System and method for monitoring and forecasting operation carbon footprint of prefabricated cabin transformer substation

Abstract

The invention discloses a prefabricated cabin transformer substation operation carbon footprint monitoring and forecasting system and method, and the system comprises an energy consumption monitoring module which is used for monitoring and calculating the energy consumption carbon footprint generated by the power utilization of a transformer substation and the energy consumption of a transport vehicle; the SF6 gas monitoring module is used for monitoring greenhouse gas released during operation of the power distribution device and calculating SF6 leakage carbon footprint; the renewable energy compensation module is used for monitoring carbon compensation of a new energy power generation facility matched with the transformer substation to the transformer substation and calculating a renewable energy compensation carbon footprint; the carbon sink absorption module is used for calculating the carbon absorption amount and the carbon sink carbon footprint of green plants and lawns; the sum of the energy consumption carbon footprint, the SF6 leakage carbon footprint, the renewable energy compensation carbon footprint and the carbon sink carbon footprint is the operation carbon footprint of the prefabricated cabin transformer substation; the invention provides two methods for predicting the carbon footprint in the full operation period of the prefabricated cabin transformer substation, namely mean value estimation and model prediction, reference is provided for low-carbon stable operation of the prefabricated cabin transformer substation, and low-carbon transformation of a power grid is assisted.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grids, and in particular relates to a system and method for monitoring and predicting the operation carbon footprint of a prefabricated cabin substation. Background Art

[0002] Carbon emissions from the operation of prefabricated cabin substations are an important part of the carbon emissions of the power system. The operation cycle of prefabricated cabin substations is long, and there are many carbon emission sources involved. It is difficult to monitor the carbon emissions from the operation of prefabricated cabin substations. Tracking and detecting carbon emissions during the operation of prefabricated cabin substations is an important basis for achieving carbon management and carbon evaluation of substations, and is of great significance to helping the low-carbon development of the power grid.

[0003] Existing technologies mainly measure the total carbon emissions of traditional civil substations, without considering the differences in carbon emissions in each year during the operation of the substation. Existing technologies mainly focus on measuring carbon emissions caused by energy consumption such as electricity consumption in substations, and there is little research on carbon emissions from SF6 equipment. For prefabricated cabin substations, the error in calculating the carbon footprint of traditional substations is large. The carbon footprint of a prefabricated cabin substation during operation mainly consists of four parts: one is carbon emissions generated by energy consumption (including air conditioning, heating, fuel consumption of operation and maintenance vehicles, etc.); the second is greenhouse gases directly released by electrical equipment, such as SF6; the third is the compensation for carbon emissions by microgrid systems such as renewable energy power generation and energy storage that are equipped with prefabricated cabin substations. Fourth, the contribution of carbon sinks such as green plants to the carbon footprint of substations.

[0004] Compared with civil substations, the modular design, compact layout and efficient heat dissipation system of prefabricated cabin substations reduce the energy consumption during operation, and improve the reliability of equipment and operation stability. Based on the above characteristics of prefabricated cabin substations, a carbon footprint monitoring and forecasting system for the operation of prefabricated cabin substations is established, which provides effective carbon footprint monitoring technology for the operation of prefabricated cabin substations in multiple scenarios such as urban power grids, industrial parks, and new energy power supply, strengthens the low-carbon operation and maintenance management of substations, and promotes the green and low-carbon development of power grids. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the background technology, provide a prefabricated cabin substation operation carbon footprint monitoring and forecasting system, provide a scientific method for monitoring and evaluating the operation carbon footprint of the prefabricated cabin substation, provide a theoretical basis for calculating the total carbon footprint of the prefabricated cabin substation during operation, improve the effectiveness of carbon footprint monitoring of the prefabricated cabin substation, and promote the low-carbon green development of the power grid. .

[0006] To achieve the above objectives, the technical solution of the present invention is:

[0007] A prefabricated cabin substation operation carbon footprint monitoring and forecasting system, the system comprising:

[0008] Energy consumption monitoring module, used to monitor and calculate the carbon footprint of energy consumption generated by substation electricity consumption and transportation vehicle energy consumption;

[0009] SF6 gas monitoring module, used to monitor greenhouse gases released by power distribution equipment during operation;

[0010] Renewable energy compensation module, used to monitor the carbon compensation of substations by supporting new energy power generation facilities;

[0011] The carbon sink absorption module is used to monitor the carbon absorption of green plants and lawns.

[0012] The sum of the carbon footprint of energy consumption, the carbon footprint of SF6 leakage, the carbon footprint of renewable energy compensation, and the carbon footprint of carbon sink is the operating carbon footprint of the prefabricated cabin substation.

[0013] Furthermore, the energy consumption monitoring module includes an intelligent operation and maintenance management system and a vehicle management platform of the substation;

[0014] The SF6 gas monitoring module includes an SF6 online monitoring alarm system;

[0015] The renewable energy compensation module includes a power grid energy management system;

[0016] The carbon sink absorption module includes a green and lawn planting area acquisition unit.

[0017] A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation operation, the method comprising:

[0018] Construct four carbon footprint monitoring and forecasting system modules: energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module, and carbon sink absorption module;

[0019] The data of the four modules, namely, energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module and carbon sink absorption module, are traced to obtain the annual operating energy consumption data of the prefabricated cabin substation, the greenhouse gas data released by the distribution device during operation, the renewable energy power generation and the green and lawn planting area;

[0020] Construct a carbon footprint factor library and a carbon sink factor library, introduce the global warming potential value GWP to convert the carbon footprint of the prefabricated cabin substation into a carbon dioxide scalar; the product of the operating energy consumption and the energy consumption carbon footprint factor is the energy consumption carbon footprint of the prefabricated cabin substation, the product of the released greenhouse gas data and the global warming potential value GWP value is the SF6 leakage carbon footprint, the product of the renewable energy power generation and the national power grid average emission factor is the negative value of the renewable energy compensation carbon footprint, and the product of the green and lawn planting area and the carbon sink factor is the negative value of the carbon sink carbon footprint;

[0021] The carbon footprint data of each year in the operating life of the prefabricated cabin substation is obtained by summing up the carbon footprint of energy consumption, SF6 leakage, renewable energy compensation and carbon sink in each year of the operating life of the prefabricated cabin substation.

[0022] The carbon footprint data of each year within the operating years is used to predict the carbon footprint of the prefabricated cabin substation in future operating years.

[0023] Furthermore, the power consumption of the air-conditioning system, lighting system, fan, fire protection system, control system and other stations in the operating energy consumption comes from the intelligent operation and maintenance management system of the substation, the fuel consumption data of the operating vehicles is obtained from the vehicle management platform, the released greenhouse gas data is obtained from the SF6 online monitoring and alarm system, the renewable energy power generation comes from the microgrid energy management system, and the planting area of ​​green, lawn, etc. is obtained from on-site measurements of the substation.

[0024] Furthermore, the prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: calculating according to the average carbon footprint of the operating years, taking the average of the operating years as the carbon footprint data for other years, and calculating the total carbon footprint generated by the prefabricated cabin substation during the entire operating period.

[0025] Furthermore, the prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: training an extreme learning machine ELM neural network model through the carbon footprint data within the operating years, predicting the carbon footprint data of other years through the trained ELM neural network model, and calculating the total carbon footprint generated by the prefabricated cabin substation during the entire operating period.

[0026] Furthermore, the carbon footprint factor library includes the average carbon emission factor of the power grid, the carbon footprint factors of gasoline and diesel, and other energy consumption carbon footprint factors; the carbon sink factor library includes the carbon sequestration amount of common vegetation.

[0027] Furthermore, the prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: determining the year with the greatest impact on the carbon footprint of the current year through partial autocorrelation analysis PACF, and after determining the carbon footprint data of the year with the greatest impact on the current carbon footprint, using the carbon footprint data of the year with the greatest impact to train an extreme learning machine ELM neural network model, taking the carbon footprint data of the year with the greatest impact as input, and predicting the carbon footprint data of future operating years through the trained ELM neural network model.

[0028] Furthermore, the method for calculating the autocorrelation coefficient ψ_kk value by the partial autocorrelation analysis PACF is:

[0029]

[0030] in,

[0031]

[0032] For a time series X t , t is the current forecast year, where t = 1, 2, ..., n; k is the year, and k takes values ​​between the years that have been run.

[0033] Furthermore, the ELM neural network model has k nodes in the input layer, m nodes in the hidden layer, and 1 node in the output layer. For a given n samples (x i , t i ), input variable x i =(x i1 , x i2 , …, x ik ) T , t i is the carbon footprint value of the current year, i = 1, 2, ..., n, and the ELM neural network model is expressed as:

[0034]

[0035] Among them, j is the output value of ELM, i.e. the predicted future operation carbon footprint value of the prefabricated cabin substation, g(x) is the Sigmoid activation function, which is the fixed activation function used by the neural network; w i is the weight between the i-th hidden node and the input node, and the corresponding weight matrix is ​​an m×k matrix; b i is the bias of the i-th hidden node, and the corresponding bias matrix is ​​an m×1 matrix; β i is the weight between the i-th hidden node and the output node.

[0036] The beneficial effects of the present invention are:

[0037] 1. The present invention aims at the carbon footprint of the prefabricated cabin substation during its operation period, traces the source of the carbon footprint, deeply analyzes the formation process, data acquisition, and calculation methods of different carbon footprints, and calculates the carbon footprint of the substation during the entire operation period from different sources such as energy consumption, greenhouse gas emission, renewable energy power generation, and carbon sinks, providing a new idea for the carbon footprint management of the prefabricated cabin substation during the operation stage.

[0038] 2. Based on the characteristics of prefabricated cabin substation, the present invention constructs a prefabricated cabin substation operation carbon footprint monitoring and forecasting system, which provides a systematic solution for the measurement of the carbon footprint of the prefabricated cabin substation during operation. It also provides two methods for predicting the carbon footprint of the prefabricated cabin substation during the entire operation period: mean estimation and model prediction. It provides a reference for the low-carbon and stable operation of the prefabricated cabin substation and helps the low-carbon transformation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the process of the present invention.

[0040] Figure 2 This is the operating energy consumption diagram of the prefabricated cabin substation of the present invention.

[0041] Figure 3 This is the energy consumption carbon footprint diagram of the prefabricated cabin substation of the present invention.

[0042] Figure 4 This is the PACF analysis of the carbon footprint of the prefabricated cabin substation of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the invention clearer, the present invention is further described below with reference to the accompanying drawings.

[0044] According to the sources of the carbon footprint of the prefabricated cabin substation operation, the carbon footprint of the prefabricated cabin substation operation includes four parts, namely, the carbon footprint of energy consumption, the carbon footprint of SF6 leakage, the carbon footprint of renewable energy compensation, and the carbon footprint of carbon sink. The sum of the carbon footprint of energy consumption, the carbon footprint of SF6 leakage, the carbon footprint of renewable energy compensation, and the carbon footprint of carbon sink is the carbon footprint of the prefabricated cabin substation operation.

[0045] Regarding the prefabricated cabin substation operation carbon footprint monitoring and forecasting system of the present invention, according to the source of the prefabricated cabin substation operation carbon footprint, four modules are constructed, namely, energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module, and carbon sink absorption module.

[0046] The carbon footprint monitoring and forecasting process of the prefabricated cabin substation operation of the present invention is as follows Figure 1 shown.

[0047] The carbon footprint of the prefabricated cabin substation during its operating life can be expressed as:

[0048]

[0049] Where n is the design operating life, CF E is the carbon footprint of energy consumption for one year, CF S The carbon footprint of SF6 leakage in one year, CF B Compensate for one year of renewable energy carbon footprint, CF H It is the carbon footprint of carbon sink for one year, which is calculated by energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module and carbon sink absorption module respectively.

[0050] The energy consumption monitoring module is used to monitor and calculate the carbon footprint of energy consumption generated by the power consumption of the substation and the energy consumption of the transportation vehicles; the energy consumption monitoring module includes the intelligent operation and maintenance management system of the substation and the vehicle management platform; the power consumption of the air conditioning system, the lighting system, the fan, the fire protection system, the control system and other stations are from the intelligent operation and maintenance management system of the substation; the fuel consumption data of the running vehicles can be obtained from the vehicle management platform. Figure 2 shown.

[0051] In the energy consumption monitoring module, the air conditioning system, lighting system, fan, fire protection system, control system, other station electricity consumption and fuel consumption of operation and maintenance vehicles constitute the operating energy consumption of the prefabricated cabin substation. The product of the operating energy consumption and the carbon footprint factor is the energy consumption carbon footprint of the prefabricated cabin substation. The energy consumption carbon footprint of the prefabricated cabin substation of the present invention is as follows: Figure 3 shown.

[0052] Energy consumption carbon footprint represents the carbon footprint of various energy sources such as electricity and fuel consumed during the operation of the prefabricated cabin substation. The annual energy consumption carbon footprint CF E Calculate as follows:

[0053] CF E =(E k +E z +E f +E x +E c +E q )·F e +o q ·F q +o c ·F c +o r ·F r

[0054] Among them, E k is the annual air conditioning energy consumption (cooling and heating), E z The annual electricity consumption for lighting, E f is the annual electricity consumption of the fan, E x The annual electricity consumption of the fire protection system, E c is the annual power consumption of the control system, E q For the annual electricity consumption of other stations, q is the annual fuel consumption (gasoline) of operation and maintenance vehicles, o c The annual fuel consumption of operation and maintenance vehicles (diesel), o r F is the annual fuel consumption of operation and maintenance vehicles (other fuels); e is the average emission factor of the power grid, F q is the gasoline carbon footprint factor, F c is the diesel carbon footprint factor, Fr is the carbon footprint factor for other fuels.

[0055] Taking a 110kV prefabricated cabin substation as an example, the power consumption of air conditioning, lighting, fans, fire protection systems, control systems, other stations and fuel consumption of operation and maintenance vehicles in a certain year are shown in Table 1.

[0056] Table 1 Energy consumption table of a 110kV substation

[0057]

[0058]

[0059] In recent years, the average emission factor of the power grid has been reduced from 0.6101kgCO2 / kWh in 2015 to 0.5568kgCO2 / kWh, and the carbon emission reduction effect of the power grid has initially emerged. The average emission factor of the national power grid is 0.5568kgCO2 / kWh, which is used as the basis for calculating the carbon footprint of the energy consumption (power part) of the prefabricated cabin substation. At the same time, according to the calorific value, carbon content and carbon oxidation rate, the carbon footprint factors of gasoline and diesel are 2.925kgCO2 / kg and 3.089kgCO2 / kg respectively.

[0060] The carbon footprint factor library is composed of the average carbon emission factor of the power grid, the carbon footprint factors of gasoline and diesel, and other energy consumption carbon footprint factors, which can be used to calculate the carbon footprint of substation energy consumption.

[0061] According to the substation operation energy consumption data and carbon footprint factor, it is calculated that the carbon footprint of the prefabricated cabin substation in one operating year is 65.84t. The distribution of carbon footprint in each month is shown in Table 2.

[0062] Table 2 Carbon footprint of energy consumption of a 110kV substation

[0063]

[0064]

[0065] The SF6 gas monitoring module is used to monitor the greenhouse gases released by the power distribution device during operation and calculate the carbon footprint of SF6 leakage; the SF6 gas monitoring module includes an SF6 online monitoring alarm system.

[0066] The product of SF6 gas content, SF6 gas leakage rate and SF6 global warming potential is the SF6 leakage carbon footprint.

[0067] The carbon footprint of SF6 leakage is an important part of the carbon footprint of the prefabricated cabin substation during operation. The SF6 gas leakage amount is obtained according to the SF6 online monitoring and alarm system, and combined with the global warming potential of SF6 gas, the carbon footprint of SF6 gas leakage during the operation of the substation can be calculated. The carbon footprint of SF6 leakage can be expressed as:

[0068] CF S =C SF6 R SF6 ·GWP SF6

[0069] Among them, CF S is the carbon footprint of SF6 leakage in one year, C SF6 is the SF6 gas content in the power distribution device, R SF6 is the SF6 gas leakage rate, GWP SF6 is the global warming potential (GWP) of SF6.

[0070] The renewable energy compensation module is used to monitor the carbon compensation of the substation by the new energy power generation facilities supporting the substation, that is, the power generation and calculate the carbon footprint of renewable energy compensation; the renewable energy compensation module includes a power grid energy management system.

[0071] The amount of electricity generated by renewable energy is multiplied by the national grid average emission factor. If the product of the two is negative, the carbon footprint of renewable energy compensation is obtained.

[0072] Renewable energy generation can be considered as zero-carbon. According to the renewable energy generation obtained from the microgrid energy management system, combined with the national average grid emission factor of 0.5568kgCO2 / kWh, the renewable energy compensation carbon footprint (negative value) can be calculated. Specifically, the annual renewable energy generation obtained from the microgrid energy management system, the annual generation multiplied by the national average grid emission factor, the product of the two takes a negative value, and the annual renewable energy compensation carbon footprint CF is obtained. B The value of .

[0073] The carbon sink absorption module is mainly used to calculate the carbon absorption and carbon sink carbon footprint of green plants and lawns.

[0074] The green plant planting area is multiplied by the reference value of CO2 fixation (carbon sink factor), and the product of the two is a negative value, which can give the carbon sink carbon footprint.

[0075] Table 3 Reference values ​​of CO2 fixation in one year for different plants

[0076]

[0077]

[0078] Green plants, lawns and other vegetation are important ways to absorb carbon emissions in substations. By measuring the green plant planting area on site and referring to the reference value of CO2 fixation in one year for different plants (carbon sink factor) in Table 3, the carbon footprint of carbon sink in the prefabricated cabin substation can be calculated. Specifically, the green plant planting area is multiplied by the reference value of CO2 fixation in one year (carbon sink factor). The product of the two is negative, and the annual carbon sink carbon footprint CF can be obtained. H value.

[0079] Table 4 first calculates the monthly carbon footprint of the prefabricated cabin substation and then calculates the annual carbon footprint. The monthly energy consumption carbon footprint of the prefabricated cabin substation is obtained by multiplying the monthly operating energy consumption by the carbon footprint factor. The product of the SF6 gas content, SF6 gas leakage rate and SF6 global warming potential value is the SF6 leakage carbon footprint; the monthly power generation is multiplied by the national power grid average emission factor, and the product of the two is negative to obtain the monthly renewable energy compensation carbon footprint. The green plant planting area is multiplied by the monthly CO2 fixed amount reference value (carbon sink factor), and the product of the two is negative to obtain the monthly carbon sink carbon footprint.

[0080] Table 4 Carbon footprint of a 110kV substation in operation within one year

[0081]

[0082]

[0083] In the present invention, carbon footprint factors and carbon sink factors are used to calculate carbon footprint. The present invention introduces the global warming potential (GWP) to convert the carbon footprint of the prefabricated cabin substation into a carbon dioxide scalar. The carbon footprint factor calculated based on parameters such as energy calorific value, carbon content, and carbon oxidation rate is used as the energy carbon footprint factor, and the latest published electricity carbon dioxide emission factor is used as the station electricity carbon footprint factor to construct a carbon footprint factor library for calculating the carbon footprint generated by energy consumption in the energy consumption monitoring module. At the same time, the carbon footprint factor library can also be used in the renewable energy compensation module to calculate the renewable energy compensation carbon footprint; for SF6 gas leakage, considering that the greenhouse effects of SF6 and carbon dioxide are different, there are differences in the impact on the atmosphere, so GWP is introduced to uniformly convert greenhouse gases into carbon dioxide scalars, and the carbon footprint generated by the prefabricated cabin substation is uniformly expressed in carbon dioxide scalars. The average emission factor of the national power grid is 0.5568kgCO2 / kWh, and this emission factor is used as the basis for calculating the carbon footprint of the energy consumption (power part) of the prefabricated cabin substation. At the same time, based on the calorific value, carbon content and carbon oxidation rate, the carbon footprint factors of gasoline and diesel are 2.925kgCO2 / kg and 3.089kgCO2 / kg respectively.

[0084] The carbon footprint factor library is composed of the average carbon emission factor of the power grid, the carbon footprint factors of gasoline and diesel, and other energy consumption carbon footprint factors, to achieve the calculation of the carbon footprint of substation energy consumption. The carbon sequestration amount of common vegetation is queried to build a carbon sink factor library. Specifically, the carbon sink factor of common green plants is introduced, and a carbon sink factor library is built in the carbon sink absorption module to calculate the carbon footprint generated by carbon sinks in prefabricated cabin substations.

[0085] In the present invention, according to the fixed asset pricing and depreciation period of power grid enterprises provided in the "Cost Supervision and Audit Methods for Transmission and Distribution Prices" in Table 5, and considering the high operating reliability of prefabricated cabin substations, the upper limit of the reference period is selected as the design operating period n of the prefabricated cabin substation.

[0086] Table 5 Depreciation Period of Audit Method

[0087]

[0088]

[0089] In the present invention, the carbon footprint in the years of operation is calculated according to the actual carbon footprint, and the carbon footprint in the future years of operation is calculated according to the present invention. The first is the mean method, which is calculated according to the mean of the carbon footprint in the years of operation, and the mean of the years of operation is used as the carbon footprint data of other years to calculate the total carbon footprint generated by the prefabricated cabin substation during the entire operation period. The second is to train an extreme learning machine (ELM) neural network model with the carbon footprint data in the years of operation, predict the carbon footprint data of other years with the trained (ELM) neural network model, and calculate the total carbon footprint generated by the prefabricated cabin substation during the entire operation period.

[0090] The total carbon footprint of the prefabricated cabin substation during the entire operation period is calculated by the first average method as follows:

[0091]

[0092] Where, l is the number of years in operation, and n is the designed number of years in operation.

[0093] Taking the 110kV substation as an example, it has been in operation for 8 years. According to Table 5, the operational life is 32 years. The carbon footprint of the prefabricated cabin substation in the first 8 years is calculated according to the actual measurement results, and the carbon footprint of the next 24 years is calculated according to the average carbon footprint of the first 8 years. The total carbon footprint of the project during the operation period is shown in Table 6. In Table 6, the data from the 1st to the 8th year are known values, and the data from the 9th to the 32nd year are estimated values ​​obtained by the average value of the first 8 years.

[0094] Table 6 Carbon footprint of a 110kV prefabricated substation operation

[0095]

[0096]

[0097] Estimating the carbon footprint of future substation operation based on the carbon footprint of the years of operation can make a reasonable estimate of the carbon footprint level of the prefabricated cabin substation throughout its life cycle, but this method fails to effectively predict the trend of carbon footprint changes over time. Due to energy-saving transformation of equipment, improved SF6 monitoring level, increased green plant area in the plant and other reasons, the carbon footprint of future operation may further decrease. The error of estimating the total carbon footprint of the operation period based on the average carbon footprint of the years of operation will increase.

[0098] Therefore, the present invention proposes a second method for predicting carbon footprint data of other years through an (ELM) neural network model.

[0099] The present invention proposes a carbon footprint prediction model based on the combination of partial autocorrelation analysis (PACF) and extreme learning machine (ELM). The PACF-ELM model is used to predict the future operation carbon footprint of the prefabricated cabin substation, and the total predicted value of the operation carbon footprint of the prefabricated cabin substation is obtained.

[0100] Among them, the partial autocorrelation analysis (PACF) is used to determine which years' carbon footprints have the greatest impact on the current carbon footprint. After determining the carbon footprint data of the year with the greatest impact on the current carbon footprint, an extreme learning machine (ELM) neural network model is trained with the carbon footprint data of the year with the greatest impact, and the trained (ELM) neural network model is used to predict the carbon footprint data of other years.

[0101] Extreme learning machine (ELM) is a neural network with a single hidden layer structure. After the input weights are initialized, they generally do not need to be adjusted again. Compared with traditional neural networks that require a large number of training samples, extreme learning machines are more flexible and can adapt to complex machine learning needs. They have the advantages of fast convergence and are not prone to falling into local extreme values. The calculation steps of PACF are as follows:

[0102] (1) For a time series X t , t is the current forecast year. In the embodiment of the present invention, the substation has been in operation for 8 years, and the value of t starts from 9. k is the year, and k is between the years of operation. In the embodiment of the present invention, the substation has been in operation for 8 years, and k is between 1 and 8. The autocorrelation coefficient is calculated as follows:

[0103]

[0104] (2) When k=1 is initialized, the partial autocorrelation coefficient ψ 11 =ρ1

[0105] (3) For k>1, calculate the partial autocorrelation coefficient ψkk Values:

[0106]

[0107] Here, j takes values ​​between 1 and k-1.

[0108] according to Figure 4 The partial autocorrelation analysis results of the carbon footprint shown show that the carbon footprint 4 and 5 years ago has the greatest impact on the current carbon footprint.

[0109] The carbon footprint data from the 1st to the 8th year is known. Regarding the ELM neural network model training to determine the model parameters, since the partial autocorrelation analysis (PACF) has been used to determine that the carbon footprint data from the first 4 years and the first 5 years have the greatest impact on the current carbon footprint, during the model training, the carbon footprint of the 6th year can be predicted by the carbon footprint of the 2nd and 1st years, the carbon footprint of the 7th year can be predicted by the carbon footprint of the 3rd and 2nd years, and the carbon footprint of the 8th year can be predicted by the data of the 4th and 3rd years. The actual carbon footprint data of the 6th, 7th and 8th years are known. The actual carbon footprint data from the 1st to the 4th year is the training set, and the actual carbon footprint data of the 6th, 7th and 8th years is the validation set. When the model parameters with the smallest error between the predicted values ​​of the carbon footprints of the 6th, 7th and 8th years and the actual values ​​of the carbon footprints of the 6th, 7th and 8th years are the optimal model parameters, the optimal model parameters are determined to complete the ELM neural network model training.

[0110] According to the partial autocorrelation analysis (PACF), Figure 4 The results of partial autocorrelation analysis of carbon footprints shown in the figure show that the carbon footprints 4 years ago and 5 years ago have the greatest impact on the current carbon footprint. Therefore, the carbon footprint data of 4 years ago and 5 years ago are used as input variables, and the carbon footprint data of subsequent operating years are predicted through the trained ELM carbon footprint prediction model. For example, when predicting the carbon footprint data of the 9th year, the carbon footprints 4 years ago and 5 years ago have the greatest impact on the carbon footprint of the 9th year, so the carbon footprint data of 4 years ago, i.e. the 5th year, and 5 years ago, i.e. the 4th year, are used to predict the carbon footprint of the 9th year. When predicting the carbon footprint data of the 10th year, the carbon footprints of 4 years ago and 5 years ago have the greatest impact on the carbon footprint of the 10th year, so the carbon footprint data of 4 years ago, i.e. the 6th year, and 5 years ago, i.e. the 5th year, are used to predict the carbon footprint of the 10th year. Similarly, when predicting the carbon footprint data for the 15th year, the carbon footprint data from 4 years ago and 5 years ago have the greatest impact on the carbon footprint for the 15th year. Therefore, the carbon footprint data from 4 years ago, i.e. the 11th year, and 5 years ago, i.e. the 10th year, are used to predict the carbon footprint for the 15th year.

[0111] The number of nodes in the input layer of the ELM neural network model is k, the number of nodes in the hidden layer is m, and the number of nodes in the output layer is 1. For a given n samples (x i, t i ), input variable x i =(x i1 , x i2 , …, x ik ) T , t i is the carbon footprint value of the current year, i = 1, 2, ..., n, the ELM neural network can be expressed as:

[0112]

[0113] Among them, g(x) is the Sigmoid activation function, which is a fixed activation function used by neural networks; w i is the weight between the i-th hidden node and the input node, and the corresponding weight matrix is ​​an m×k matrix; b i is the bias of the i-th hidden node, and the corresponding bias matrix is ​​an m×1 matrix; β i is the weight between the i-th hidden node and the output node, obtained through sample training, and the corresponding weight matrix β is a 1×m matrix; j The output value of ELM is the predicted future operating carbon footprint value of the prefabricated cabin substation.

[0114] H represents the m×1 output matrix of the ELM hidden layer, β represents the output weight matrix, and T represents the 1×1 expected output matrix (corresponding to the t i ), ELM can be expressed as:

[0115] βH=T

[0116] Solving the above equation, we can get:

[0117]

[0118] in, is the solution of the weight matrix β, H + is the Moore-Penrose generalized inverse matrix of the hidden layer output matrix H.

[0119] according to Figure 4 The partial autocorrelation analysis results of the carbon footprint shown show that the carbon footprint 4 and 5 years ago has the greatest impact on the current carbon footprint.

[0120] Therefore, taking the carbon footprint data from 4 and 5 years ago as input variables, the carbon footprint of a 110 kV substation during the next 24 years of operation is predicted through the trained ELM carbon footprint prediction model. The prediction results are shown in Table 7.

[0121] In Table 7, the data from the 1st to the 8th year are known values, and the data from the 9th to the 32nd year are predicted values ​​calculated by the neural network model.

[0122] Table 7 Carbon footprint of a 110kV prefabricated substation operation

[0123]

[0124]

[0125] The ELM neural network model is a trend prediction model. Comparing Table 6 and Table 7, the predicted carbon footprint of the prefabricated cabin substation based on the PACF-ELM model takes into account the influence of the following factors:

[0126] (1) With the adoption of energy-saving equipment and energy-saving transformation of substations, station electricity consumption will further decrease;

[0127] (2) As the SF6 monitoring accuracy improves, the SF6 leakage rate will further decrease, and the greenhouse gas leakage in the substation will gradually decrease;

[0128] (3) As time goes by, the output of photovoltaic power stations gradually decreases. If the average carbon footprint of the previous few years is used for calculation, the calculated carbon footprint of the operation period will be too high. The results predicted by the PACF-ELM model are closer to the actual level of carbon footprint;

[0129] (4) As green plants and lawns grow, carbon sequestration capacity will be further enhanced. At the same time, the increase in green and lawn planting area and the change in vegetation types will also affect carbon sequestration capacity.

[0130] The present invention combines the characteristics of prefabricated cabin substations to construct a prefabricated cabin substation operation carbon footprint monitoring and forecasting system, which provides a systematic solution for the measurement of the carbon footprint of the prefabricated cabin substation during operation. It also provides two methods for predicting the carbon footprint of the prefabricated cabin substation during the entire operation period: mean estimation and model prediction. This provides a reference for the low-carbon and stable operation of prefabricated cabin substations.

[0131] Finally, it should be noted that the contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field. The above description is only the preferred embodiment of the present invention and is not used to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions recorded in the aforementioned embodiments, or to replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A prefabricated cabin substation operation carbon footprint monitoring and forecasting system, characterized in that: The system comprises: Energy consumption monitoring module, used to monitor and calculate the carbon footprint of energy consumption generated by substation electricity consumption and transportation vehicle energy consumption; SF6 gas monitoring module, used to monitor greenhouse gases released by power distribution equipment during operation; Renewable energy compensation module, used to monitor the carbon compensation of substations by supporting new energy power generation facilities; The carbon sink absorption module is used to monitor the carbon absorption of green plants and lawns. The sum of the carbon footprint of energy consumption, the carbon footprint generated by SF6 leakage, the carbon footprint of renewable energy compensation, and the carbon footprint of carbon sink is the operating carbon footprint of the prefabricated cabin substation.

2. According to claim 1, a prefabricated cabin substation operation carbon footprint monitoring and forecasting system is characterized by: The energy consumption monitoring module includes an intelligent operation and maintenance management system and a vehicle management platform for the substation; The SF6 gas monitoring module includes an SF6 online monitoring alarm system; The renewable energy compensation module includes a power grid energy management system; The carbon sink absorption module includes a green and lawn planting area acquisition unit.

3. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation operation, characterized in that: The method comprises: Construct four carbon footprint monitoring and forecasting system modules: energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module, and carbon sink absorption module; The data of the four modules, namely, energy consumption monitoring module, SF6 gas monitoring module, renewable energy compensation module and carbon sink absorption module, are traced to obtain the annual operating energy consumption data of the prefabricated cabin substation, the greenhouse gas data released by the distribution device during operation, the renewable energy power generation and the green and lawn planting area; Construct a carbon footprint factor library and a carbon sink factor library, introduce the global warming potential value GWP to convert the carbon footprint of the prefabricated cabin substation into a carbon dioxide scalar; the product of the operating energy consumption and the energy consumption carbon footprint factor is the energy consumption carbon footprint of the prefabricated cabin substation, the product of the released greenhouse gas data and the global warming potential value GWP value is the SF6 leakage carbon footprint, the product of the renewable energy power generation and the national power grid average emission factor is the negative value of the renewable energy compensation carbon footprint, and the product of the green and lawn planting area and the carbon sink factor is the negative value of the carbon sink carbon footprint; The carbon footprint data of each year in the operating life of the prefabricated cabin substation is obtained by summing up the carbon footprint of energy consumption, SF6 leakage, renewable energy compensation and carbon sink in each year of the operating life of the prefabricated cabin substation. The carbon footprint data of each year within the operating years is used to predict the carbon footprint of the prefabricated cabin substation in future operating years.

4. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 3, characterized in that: The power consumption of the air-conditioning system, lighting system, fan, fire protection system, control system and other stations in the operating energy consumption comes from the intelligent operation and maintenance management system of the substation, the fuel consumption data of the operating vehicles is obtained from the vehicle management platform, the released greenhouse gas data is obtained from the SF6 online monitoring and alarm system, the renewable energy power generation comes from the microgrid energy management system, and the planting area of ​​green, lawn, etc. is obtained from the on-site measurement of the substation.

5. The method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 3 is characterized in that: The prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: calculating according to the average carbon footprint of the operating years, taking the average of the operating years as the carbon footprint data of other years, and calculating the total carbon footprint generated by the prefabricated cabin substation during the entire operating period.

6. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 3, characterized in that: The prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: training an extreme learning machine ELM neural network model through the carbon footprint data in the operating years, predicting the carbon footprint data of other years through the trained ELM neural network model, and calculating the total carbon footprint generated by the prefabricated cabin substation during the entire operating period.

7. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 3, characterized in that: The carbon footprint factor library includes the average carbon emission factor of the power grid, the carbon footprint factors of gasoline and diesel, and other energy consumption carbon footprint factors; the carbon sink factor library includes the carbon fixation amount of common vegetation.

8. The method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 3 is characterized in that: The prediction of the carbon footprint of the prefabricated cabin substation in future operating years includes: determining the year with the greatest impact on the carbon footprint of the current year through partial autocorrelation analysis PACF, after determining the carbon footprint data of the year with the greatest impact on the current carbon footprint, using the carbon footprint data of the year with the greatest impact to train an extreme learning machine ELM neural network model, taking the carbon footprint data of the year with the greatest impact as input, and predicting the carbon footprint data of future operating years through the trained ELM neural network model.

9. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 8, characterized in that : The method for calculating the autocorrelation coefficient ψ_kk value by the partial autocorrelation analysis PACF is: in, For a time series X t , t is the current forecast year, where t = 1, 2, …, n; k is the year, and k takes values ​​between the years that have been run.

10. A method for monitoring and predicting the carbon footprint of a prefabricated cabin substation according to claim 8, characterized in that : The ELM neural network model has k nodes in the input layer, m nodes in the hidden layer, and 1 node in the output layer. For a given n samples (x i ,t i ), input variable x i =(x i1 ,x i2 ,…,x ik ) T , t i is the carbon footprint value of the current year, i=1,2,…,n, and the ELM neural network model is expressed as: Among them, j is the output value of ELM, i.e. the predicted future operation carbon footprint value of the prefabricated cabin substation, g(x) is the Sigmoid activation function, which is the fixed activation function used by the neural network; w i is the weight between the i-th hidden node and the input node, and the corresponding weight matrix is ​​an m×k matrix; b i is the bias of the i-th hidden node, and the corresponding bias matrix is ​​an m×1 matrix; β i is the weight between the i-th hidden node and the output node.

Images