TW201727559A - Management method and system of renewable energy power plant checking whether the power generation of a renewable energy power plant is normal according to the estimated power generation amount - Google Patents

Abstract

The present invention discloses a management method and system for a renewable energy power plant comprising the steps of: selecting a plurality of influence factors of power generation of a renewable energy power plant; collecting multiple power plant data and multiple historical power generation data corresponding to the influence factors of power generation in a first historical time interval; establishing a power generation estimation model according to the power plant data and the historical power generation data; calculating a first estimated power generation amount in a second historical time interval using the power generation estimation model; obtaining an actual power generation amount of the renewable energy power plant in the second historical time interval; comparing the first estimated power generation amount with the actual power generation amount and obtaining a power generation difference; and when the power generation difference is greater than an error tolerance, an abnormal alarm is issued. Through the above method, whether the power generation of the renewable energy power plant is normal can be checked according to the estimated power generation amount.

Description

Management method and system of renewable energy power plant

This creative department is concerned with a management method and system for a renewable energy power plant, in particular, a management method and system for a renewable energy power plant with application data exploration technology.

Renewable energy is an inexhaustible source of energy from nature, such as solar energy, wind power, tidal energy, geothermal energy, biomass energy or water power, and is not limited here. However, the amount of electricity generated by renewable energy power generation systems will be affected by many external factors, such as climate temperature, solar intensity, or seasonal factors. These external factors affect the power generation of renewable energy power generation systems to varying degrees, and renewable energy power generation systems. It can be regarded as an unstable power source. Therefore, it is important to study the randomness of renewable energy and the estimation technology of renewable energy power generation.

However, the shortcomings such as randomness, volatility, intermittentness and uncertainty of power generation in renewable energy power plants have caused certain difficulties in estimating renewable energy power generation. The difference between the estimation results and the actual results is too large, resulting in no reference value for the estimation results.

Therefore, there is a need to design a management method and system for a renewable energy power plant. Through data analysis and exploration techniques, a power generation prediction model is first established through past historical power generation data, climate data, and time data to generate a power generation estimation model for reference. According to the power generation estimate, the staff can estimate the amount of power generated by the historical time of the renewable energy power plant, and then determine whether the power generation of the renewable energy power plant is normal or the power supply mode of the power generation is deployed.

The purpose of this creation is to provide a management method for a renewable energy power plant, through which the power generation amount of the renewable energy power plant can be estimated, and the estimated power generation amount and the actual power generation amount are compared to confirm whether the power generation of the renewable energy power plant is normal.

According to the above purpose, the present invention provides a management method for a renewable energy power plant, comprising the following steps: selecting a plurality of power generation quantity influence factors of a renewable energy power plant; collecting a plurality of power generation quantity influence factors in a first historical time interval The power plant data and the corresponding plurality of historical power generation data; establishing a power generation estimation model according to the power plant data and the corresponding historical power generation data; using the power generation estimation model to calculate a first historical time interval An estimated power generation amount; obtaining an actual power generation amount of the renewable energy power plant in the second historical time interval; comparing the first estimated power generation amount with the actual power generation amount, and obtaining a power generation difference amount; when the power generation difference amount is greater than one The error tolerance value gives an abnormal warning.

Another purpose of this creation is to provide a management system for a renewable energy power plant. Through the management system of the renewable energy power plant, it is possible to verify whether the actual power generation amount is normal or not, and can also be used to estimate the amount of power generation in the future, thereby enabling power transmission and distribution. Provisioning.

According to the above purpose, the present invention provides a management system for a renewable energy power plant, comprising: a selection module for selecting a plurality of power generation quantity influence factors of a renewable energy power plant; a database for storing the influence of the power generation amounts a plurality of power plant data of the factor and the corresponding plurality of historical power generation data; a data collection module connecting the selection module and the database, and collecting the power plants corresponding to the power generation influence factors from the database A power generation estimation model is connected to the data collection module, and the power generation quantity estimation model is trained by the power generation data of the power generation influence factors and the corresponding historical power generation data, and an estimated power generation amount can be output. A comparison module for comparing the difference between an actual power generation amount and the estimated power generation amount.

In addition to the difference between the actual power generation of the renewable energy power plant and the estimated power generation, it can be used to compare whether the power generation of the renewable energy power plant is abnormal. In addition, the management system of the renewable energy power plant of the present invention establishes a power generation estimation model based on power plant data (such as various power generation data or environmental data, etc.) in a historical time interval, and the power plant data is closely related to the power generation capacity of the renewable energy power plant. Let the factors such as the randomness, volatility, intermittentness and uncertainty of the power generation of the renewable energy power plant can be considered, and the power generation of the highly accurate renewable energy power generation can be estimated, so that the difference between the estimated result and the actual result is compared. small.

Figure 1 is a block diagram of the management system of the regenerative power plant of the present invention. As shown in FIG. 1 , the management system 10 of the renewable energy power plant includes a selection module 11 , a data collection module 12 , a database 13 , a power generation estimation model 14 and a comparison module 15 . The renewable energy power plant is a solar power plant, a wind power plant, a tidal power power plant, a geothermal power plant, a biomass power plant or a hydropower plant.

In the management system 10 of the regenerative power plant of the present invention, the selection module 11 is configured to select a power generation amount influence factor of the renewable energy power plant 16, in other words, the selection module 11 is configured to select a power generation influence factor that affects the power generation amount, These power generation impact factors are the types of power plant data for renewable energy power plants. The selection module 11 can be an operation interface of the electronic device, and the worker selects a power generation amount influence factor for estimating the power generation amount through the selection module 11. Different workers can choose different power generation impact factors. The data collection module 12 is connected to the selection module 11 and the database 13. According to the power generation influence factor selected by the selection module 11, the data collection module 12 extracts the power generation data of the power generation influence factor from the database 13 (for example, power generation). Data or environmental data), these power plant data are the power plant data of the renewable energy power plant 16 , which are automatically taken or remitted by the power plant monitoring equipment to automatically capture or link external data (such as weather station climate data or other climate data providing units) Etc.) or the staff records them in the daily power plant management work and stores them in the database 13.

In addition, the data collection module 12 can also be used to collect climate data. The data collection module 12 can connect to a weather website through a wired or wireless network to retrieve historical meteorological data or future predicted meteorological data or stretched storage devices. The data collection module 12 is connected to the power generation estimation model 14 and transmits the collected power plant data and the corresponding historical power generation data to the power generation estimation model 14 for training. After the trained model is established, the weather station or other climate is further transmitted. The data providing unit obtains the environmental climate data needed for a period of time in the future, and substitutes it into the established model to obtain the estimated power generation for a period of time in the future. For example, if the renewable energy power plant is a solar power plant, its scale is large or small. For example, a solar power plant can contain only one tandem solar module, and a tandem solar module contains multiple solar modules. In an embodiment, the solar power plant may also include a plurality of tandem solar modules, or multiple solar modules included in one or more Maximum Power Point Tracking (MPPT), or even one or The amount of solar module power that can be received by multiple inverters is not limited here.

Then, the power generation amount estimation model 14 can be used to detect whether the power generation amount of the regenerative power plant is abnormal. For example, if it is necessary to check whether the power generation of the renewable energy power plant is normal yesterday, the estimated power generation is calculated from the power generation estimation model 14 and the known climate data yesterday from the historical climate data and power generation data required by one of the historical time before yesterday. And comparing the estimated power generation amount with the actual power generation amount obtained from the database 13 by the comparison module 15, and if the difference between the estimated power generation amount and the actual power generation amount is greater than an error tolerance value, the regeneration is performed. The management system 10 of the energy power plant outputs a warning signal to inform the staff of the renewable energy power plant that the power generation yesterday is abnormal and needs further testing. In addition, the power generation estimation model 14 can also substitute the climate prediction data for a certain period of time into the established model, and calculate the estimated power generation amount for a certain period of time. According to the estimated power generation amount, the staff can allocate the power supply and distribution mode of the area to Avoid the problem of insufficient or excess power supply in the region due to unstable power generation output from renewable energy power plants. It should be noted here that each time period, the power generation amount estimation model 14 must be retrained, and the power generation amount estimation model 14 is updated through retraining. For example, if the renewable energy power plant is a solar power plant, because the solar module's polycrystalline silicon, single crystal germanium or other materials will reduce the conversion efficiency over time, the power generation estimation model 14 needs to be updated at intervals. In order to avoid the occurrence of abnormal conditions in the solar power plant, the power generation estimated by the power generation estimation model 14 is still too large from the actual power generation, resulting in a false positive.

Figure 2 is a flow chart showing the steps of the management method of the regenerative power plant of the present invention. As shown in FIG. 2, in step S201, a plurality of power generation amount influence factors of a regenerative power plant 16 are selected through the selection module 11. There are many factors affecting the amount of power generation in renewable energy power plants16. These power generation factors include time and environmental data. The environmental data can be wind speed, wind direction, temperature, sunshine, humidity, rainfall or cloud cover, air pressure or ground temperature. (soil temperature), etc., is not limited here, and the power generation data may be direct current (DC) data of the renewable energy power plant 16, such as voltage, current, and regenerative power generation efficiency (kWh/kWp/h), kilowatts (kW) , per kilowatt hour (kWh) or power per unit (kWh/kWp), or alternating current (AC) data, such as voltage, current, renewable energy generation efficiency (kWh/kWp/h), kilowatts (kW) , per kilowatt hour (kWh) or power generation per unit (kWh/kWp), etc., is not limited here.

In step S202, the plurality of power plant data of the power generation quantity influence factor in the first historical time interval and the corresponding plurality of historical power generation data are collected from a database 13 through a data collection module 12. Because these power generation impact factors are factors affecting the power generation of renewable energy power plants, the power plant data of these power generation impact factors and the corresponding historical power generation data are collected, and these power plant data are further analyzed, data and summarized. analysis. In step S203, a power generation amount estimation model 14 is established based on the power plant data and the historical power generation data. The power generation estimation model 14 may be a computer module including a data analysis and data exploration algorithm, and input power plant data of the power generation influence factor to the power generation estimation model 14, using multiple power plant data and multiple historical power generation data. The power generation amount estimation model 14 is trained into an optimal power generation amount estimation model 14. The data analysis and data mining algorithms used in the power generation estimation model 14 are data exploration using Artificial Neural Networks (NNs), K-Nearest Neighbor (KNN) or Linear Regression. The calculation method of the technology is then repeatedly corrected by the historical climate data of the power generation influence factor of the renewable energy power plant and the corresponding historical power generation data to obtain the optimal power generation estimation model 14 . The data exploration algorithm is a calculation method using data mining techniques such as neural network, nearest neighbor method or linear regression. It has been applied in the fields of stock analysis or meteorological analysis, and is well known to those of ordinary skill in the art, and is applied to regeneration. A reliable power generation estimate can be obtained from the management analysis of the energy power plant.

Next, in step S204, the first estimated power generation amount in a second historical time interval is calculated using the power generation amount estimation model 14. After the power generation amount estimation model 14 is established in step S203, the power generation amount estimation model 14 can be applied to calculate the first estimated power generation amount in the second historical time interval. Then, in step S205, an actual power generation amount of the renewable energy power plant 16 in the second historical time interval is obtained from the database 13, and the first estimated power generation amount in the second historical time interval is estimated to be used for The actual amount of power generation in the second historical time interval is compared. In step S206, the comparison estimated module 15 compares the first estimated power generation amount with the actual power generation amount to obtain a power generation difference amount, and compares the first estimated power generation amount with the actual power generation amount to analyze whether the actual power generation amount is Whether the gap between the first estimated power generation is too large, if the gap is too large, the actual power generation may be abnormal, and the renewable energy power plant 16 may have an abnormality. Therefore, in step S207, when the power generation difference amount is greater than an error tolerance value, an abnormal alarm is issued. News. Through the management method of the renewable energy power plant 16 of the present invention, the estimation result can be compared with the actual result, and then it is detected whether the regenerative power plant 16 is abnormal.

In addition, in the present creation, step S208 is further included, and the power generation amount estimation model 14 is used to collect the plurality of future data of the power generation quantity influence factors in a future time interval to calculate a second estimated power generation amount, and the future data may be It is a climate prediction data that will affect power generation in the future, and these climate prediction data may be wind speed, wind direction, temperature, sunshine, humidity, rainfall or cloud cover, air pressure or ground temperature (soil temperature), etc. Not limited. A second estimated power generation amount is calculated at a future time, and an output power generation amount of the renewable energy power plant 16 is adjusted according to the second estimated power generation amount. For example, if tomorrow's second estimated power generation is larger than the actual power generation of the renewable energy power plant 16 today, the output power generation of the renewable energy power plant 16 is increased, if the second estimated power generation capacity of tomorrow is higher than that of the renewable energy power plant. 16 If the actual power generation is small today, the output power generation of the renewable energy power plant 16 will be reduced, and the power transmission and distribution regulation will be carried out according to the power consumption required tomorrow. The management method of the regenerative power plant 16 of the present invention can be used to compare the estimated power generation amount with the actual power generation amount, and can also estimate the future power generation amount, and refer to the future power generation amount and the future required power consumption, and adjust in advance in the future time. The power generation output of the renewable energy power plant 16 or integration of other power plants for transmission and distribution, allows the manager of the renewable energy power plant 16 to refer to the estimated future power generation, and then pre-adjust the power generation output of the renewable energy power plant 16 or integrate other power plants for transmission and distribution. .

A neural network includes an input layer, a hidden layer, and an output layer. The input layer receives the power generation quantity influence factor of the regenerative power plant and the corresponding data of the corresponding power generation amount (or training data or testing data), and the output layer outputs the training result of the neural network or Calculation results. In the application neural network for power generation estimation model, the power plant data is cut into training data, training data and test data according to the training mode, and then the training data and the training data are used to train and adjust the power generation estimation model. The so-called training is to train the power generation amount estimation model 14 into the optimal power generation amount estimation model 14 based on the calculation method of the error-modified power generation amount estimation model 14 through repeated tests and corrections. Since the technology of the neural network is well known to those of ordinary skill in the art, a detailed description of the neural network is not repeated here.

In addition, the proposed power generation estimation model can also use the nearest neighbor method to estimate future power generation. The nearest neighbor method is calculated based on the weather forecast data of a certain day in the future, and finds the days of multiple (K) similar weathers in the past, and estimates the amount of power generated in the future by the amount of power generated in the past several similar weather days. This method uses multiple meteorological factors, such as solar radiation, temperature or wind speed, to calculate the similarity, determine the K value of the day of similar weather, and then integrate the power generation of multiple (K) days of similar weather. Integration methods include averaging and weighted averaging. The weighted average method sets the weighting value according to the degree of similarity, and the more similar, the larger the weight. Since the techniques of the nearest neighbor method are well known to those of ordinary skill in the art, a detailed description of the nearest neighbor law will not be repeated.

In addition, the proposed power generation estimation model can also use the linear regression prediction model to estimate the future power generation. The linear regression prediction model considers currently known impact factors as well as unknown fixed and random impact factors. There are many factors affecting output power, including design factors, environmental factors, equipment factors, and weather factors. In order to understand the substantive impact of each factor, the number of impact factors or control factors should be reduced as much as possible to avoid interaction effects. The impact factor is random, not long-term, such as temporary shading, wind direction, and dirt. Such effects can be estimated, and since the techniques of linear regression are well known to those of ordinary skill in the art, a detailed description of linear regression will not be repeated.

Through the above description, the calculation method of the data exploration technology such as the neural network, nearest neighbor method or linear regression of the creation application can obtain reliable power generation estimation results. The results of the power generation estimation can be used to compare whether the actual power generation of the renewable energy power plant is different from the estimated power generation, and then whether the power generation of the renewable energy power plant is abnormal, or the future power generation of the renewable energy power plant can be estimated. Adjust the future power output of renewable energy power plants.

10‧‧‧Management system for renewable energy power plants
11‧‧‧Selection module
12‧‧‧ Data Collection Module
13‧‧‧Database
14‧‧‧Power generation estimation model
15‧‧‧Comparative Module
16‧‧‧Renewable Energy Power Plant

Figure 1 is a block diagram of the management system of the regenerative power plant of the present invention. Figure 2 is a flow chart showing the steps of the management method of the regenerative power plant of the present invention.

Claims (10)

A method for managing a renewable energy power plant, comprising the steps of: selecting a plurality of power generation quantity influence factors of a renewable energy power plant; collecting a plurality of power plant materials corresponding to the power generation quantity influence factors in a first historical time interval and corresponding plural numbers a historical power generation data; a power generation quantity estimation model is established according to the power plant data and the corresponding historical power generation data; and the first estimated power generation amount is calculated by using the power generation quantity estimation model in a second historical time interval; An actual power generation amount of the renewable energy power plant in the second historical time interval; comparing the first estimated power generation amount with the actual power generation amount, and obtaining a power generation difference amount; when the power generation difference amount is greater than an error tolerance value, issuing an abnormality Warning. The method for managing a renewable energy power plant according to claim 1, wherein the power generation amount estimation model is established by using a nearest neighbor method, a linear regression, or a type of neural network. The method for managing a renewable energy power plant according to claim 1, further comprising calculating a second estimated power generation amount at a future time by using the power generation amount estimation model, and adjusting the renewable energy power plant according to the second estimated power generation amount One can output power generation. The method for managing a renewable energy power plant according to claim 3, wherein the second estimated power generation amount is calculated by using the power generation amount estimation model to collect a plurality of future data of the power generation quantity influence factors in the future time interval, And these future data are climate prediction data that affect the power generation of the renewable energy power plant. The method for managing a renewable energy power plant as claimed in claim 1 or 2, wherein the power plant data comprises climate data and power generation data. The method for managing a renewable energy power plant according to claim 5, wherein the climatic data includes wind speed, wind direction, temperature, sunshine, humidity, rainfall, cloud cover, air pressure or soil temperature, and the power generation data includes DC power generation or AC power generation. The method for managing a renewable energy power plant according to claim 5, wherein the renewable energy power plant is a solar power plant, a wind power plant, a tidal power power plant, a geothermal power plant, a biomass power plant, or a hydropower plant. A management system for a renewable energy power plant, comprising: a selection module for selecting a plurality of power generation quantity influence factors of a renewable energy power plant; a database for storing a plurality of power plant data and facilities of the power generation quantity influence factors Corresponding plural historical power generation data; a data collection module, connecting the selection module and the database, collecting the power plant data corresponding to the power generation influence factors from the database and corresponding to the history Power generation data; a power generation estimation model is connected to the data collection module, and the power generation data corresponding to the power generation influence factors and the corresponding historical power generation data are trained to output an estimated power generation amount; A module for comparing the difference between an actual power generation amount and the estimated power generation amount. A management system for a renewable energy power plant according to claim 8, wherein the power generation estimation model is established using a nearest neighbor method, a linear regression, or a type of neural network. The management system for a renewable energy power plant according to claim 8, wherein the power plant data includes climatic data and power generation data, and the climatic data includes wind speed, wind direction, temperature, sunshine, humidity, rainfall, cloud cover, air pressure or soil. Temperature, and the power generation data includes DC power generation or AC power generation.