CN-122019938-A - Method for calculating dust accumulation of photovoltaic power station

CN122019938ACN 122019938 ACN122019938 ACN 122019938ACN-122019938-A

Abstract

The application discloses a method for calculating dust accumulation of a photovoltaic power station, and relates to the technical field of data processing of photovoltaic systems. The method comprises the steps of obtaining a PM2.5 concentration value, an ambient temperature and an ambient humidity of a photovoltaic power station at the current time, obtaining the latest cleaning time of the photovoltaic power station, constructing a dust accumulation dynamic model which is a differential equation model based on physics and used for describing the process of dynamic change of a dust shielding coefficient along with the time, and calculating the dust shielding coefficient at the current time according to the PM2.5 concentration value, the ambient temperature, the ambient humidity and the cleaning time through the dust accumulation dynamic model. The application can accurately calculate and monitor dust accumulation of the photovoltaic power station in real time and at low cost.

Inventors

ZHENG QINGYUE
CAO LEI
Deng Hanbiao
YANG MINGJING
ZHANG TINGKAI
LI ZHICAI
XIE HAOLIN
HE JIASHENG
HUANG ZHONGBIN
Wei Haimei
SHEN YUBIN

Assignees

广州南方电力集团科技发展有限公司

Dates

Publication Date: 20260512
Application Date: 20251226

Claims (10)

1. A method for calculating dust accumulation in a photovoltaic power plant, comprising the steps of: acquiring a PM2.5 concentration value, an ambient temperature and an ambient humidity of a photovoltaic power station at the current time; obtaining the latest cleaning time of a photovoltaic power station; constructing a dust accumulation dynamic model, wherein the dust accumulation dynamic model is a differential equation model based on physics and is used for describing the process of dynamic change of a dust shielding coefficient along with time; and calculating a dust shielding coefficient at the current moment through the dust accumulation dynamic model according to the PM2.5 concentration value, the ambient temperature, the ambient humidity and the cleaning time.
2. The method of claim 1, wherein the functional expression of the dust accumulation dynamic model is: Wherein p_rising (t) represents a dust shielding coefficient at the current t moment, p_clean is a dust shielding coefficient reference value, k is a station pollution sensitivity coefficient, PM2.5 (tau) is a PM2.5 concentration value at the tau moment, alpha is a nonlinear accumulation index, lambda () is a dust deposition attenuation constant which changes with time, and t0 is the latest cleaning time.
3. The method according to claim 2, wherein the dust deposit decay constant λ () is a function of the ambient temperature T and the ambient humidity RH, and is related by: Wherein lambda 0 is a dust deposition base attenuation constant, which represents the proportion of settled dust blown away by natural wind every day under standard conditions, beta T is a temperature compensation coefficient, beta RH is a humidity compensation coefficient, and T ref and RH ref are reference temperature and reference humidity respectively.
4. The method according to claim 2, characterized in that the value of the nonlinear accumulation index α is greater than 1.
5. The method according to claim 4, characterized in that the method further comprises the step of: and carrying out self-learning update on the site pollution sensitivity coefficient k, wherein the self-learning update is triggered after the photovoltaic power station is cleaned.
6. The method according to claim 5, wherein the self-learning update is specifically: According to the shielding coefficient p actual of the photovoltaic module obtained through actual measurement after cleaning, the predicted value p clean of the shielding coefficient of the dust obtained through calculation of the dust accumulation dynamic model after cleaning and the predicted value p predicted of the shielding coefficient of the dust obtained through calculation of the dust accumulation dynamic model according to k n before cleaning, updating the k value according to the following formula by adopting a gradient descent method: where k n is the site pollution sensitivity coefficient before updating, k n+1 is the site pollution sensitivity coefficient after updating, and η is the learning rate.
7. The method of claim 6, wherein the loss function of the self-learning update of the site contamination sensitivity coefficient k is: in the self-learning updating process of site pollution sensitivity coefficient k to minimize For the purpose, the new k value is shifted in the negative gradient direction of the loss function.
8. The method of claim 7, wherein the site contamination sensitivity k has a value in the range of 0.01 to 0.05 (μg/m 3) - 1.
9. The method according to claim 2, wherein the dust blocking coefficient p_blocking (t) is reset to the dust blocking coefficient reference value p_clean every time after cleaning, and the latest cleaning time t0 is updated to the current cleaning time.
10. Method according to claim 2, characterized in that the dust blocking coefficient p_blocking (t) is used for determining an optimal cleaning timing of a photovoltaic power plant and/or for evaluating the loss of power generation due to dust accumulation.

Description

Method for calculating dust accumulation of photovoltaic power station Technical Field The application relates to the technical field of data processing of photovoltaic systems, in particular to a method for calculating dust accumulation of a photovoltaic power station. Background With the acceleration of global energy transformation, the duty ratio of photovoltaic power generation in a global energy structure continues to increase rapidly due to the clean and renewable characteristics of the photovoltaic power generation. Grid-connected operation of large-scale photovoltaic power stations has become a normal state. The photovoltaic power generation has remarkable intermittence and volatility, and the output power of the photovoltaic power generation is directly influenced by real-time changes of meteorological factors such as solar irradiance, ambient temperature, cloud cover, humidity, wind speed, atmospheric pollution degree and the like. The dust of the photovoltaic module can shield light, so that the photovoltaic panel cannot fully absorb solar energy, a shadow effect is caused, and the generated energy can be obviously reduced. Studies have shown that severe dust accumulation can lead to power losses of up to 15% -30% even higher in sand storm frequent areas. By accurately calculating the efficiency loss caused by dust accumulation, the operation and maintenance team can scientifically determine the optimal cleaning time. The waste of water resources and manpower caused by too early cleaning can be avoided, and the loss of generating income caused by too late cleaning can be avoided. The dust accumulation is accurately calculated and monitored, and the method is a key ring for improving the yield of the photovoltaic power station, guaranteeing the safety and realizing efficient operation. In the conventional model, the dust blocking coefficient is a static constant that needs to be set manually and periodically. This means that the model completely disregards the continuous accumulation of dust in the environment from one wash to the next, resulting in a linear increase of prediction error over time. The assessment and management of dust accumulation in photovoltaic power plants is mostly in a passive, hysteretic, rough state. Depending on either expensive hardware or human experience, or using stiff statistical models, accurate, real-time, low-cost sensing and prediction is difficult to achieve. Disclosure of Invention Based on this, the object of the present invention is to solve the above-mentioned technical problems, enabling accurate, real-time and low-cost calculation and monitoring of dust accumulation in photovoltaic power stations. To achieve the above object, the present application provides a method for calculating dust accumulation in a photovoltaic power station, comprising: acquiring a PM2.5 concentration value, an ambient temperature and an ambient humidity of a photovoltaic power station at the current time; obtaining the latest cleaning time of a photovoltaic power station; constructing a dust accumulation dynamic model, wherein the dust accumulation dynamic model is a differential equation model based on physics and is used for describing the process of dynamic change of a dust shielding coefficient along with time; and calculating a dust shielding coefficient at the current moment through the dust accumulation dynamic model according to the PM2.5 concentration value, the ambient temperature, the ambient humidity and the cleaning time. Preferably, the functional expression of the dust accumulation dynamic model is: Wherein p_rising (t) represents a dust shielding coefficient at the current t moment, p_clean is a dust shielding coefficient reference value, k is a station pollution sensitivity coefficient, PM2.5 (tau) is a PM2.5 concentration value at the tau moment, alpha is a nonlinear accumulation index, lambda () is a dust deposition attenuation constant which changes with time, and t0 is the latest cleaning time. Preferably, the dust deposit attenuation constant λ () is a function of the ambient temperature T and the ambient humidity RH, and has the relation: Wherein lambda 0 is a dust deposition base attenuation constant, which represents the proportion of settled dust blown away by natural wind every day under standard conditions, beta T is a temperature compensation coefficient, beta RH is a humidity compensation coefficient, and T ref and RH ref are reference temperature and reference humidity respectively. Preferably, the value of the nonlinear accumulation index α is greater than 1. Preferably, the method further comprises the step of performing self-learning update on the site pollution sensitivity coefficient k, wherein the self-learning update is triggered after the photovoltaic power station is cleaned. Preferably, the self-learning update is specifically: According to the shielding coefficient p actual of the photovoltaic module obtained through actual measurement afte