during the unit's operation, ultimately achieving advanced regulation.

[0008] Seawater temperature prediction models include:

[0009] Forget gate: determines which information is discarded or retained from the cell state;

[0010] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0011] Input gate: determines which new information is updated in the cell state;

[0012] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0013]

[0014] Cell state update: Update the cell state by combining the decisions of the forget gate and the input gate;

[0015]

[0016] Output gate: Determines which information based on the cell state will be output to the next hidden state.

[0017] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0018] h t =o t ×tanh(C t )

[0019] Where: σ is the sigmoid activation function; × represents element-wise multiplication; W and b are the weights and biases.

[0020] As can be seen from the above scheme, the method for adjusting the combined operation mode of circulating pumps in coastal power plants provided by this invention allows for two operating modes for a single circulating water pump: high-speed operation and low-speed operation. If a unit is equipped with multiple circulating water pumps, there are multiple combined operation modes. The circulating water volume and power consumption of the circulating water pumps are constant under each operating mode. That is, after determining the unit's combined operation mode, the unit's circulating water flow rate and power consumption can be determined. After predicting the seawater temperature and determining the unit load, the unit output and net power can be calculated. With a fixed steam inlet and seawater temperature, each combined operation mode corresponds to a unit net power. By comparing the unit net power under multiple combined operation modes, the operating mode that maximizes the net power is the optimal combined operation mode for that condition. Based on the predicted seawater temperature and unit load, the optimal combined operation mode for future times is confirmed. If the current operating mode is no longer the optimal mode for the next time and a switch is required, the optimal adjustment time corresponding to the maximum net power is determined by calculating the unit net power obtained at each time. This allows for the prediction of seawater temperature to determine the optimal circulating pump combination operation mode at every moment in advance, and to calculate the optimal adjustment time when it is necessary to switch the circulating pump operation mode, thereby ensuring that the circulating water pump always operates in the most economical way during the unit's operation. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A schematic flowchart illustrating the adjustment method for the combined operation mode of circulating pumps in a coastal power plant, provided in an embodiment of the present invention.

[0023] Figure 2 A trend diagram of the net power of a 450MW unit as a function of circulating water temperature is provided for an embodiment of the present invention.

[0024] Figure 3 An optimized operating curve of the circulating water pump provided in an embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] Please see Figure 1 This invention provides a method for adjusting the combined operation mode of circulating pumps in a coastal power plant, the method comprising:

[0027] By predicting seawater temperature, the optimal circulating pump combination operation mode is determined at every moment through iterative optimization. By considering pipeline dynamics and condenser dynamic response time, various advance adjustment strategies are explored and compared to find the optimal operating condition point and obtain the optimal adjustment time. This allows the circulating water pumps to always operate in the most economical way during the unit's operation, ultimately achieving advanced regulation.

[0028] Seawater temperature prediction models include:

[0029] Forget gate: determines which information is discarded or retained from the cell state;

[0030] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0031] Input gate: determines which new information is updated in the cell state;

[0032] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0033]

[0034] Cell state update: Update the cell state by combining the decisions of the forget gate and the input gate;

[0035]

[0036] Output gate: Determines which information based on the cell state will be output to the next hidden state.

[0037] o t =σ(W o ·[h t-1 ,x t ]+bo )

[0038] h t =o t ×tanh(C t )

[0039] Where: σ is the sigmoid activation function; × represents element-wise multiplication; W and b are the weights and biases.

[0040] In this embodiment, a circulating water pump has two operating modes: high-speed and low-speed. If a unit is equipped with multiple circulating water pumps, there are multiple circulating pump combination operating modes. The circulating water volume and power consumption of the circulating water pumps are constants under each operating mode. That is, once the circulating pump operating mode of the unit is determined, the circulating water flow rate and circulating pump power consumption can be determined. After predicting the seawater temperature and determining the unit load, the unit output and net power can be calculated. With a constant steam inlet and seawater temperature, each circulating pump combination operating mode corresponds to a unit net power. By comparing the unit net power under multiple circulating pump combination operating modes, the operating mode that maximizes the net power is the optimal circulating pump combination operating mode for that operating condition. Based on the predicted seawater temperature and unit load, the optimal circulating pump combination operating mode for future times is determined. If the current operating mode is no longer the optimal operating mode for the next time time and a switch is required, the optimal adjustment time corresponding to the maximum net power is determined by calculating the unit net power obtained from the adjustment at each time time. This allows for the prediction of seawater temperature to determine the optimal circulating pump combination operation mode at every moment in advance, and to calculate the optimal adjustment time when it is necessary to switch the circulating pump operation mode, thereby ensuring that the circulating water pump always operates in the most economical way during the unit's operation.

[0041] In seawater temperature time series prediction, LSTM models can learn and remember patterns in long-term series. Due to their unique structure, LSTMs can reduce the gradient vanishing problem during backpropagation. In the modeling process of seawater temperature prediction models based on LSTM, model parameter tuning and optimization are the main tasks. This can be achieved by adjusting the learning rate or using an exponential learning rate decay strategy to avoid the model getting trapped in local optima. The exponential learning rate decay formula is:

[0042] Learning rate = initial learning rate × e -衰减率×当前周期数

[0043] The following section uses a newly built coastal power plant as an example to introduce a specific technical solution implementation.

[0044] The power plant's two generating units are equipped with a total of four circulating water pumps, each capable of switching between high-speed and low-speed operation. The following analysis examines the trend of net power output as a function of circulating water temperature under the same steam inflow near the 450MW operating condition. Figure 2 As shown. Therefore, the following results are obtained:

[0045] The optimal operating mode for the circulating water pumps that maximizes the net power of the units varies depending on the seawater temperature. When the seawater temperature is between 5℃ and 15℃, the most economical operating mode is two units operating two circulating water pumps. When the seawater temperature is between 15℃ and 22.5℃, the most economical operating mode is two units operating three circulating water pumps. When the seawater temperature is above 22.5℃, the most economical operating mode is two units operating four circulating water pumps.

[0046] The most economical combination operation mode of circulating water pumps under different seawater temperatures and loads, such as Figure 3 As shown, when the unit load is 300MW, the circulating pump combination operation mode is switched at 15℃ and 27.5℃, and when the unit load is 600MW, the circulating pump combination operation mode is switched at 11℃ and 17.5℃.

[0047] Table 1 Comparison of Net Power of Advanced Adjustment in Combined Operation Modes of Circulating Water Pumps

[0048]

[0049] Table 1 shows the net power and optimal adjustment time for the two hours before and after switching the circulating pump combination operation mode near the 450MW operating condition. Comparing the net power of the unit with advanced adjustment and the unit without advanced adjustment, it can be seen that the net power of the unit with advanced adjustment based on the seawater temperature prediction to calculate the optimal adjustment time is higher than that without advanced adjustment, thus improving the economic operation level of the unit.

[0050] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is only used to illustrate the technical solution of the present invention, and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for regulating the combined operation of circulating pumps in a coastal power plant, characterized in that, include: By predicting seawater temperature, the optimal circulating pump combination operation mode is determined at every moment through iterative optimization. By considering pipeline dynamics and condenser dynamic response time, various advance adjustment strategies are explored and compared to find the optimal operating condition point and obtain the optimal adjustment time. This allows the circulating water pumps to always operate in the most economical way during the unit's operation, ultimately achieving advanced regulation. Seawater temperature prediction models include: Forget gate: determines which information is discarded or retained from the cell state; f t =σ(W f ·[h t-1 ,x t ]+b f ) Input gate: determines which new information is updated in the cell state; i t =σ(W i ·[h t-1 ,x t ]+b i ) Cell state update: Update the cell state by combining the decisions of the forget gate and the input gate; Output gate: Determines which information based on the cell state will be output to the next hidden state. the t =σ(W o ·[h t-1 ,x t ]+b o ) h t =o t ×tanh(C t ) Where: σ is the sigmoid activation function; × represents element-wise multiplication; W and b are the weights and biases.

Images