Carbon emission parameter acquisition device and acquisition method thereof

Abstract

The invention discloses a carbon emission parameter acquisition device and an acquisition method thereof. Through joint analysis of a harmonic distribution characteristic spectrum and a power fluctuation characteristic, an energy type classification model can accurately identify energy characteristic information of a current power grid; and based on a dynamic matching mechanism of the layered carbon emission factor library, weighting correction is further carried out on the reference emission factor by using a real-time power grid load rate and a dynamic correction coefficient, and a real-time carbon emission factor closer to reality is generated. According to the method, the comprehensive carbon emission is calculated by adopting a hierarchical weighting algorithm, and the change rate of the carbon emission intensity is predicted, so that refined dynamic accounting of carbon emission parameters is realized. Finally, a complete carbon emission parameter set is uploaded to a supervision platform through an edge communication module, the problem of errors caused by dependence on fixed emission factors in a traditional method is solved, and high-precision and real-time data support is provided for carbon management of industrial parks.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a carbon emission parameter acquisition device and a carbon emission parameter acquisition method. Background Art

[0002] Traditional carbon emissions collection methods typically estimate based on rated power or fixed emission factors, making it difficult to reflect the dynamic emission characteristics under real-world operating conditions. For example, electricity carbon emission factors typically use the average value of the regional power grid, ignoring the impact of factors such as renewable energy penetration and load fluctuations. Variables such as the actual operating efficiency of industrial equipment and the degree of fuel combustion are also not accurately included in the calculation. Furthermore, in multi-energy complementary scenarios, the coupling effect between different energy sources further complicates carbon emissions accounting. This extensive collection method results in insufficiently accurate carbon emissions data, making it difficult to support refined carbon management needs. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a carbon emission parameter collection device and a collection method thereof.

[0004] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0005] In a first aspect, the present invention provides a carbon emission parameter collection device, which is suitable for a power grid. The device includes a power quality monitoring module, a carbon emission calculation module, an edge communication module and a local storage module; the power quality monitoring module is used to collect the original electrical signals of the power grid in real time and calculate the electrical signal parameters; the carbon emission calculation module is connected to the power quality monitoring module, and is used to dynamically calculate the carbon emission parameters based on the electrical signal parameters and the energy type of the equipment in the power grid. The carbon emission calculation module includes an energy type identification unit and a carbon emission factor dynamic matching unit. The energy type identification unit automatically identifies the energy type by analyzing the harmonic characteristics and power fluctuation characteristics of the power grid. The carbon emission factor dynamic matching unit calls the corresponding carbon emission factor in real time according to the identified energy type for calculation; the edge communication module supports multi-protocol data encapsulation and transmission, and is used to upload carbon emission parameters to the supervision platform; the local storage module is used to cache carbon emission parameters and support continued transmission after network disconnection.

[0006] Furthermore, the original electrical signal includes an original voltage signal and an original current signal. The power quality monitoring module includes a voltage transformer, a current transformer, a signal conditioning circuit, an analog-to-digital conversion unit and a digital signal processor. The voltage transformer is used to collect the original voltage signal; the current transformer is used to collect the original current signal; the signal conditioning circuit is used to filter and amplify the collected original electrical signal to obtain a first processed signal; the analog-to-digital conversion unit uses synchronous sampling to digitize the first processed signal to obtain a second processed signal; the digital signal processor is used to perform real-time calculation on the second processed signal to obtain electrical signal parameters, which include electrical power parameters and harmonic content.

[0007] Furthermore, the energy type identification unit includes a harmonic feature analysis module, a power fluctuation analysis module and a machine learning classifier. The harmonic feature analysis module is used to extract the harmonic features of the power grid through FFT transformation; the power fluctuation analysis module is used to monitor the power fluctuation characteristics; the machine learning classifier is configured to be trained based on historical data, and the machine learning classifier is used to classify energy types according to the harmonic features of the power grid and the power fluctuation characteristics.

[0008] In a second aspect, the present invention further provides a carbon emission parameter collection method, applicable to the collection device described in the first aspect, the method comprising:

[0009] The power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters;

[0010] The energy type identification unit performs a dynamic energy type identification process, including:

[0011] Perform real-time harmonic feature extraction on electrical signal parameters to obtain harmonic distribution characteristic spectrum;

[0012] Synchronously monitor the power fluctuation curve and extract the power fluctuation characteristics, which include fluctuation periodicity and amplitude characteristics;

[0013] The harmonic distribution characteristic spectrum and power fluctuation characteristics are input into the pre-trained energy type classification model to output the energy characteristic information of the current power grid;

[0014] Input the energy characteristic information into the carbon emission factor dynamic matching unit to perform carbon emission factor matching, including:

[0015] Obtain a stratified carbon emission factor library, which includes benchmark emission factors and dynamic correction factors corresponding to multiple energy types;

[0016] According to the energy characteristic information, the corresponding benchmark emission factor is extracted from the stratified carbon emission factor library;

[0017] Furthermore, the benchmark emission factor is dynamically weighted and corrected according to the real-time grid load rate and the dynamic correction coefficient to generate a real-time carbon emission factor corresponding to the current energy characteristic information;

[0018] Dynamically calculate carbon emission parameters based on electrical signal parameters and matching carbon emission factors, including:

[0019] A hierarchical weighted algorithm is used to calculate comprehensive carbon emissions;

[0020] Based on the energy composition change trend within the preset time window, predict the carbon emission intensity change rate in the next cycle;

[0021] The complete carbon emission parameter set including energy characteristic information, real-time carbon emission factors, comprehensive carbon emissions and carbon emission intensity change rate is uploaded to the supervision platform through the edge communication module.

[0022] Furthermore, the power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters, including:

[0023] The original voltage signal is collected through a voltage transformer, and the original current signal is collected through a current transformer;

[0024] Perform signal conditioning on the original voltage signal and the original current signal, including:

[0025] Perform voltage division and filtering on the original voltage signal to obtain a conditioned voltage signal;

[0026] Convert and filter the original current signal to obtain a conditioned current signal;

[0027] The conditioned voltage and current signals are synchronously sampled and converted into analog-to-digital signals, including:

[0028] The phase-locked loop method is used to track the fundamental frequency of the power grid and realize the synchronous sampling of voltage and current signals;

[0029] Converting the analog signal obtained by synchronous sampling into a digital signal, the digital signal including a digital voltage signal and a digital current signal;

[0030] Perform real-time computation and processing on digital signals, including:

[0031] Calculate real-time electric power parameters based on digital voltage and current signals, including active power, reactive power, and apparent power;

[0032] Extract the harmonic components of digital voltage and current signals through fast Fourier transform and calculate the harmonic content parameters, which include total harmonic distortion and the content rate of each harmonic.

[0033] The electrical signal parameters are generated according to the electrical power parameters and the harmonic content parameters.

[0034] Furthermore, real-time harmonic feature extraction is performed on the electrical signal parameters to obtain the harmonic distribution characteristic spectrum including:

[0035] Normalize the harmonic content parameters to generate a standardized harmonic amplitude spectrum;

[0036] The distribution characteristics of the characteristic harmonic group are extracted based on the standardized harmonic amplitude spectrum, including:

[0037] Determine the amplitude ratio relationship of the 3rd, 5th, and 7th characteristic harmonics;

[0038] Calculate the energy gradient change rate between each harmonic;

[0039] Based on the harmonic content parameters of continuous sampling periods, a time-varying harmonic characteristic matrix is ​​constructed, including:

[0040] Perform sliding window analysis on the normalized harmonic amplitude spectrum;

[0041] Extract the time series variation characteristics of each harmonic phase angle;

[0042] The distribution characteristics of the characteristic harmonic group are fused with the time-varying harmonic characteristic matrix to generate the harmonic distribution characteristic spectrum.

[0043] Furthermore, monitoring the power fluctuation curve and extracting the power fluctuation characteristics include:

[0044] Perform sliding time window analysis on active power and calculate power fluctuation rate;

[0045] Extract the time domain features of the power fluctuation curve, including:

[0046] Determine the periodic characteristics of power fluctuations;

[0047] Calculate the amplitude envelope characteristics of power fluctuations;

[0048] Perform frequency domain analysis on power fluctuations to obtain frequency domain characteristics, including:

[0049] Extract the spectrum characteristics of power fluctuations through Fourier transform;

[0050] Identify the main frequency components of power fluctuations;

[0051] Generate frequency domain features based on main frequency components and spectrum characteristics;

[0052] The power fluctuation rate, time domain features, amplitude envelope features and frequency domain features are integrated to generate power fluctuation features.

[0053] Furthermore, the benchmark emission factor is dynamically weighted and corrected according to the real-time grid load rate and the dynamic correction coefficient to generate the real-time carbon emission factor corresponding to the current energy characteristic information, including:

[0054] Based on the energy composition ratio in the energy characteristic information, the benchmark emission factors corresponding to each energy type are extracted from the stratified carbon emission factor library;

[0055] Obtain the current grid load rate parameter and calculate the load rate correction weight, which is expressed by formula (1). Formula (1) is as follows:

[0056] w L =α·(1-e -β·L );

[0057] In formula (1), W L is the debt ratio correction weight, L is the real-time grid debt ratio, α is the first load ratio correction coefficient, β is the second debt ratio correction coefficient, and e is the base of the natural logarithm;

[0058] The initial revision of each baseline emission factor is made according to the dynamic revision factor, including:

[0059] Applying load rate correlation correction to the carbon emission factor of thermal power energy is expressed by formula (2), which is as follows:

[0060] E′ c =E c ·(1+γ·W L );

[0061] In formula (2), E′ c is the modified carbon emission factor of thermal power energy, E c is the benchmark carbon emission factor for thermal power energy, and γ is the thermal power load sensitivity coefficient;

[0062] Applying period validity correction to the carbon emission factor of renewable energy is expressed by formula (3), which is as follows:

[0063] E′ r =E r ·(1+η·A t );

[0064] In formula (3), E′ r is the carbon emission factor of renewable energy after correction, E r is the benchmark carbon emission factor of renewable energy, η is the renewable energy adjustment coefficient, A t is the time period availability rate;

[0065] The revised carbon emission factor is weighted twice using a dynamic weighting algorithm, including:

[0066] Determine the weight coefficient of each energy type according to the energy composition ratio;

[0067] Combined with the load factor correction weight to perform comprehensive weighted calculation;

[0068] The output matrix consists of the real-time carbon emission factors of each energy contribution, which is expressed by formula (4). Formula (4) is as follows:

[0069]

[0070] In formula (4), EF is the matrix of real-time carbon emission factors, E i ′ is the carbon emission factor of the i-th energy type after correction, P i is the energy composition ratio of the i-th energy type.

[0071] Furthermore, the hierarchical weighted algorithm is used to calculate the comprehensive carbon emissions, including:

[0072] The comprehensive carbon emissions within the preset period are obtained and expressed by formula (5), which is as follows:

[0073]

[0074] In formula (5), C Δt is the comprehensive carbon emissions, Δt is the preset time period, Ψ i (t) is the real-time power of energy of the i-th energy type, and t0 is the starting time of the preset time period;

[0075] The output includes a comprehensive carbon emission data set of time series, which is expressed by formula (6). Formula (6) is as follows:

[0076]

[0077] In formula (6), C Δt (t k ) is t k Comprehensive carbon emissions at the sampling time, t k is the kth sampling moment, and N is the total number of sampling points.

[0078] Furthermore, based on the energy composition change trend within the preset time window, the carbon emission intensity change rate for the next cycle is predicted to include:

[0079] Based on the energy characteristics information and the historical change rate of carbon emission intensity, perform the following steps:

[0080] Perform sliding window analysis on the energy composition ratio in the energy characteristic information to extract the temporal variation characteristics of the energy type proportion;

[0081] Modeling the correlation between temporal variation characteristics and the corresponding historical change rate of carbon emission intensity includes:

[0082] Establish a correlation coefficient matrix between changes in the proportion of each energy type and changes in carbon emission intensity;

[0083] Identify temporal cyclical patterns in the historical rate of change of carbon emission intensity;

[0084] Construct a carbon emission intensity prediction model based on the correlation coefficient matrix and time periodicity pattern;

[0085] The carbon emission intensity prediction model is used to predict the energy composition change trend of the next cycle and output the predicted value of the carbon emission intensity change rate;

[0086] The error analysis between the predicted value of the carbon emission intensity change rate and the real-time calculated carbon emission intensity change rate is carried out, and the correlation coefficient matrix and the parameters of the time periodic pattern are dynamically updated.

[0087] By adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0088] The power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters. Combined with the dynamic discrimination process of the energy type identification unit, the matching accuracy of the carbon emission factor is significantly improved. Through the joint analysis of the harmonic distribution characteristic spectrum and the power fluctuation characteristics, the energy type classification model can accurately identify the energy characteristic information of the current power grid; based on the dynamic matching mechanism of the hierarchical carbon emission factor library, the real-time grid load rate and dynamic correction coefficient are further used to perform weighted correction on the benchmark emission factor to generate a real-time carbon emission factor that is closer to reality. This method uses a hierarchical weighted algorithm to calculate the comprehensive carbon emissions and predict the rate of change of carbon emission intensity, realizing the refined dynamic accounting of carbon emission parameters. Finally, the complete carbon emission parameter set is uploaded to the supervision platform through the edge communication module, which solves the error problem caused by the traditional method relying on fixed emission factors, and provides high-precision and real-time data support for carbon management in industrial parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0090] Figure 1 It is a schematic diagram of the carbon emission parameter collection device;

[0091] Figure 2This is a schematic diagram of the first step of the carbon emission parameter collection method;

[0092] Figure 3 This is a schematic diagram of the second step of the carbon emission parameter collection method.

[0093] 1. Carbon emission parameter collection device;

[0094] 11. Power quality monitoring module;

[0095] 12. Carbon emission calculation module;

[0096] 13. Edge communication module;

[0097] 14. Local storage module. DETAILED DESCRIPTION

[0098] The present invention will be described in further detail below with reference to the accompanying drawings and examples. It is particularly noted that the following examples are intended only to illustrate the present invention and are not intended to limit the scope of the present invention. Similarly, the following examples are only some embodiments of the present invention and are not intended to be exhaustive. All other embodiments obtained by those of ordinary skill in the art without creative effort are intended to fall within the scope of protection of the present invention.

[0099] See also Figure 1 In the first aspect, this embodiment provides a carbon emission parameter collection device, which is suitable for a power grid. The device includes a power quality monitoring module, a carbon emission calculation module, an edge communication module and a local storage module; the power quality monitoring module is used to collect the original electrical signals of the power grid in real time and calculate the electrical signal parameters; the carbon emission calculation module is connected to the power quality monitoring module, and is used to dynamically calculate the carbon emission parameters based on the electrical signal parameters and the energy type of the equipment in the power grid. The carbon emission calculation module includes an energy type identification unit and a carbon emission factor dynamic matching unit. The energy type identification unit automatically identifies the energy type by analyzing the harmonic characteristics and power fluctuation characteristics of the power grid. The carbon emission factor dynamic matching unit calls the corresponding carbon emission factor in real time according to the identified energy type for calculation; the edge communication module supports multi-protocol data encapsulation and transmission, and is used to upload carbon emission parameters to the supervision platform; the local storage module is used to cache carbon emission parameters and support network disconnection and resumption.

[0100] The carbon emission parameter collection device provided in this embodiment is suitable for power grids. Its core components include a power quality monitoring module, a carbon emission calculation module, an edge communication module, and a local storage module. The power quality monitoring module is based on the hardware architecture of the LY-6640 device (using a four-layer PCB process). It collects the raw electrical signals of the power grid in real time through a voltage transformer (input range 5-420V, overload capacity 4Un / 1s) and a current transformer (range 5mA-6A, accuracy ±0.1%). It then calculates the electrical signal parameters in accordance with eight national standards, including GB / T 14549-93, using a 16-bit synchronous ADC (sampling 1024 points per cycle) and a DSP processor. The carbon emission calculation module extracts grid characteristics through a harmonic analysis unit (supporting FFT transforms for harmonics 2-65, compliant with the IEC 61000-4-30 standard) and a power fluctuation analysis unit (calculating Pst flicker values ​​within a 200ms window). It then uses a pre-trained random forest model (trained on 100,000 data sets) to automatically identify six energy types, including photovoltaic, wind power, and hydrogen. The carbon emission factor dynamic matching unit then uses the corresponding factors in real time to calculate carbon emission parameters. The edge communication module utilizes industrial Ethernet (100Mbps), RS485, and Modbus-TCP multi-protocol encapsulation technology to upload the calculated results to the industrial park's energy and carbon perception system platform. The local storage module uses a 16GB TF card for data caching and network-disconnected data resiliency, ensuring that data is temporarily stored during network outages and automatically retransmitted upon restoration. The entire device achieves real-time monitoring and dynamic accounting of grid carbon emission parameters through the coordinated integration of precise hardware acquisition, intelligent algorithm recognition, and reliable communication transmission.

[0101] The carbon emission parameter collection device provided in this embodiment realizes accurate dynamic accounting of carbon emissions from multiple energy sources such as photovoltaic and wind power through a high-precision power quality monitoring module (combined with an intelligent carbon emission calculation module). The edge communication module supports multi-protocol transmission such as industrial Ethernet and RS485, and adopts the LY-6640 hardware architecture and DSP real-time processing technology. It is particularly suitable for deployment in new power systems with a high proportion of new energy, providing a highly reliable and low-latency solution for grid carbon footprint monitoring.

[0102] Furthermore, the original electrical signal includes an original voltage signal and an original current signal. The power quality monitoring module includes a voltage transformer, a current transformer, a signal conditioning circuit, an analog-to-digital conversion unit and a digital signal processor. The voltage transformer is used to collect the original voltage signal; the current transformer is used to collect the original current signal; the signal conditioning circuit is used to filter and amplify the collected original electrical signal to obtain a first processed signal; the analog-to-digital conversion unit uses synchronous sampling to digitize the first processed signal to obtain a second processed signal; the digital signal processor is used to perform real-time calculation on the second processed signal to obtain electrical signal parameters, which include electrical power parameters and harmonic content.

[0103] In this embodiment, the original electrical signal includes an original voltage signal and an original current signal. The power quality monitoring module is composed of a voltage transformer, a current transformer, a signal conditioning circuit, an analog-to-digital conversion unit, and a digital signal processor. Specifically, the voltage transformer adopts a rated input 57.7V / 100V / 220V (optional) configuration, complies with the GB / T 12325-2008 voltage deviation standard, and has an insulation strength of 2.5kV / 1min, which is used for high-precision collection of original voltage signals; the current transformer adopts a 1A / 5A dual-range configuration, has an overload capacity of 10In / 1s and a low power consumption characteristic of ≤0.5VA / channel, and is specifically used for collecting original current signals; the signal conditioning circuit is manufactured using SMT technology, with an input impedance of >100kΩ, and has passed GB / T 17626.4 Level IV fast transient pulse immunity test: the collected original electrical signal is filtered and amplified to obtain a first processed signal; the analog-to-digital conversion unit uses synchronous sampling technology to digitize the first processed signal into a second processed signal; the digital signal processor uses an ARM+DSP dual-core architecture and runs the Linux system to perform real-time calculations on the second processed signal, outputting 58 electrical signal parameters including electrical power parameters and harmonic content. The harmonic analysis covers key indicators such as THD (total harmonic distortion) and 5-2495Hz interharmonics.

[0104] This embodiment uses phase-locked loop technology to achieve a fundamental frequency tracking accuracy of ±0.01Hz and a harmonic phase angle measurement error of ≤±5°, providing millisecond-level data updates for carbon emission calculations; the multi-range configuration of the mutual inductor is combined with a high-anti-interference signal conditioning circuit to ensure that measurement accuracy can be maintained even in complex power grid environments; the dual-core processor architecture design not only realizes real-time calculation of 58 parameters, but also reserves sufficient computing power for subsequent algorithm upgrades; it can still work reliably under extreme conditions such as voltage deviation and harmonic distortion, providing high-precision and high-reliability technical support for carbon emission monitoring of new power systems, and effectively solving the carbon emission measurement problem caused by the access of new energy.

[0105] Furthermore, the energy type identification unit includes a harmonic feature analysis module, a power fluctuation analysis module and a machine learning classifier. The harmonic feature analysis module is used to extract the harmonic features of the power grid through FFT transformation; the power fluctuation analysis module is used to monitor the power fluctuation characteristics; the machine learning classifier is configured to be trained based on historical data, and the machine learning classifier is used to classify energy types according to the harmonic features of the power grid and the power fluctuation characteristics.

[0106] In this embodiment, the energy type identification unit adopts an intelligent identification solution based on multi-dimensional feature fusion. The complete technical chain consists of a harmonic feature analysis module, a power fluctuation analysis module, and a machine learning classifier. The harmonic feature analysis module strictly complies with the IEC 61000-4-30 standard. It not only extracts the 3rd / 5th / 7th harmonics (the typical ratio for thermal power is 1:0.8:0.6), but also innovatively constructs a time-varying harmonic matrix that includes the temporal variation characteristics of the harmonic phase angle (sampling accuracy ±0.5°), effectively capturing the dynamic harmonic characteristics of different energy sources. The power fluctuation analysis module calculates Pst short-term flicker (compliant with GB / T12326-2008) and the 0.1-25Hz frequency band fluctuation envelope to accurately identify the 0.5-2Hz fluctuation characteristics unique to photovoltaic power generation. The machine learning classifier uses a random forest algorithm and is constructed based on a training set containing 100,000 sets of multi-energy operation data. The output not only includes energy type classification but also provides confidence assessment.

[0107] This embodiment uses a deep fusion analysis of the dynamic characteristics of the harmonic phase angle (±2° / s rate of change) and the frequency domain characteristics of power fluctuations to increase the identification accuracy of wind power and hydrogen energy, which are difficult to distinguish with traditional methods, from 72% to 91%. The use of B-code timing technology enables μs-level synchronized multi-node data comparison, solving the problem of spatiotemporal consistency in distributed energy monitoring. The system's built-in adaptive feature weighting algorithm can dynamically adjust the contribution ratio of harmonic and fluctuation characteristics according to the operating status of the power grid, and supports online learning functions. When a new energy source is detected, it can automatically trigger model incremental training (update cycle <24 hours) to ensure the continuous evolution of the classification system.

[0108] See also Figure 2 In a second aspect, this embodiment further provides a carbon emission parameter collection method, applicable to the collection device described in the first aspect, the method comprising:

[0109] S101, collecting the original electrical signals of the power grid in real time and calculating the electrical signal parameters through the power quality monitoring module;

[0110] S102: Executing a dynamic energy type identification process by an energy type identification unit, including:

[0111] Perform real-time harmonic feature extraction on electrical signal parameters to obtain harmonic distribution characteristic spectrum;

[0112] Synchronously monitor the power fluctuation curve and extract the power fluctuation characteristics, which include fluctuation periodicity and amplitude characteristics;

[0113] The harmonic distribution characteristic spectrum and power fluctuation characteristics are input into the pre-trained energy type classification model to output the energy characteristic information of the current power grid;

[0114] S103. Inputting the energy characteristic information into the carbon emission factor dynamic matching unit to perform carbon emission factor matching, including:

[0115] Obtain a stratified carbon emission factor library, which includes benchmark emission factors and dynamic correction factors corresponding to multiple energy types;

[0116] According to the energy characteristic information, the corresponding benchmark emission factor is extracted from the stratified carbon emission factor library;

[0117] Furthermore, the benchmark emission factor is dynamically weighted and corrected according to the real-time grid load rate and the dynamic correction coefficient to generate a real-time carbon emission factor corresponding to the current energy characteristic information;

[0118] S104. Dynamically calculate carbon emission parameters based on the electrical signal parameters and the matching carbon emission factors, including:

[0119] A hierarchical weighted algorithm is used to calculate comprehensive carbon emissions;

[0120] Based on the energy composition change trend within the preset time window, predict the carbon emission intensity change rate in the next cycle;

[0121] S105. Upload the complete carbon emission parameter set including energy characteristic information, real-time carbon emission factor, comprehensive carbon emissions and carbon emission intensity change rate to the supervision platform through the edge communication module.

[0122] The power quality monitoring module collects the grid's raw electrical signals in real time and calculates their parameters. Specifically, the raw voltage and current signals are collected using a voltage transformer (input range 5-420V, compliant with GB / T 12325-2008) and a current transformer (1A / 5A dual-range, overload capacity 10In / 1s). After processing by a signal conditioning circuit (four-layer PCB design, EMC immunity reaching GB / T 17626.4 Level IV), the signals are digitized using a 16-bit synchronous ADC with a 51.2kHz sampling rate. Finally, an ARM+DSP dual-core processor calculates the electrical signal parameters, including power parameters and harmonic content (2nd to 65th harmonics, compliant with IEC 61000-4-30).

[0123] Real-time harmonic feature extraction is performed on electrical signal parameters. A time-varying harmonic matrix is ​​constructed using FFT transforms to extract the 3rd, 5th, and 7th harmonic content and phase angle time series characteristics. Power fluctuation curves are simultaneously monitored, and the Pst flicker value (GB / T 12326-2008) and the fluctuation envelope characteristics in the 0.1-25Hz frequency band are calculated using a 200ms sliding window. The harmonic distribution characteristic spectrum and power fluctuation characteristics are input into a pre-trained random forest classification model (training data includes 100,000 sets of data for six energy categories, including photovoltaic, wind power, and hydrogen energy). The model outputs energy characteristic information and a ≥95% confidence level. A baseline emission factor is extracted from a hierarchical carbon emission factor library and dynamically modified based on the real-time grid load rate and time period weighting coefficient to generate a real-time carbon emission factor. A hierarchical weighted algorithm is used to calculate comprehensive carbon emissions, and the rate of change in carbon emission intensity for the next period is predicted based on the ARIMA model. Finally, the complete carbon emission parameter set is uploaded to the industrial park's energy and carbon management platform via an edge communication module using Industrial Ethernet (100Mbps) or Modbus-TCP protocol.

[0124] In this embodiment, the fusion analysis of harmonic phase angle (±5° accuracy) and power fluctuation frequency domain characteristics (photovoltaic 0.5-2Hz) reduces the misjudgment rate of energy types and improves the accuracy of distinguishing between wind power and hydrogen energy; the dynamic correction model improves the correction sensitivity of thermal power carbon emission factors through real-time load rate and time period availability adjustment; μs-level B code synchronization ensures the synchronization accuracy of multi-node data and supports precise carbon metering of distributed energy; the entire workflow can be completed within a few seconds, meeting the real-time carbon monitoring needs of new power systems.

[0125] Furthermore, the power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters, including:

[0126] The original voltage signal is collected through a voltage transformer, and the original current signal is collected through a current transformer;

[0127] Perform signal conditioning on the original voltage signal and the original current signal, including:

[0128] Perform voltage division and filtering on the original voltage signal to obtain a conditioned voltage signal;

[0129] Convert and filter the original current signal to obtain a conditioned current signal;

[0130] The conditioned voltage and current signals are synchronously sampled and converted into analog-to-digital signals, including:

[0131] The phase-locked loop method is used to track the fundamental frequency of the power grid and realize the synchronous sampling of voltage and current signals;

[0132] Converting the analog signal obtained by synchronous sampling into a digital signal, the digital signal including a digital voltage signal and a digital current signal;

[0133] Perform real-time computation and processing on digital signals, including:

[0134] Calculate real-time electric power parameters based on digital voltage and current signals, including active power, reactive power, and apparent power;

[0135] Extract the harmonic components of digital voltage and current signals through fast Fourier transform and calculate the harmonic content parameters, which include total harmonic distortion and the content rate of each harmonic.

[0136] The electrical signal parameters are generated according to the electrical power parameters and the harmonic content parameters.

[0137] In this embodiment, the original voltage signal is collected by a voltage transformer, and the original current signal is collected by a current transformer; the original voltage signal is divided and filtered to obtain a conditioned voltage signal, and the original current signal is converted and filtered to obtain a conditioned current signal; the conditioned voltage signal and current signal are synchronously sampled and analog-to-digital converted, including using a phase-locked loop method (hardware PLL implementation, fundamental frequency tracking range 42.5-57.5Hz, error ±0.01Hz, phase difference ≤1μs) to track the fundamental frequency of the power grid, so as to realize the voltage signal and current signal. Synchronous sampling is performed, and the analog signals obtained by synchronous sampling are converted into digital signals (digital voltage signals and digital current signals) through a 16-bit ADC; real-time calculation and processing of the digital signals are performed, including calculation of real-time electric power parameters (active power, reactive power and apparent power, with an accuracy of ±0.1%) based on the digital voltage signal and digital current signal, extraction of harmonic components of the digital voltage signal and digital current signal through fast Fourier transform (1024 points per cycle, 2-65th harmonics, THD accuracy of ±5%), calculation of harmonic content parameters (total harmonic distortion rate and each harmonic content rate, in accordance with GB / T 14549-93), and finally, generation of a complete electric signal parameter set based on the electric power parameters and harmonic content parameters.

[0138] In this embodiment, a phase-locked loop (PLL) method effectively eliminates spectrum leakage, keeping harmonic measurement errors to an ultra-low 0.05% Un (when Uh < 1% Un). A dual-core DSP + ARM processor architecture enables efficient parallel computing, enabling accurate analysis of 65th harmonics within 3 seconds. The power quality monitoring module maintains stable operation across a wide temperature range of -25°C to 65°C, with all indicators meeting or exceeding Class A power quality monitoring standards. These features provide accurate, real-time electrical signal parameters for carbon emissions calculations, meeting the stringent carbon metering accuracy requirements of emerging power systems and providing reliable data support for the low-carbon operation of the power grid.

[0139] Furthermore, real-time harmonic feature extraction is performed on the electrical signal parameters to obtain the harmonic distribution characteristic spectrum including:

[0140] Normalize the harmonic content parameters to generate a standardized harmonic amplitude spectrum;

[0141] The distribution characteristics of the characteristic harmonic group are extracted based on the standardized harmonic amplitude spectrum, including:

[0142] Determine the amplitude ratio relationship of the 3rd, 5th, and 7th characteristic harmonics;

[0143] Calculate the energy gradient change rate between each harmonic;

[0144] Based on the harmonic content parameters of continuous sampling periods, a time-varying harmonic characteristic matrix is ​​constructed, including:

[0145] Perform sliding window analysis on the normalized harmonic amplitude spectrum;

[0146] Extract the time series variation characteristics of each harmonic phase angle;

[0147] The distribution characteristics of the characteristic harmonic group are fused with the time-varying harmonic characteristic matrix to generate the harmonic distribution characteristic spectrum.

[0148] In this embodiment, the harmonic content parameters are normalized to generate a standardized harmonic amplitude spectrum (according to the limits set in GB / T14549-93, such as the 4.0% odd harmonic limit for a 0.38kV power grid). The distribution characteristics of the characteristic harmonic group are extracted from the standardized harmonic amplitude spectrum, including determining the amplitude ratio of the 3rd, 5th, and 7th characteristic harmonics (a typical ratio of 1:0.8:0.6 for thermal power generation) and calculating the energy gradient change rate between the harmonics. A time-varying harmonic characteristic matrix is ​​constructed based on the harmonic content parameters of consecutive sampling cycles, including performing a sliding window analysis on the standardized harmonic amplitude spectrum to extract the temporal variation characteristics of the phase angles of each harmonic. Finally, the distribution characteristics of the characteristic harmonic group are integrated with the time-varying harmonic characteristic matrix to generate a harmonic distribution characteristic spectrum. This process is synchronized with B-code timing technology to ensure time alignment accuracy of multi-node data at the μs level.

[0149] In this embodiment, analysis of the amplitude ratio of characteristic harmonic groups improves the accuracy of distinguishing wind power (dominated by the fifth harmonic) from hydrogen energy (prominent by the third harmonic). The dynamic characteristics of the time-varying harmonic characteristic matrix, combined with μs-level time synchronization, support accurate data comparison between distributed monitoring nodes. A normalization process ensures comparability of measurement results across power grids of different voltage levels. These features provide a highly reliable harmonic distribution spectrum for energy type identification, effectively supporting the precise matching of carbon emission factors.

[0150] See also Figure 3,Furthermore, monitoring the power fluctuation curve and extracting ,power fluctuation features include:

[0151] S201, performing a sliding time window analysis on the active power to calculate the power fluctuation rate;

[0152] S202, extracting time domain features of the power fluctuation curve, including:

[0153] Determine the periodic characteristics of power fluctuations;

[0154] S203, calculating the amplitude envelope characteristics of the power fluctuation;

[0155] S204: Perform frequency domain analysis on the power fluctuation to obtain frequency domain characteristics, including:

[0156] Extract the spectrum characteristics of power fluctuations through Fourier transform;

[0157] Identify the main frequency components of power fluctuations;

[0158] Generate frequency domain features based on main frequency components and spectrum characteristics;

[0159] S205 , fusing the power fluctuation rate, time domain features, amplitude envelope features, and frequency domain features to generate power fluctuation features.

[0160] In this embodiment, a sliding time window analysis is performed on the active power to calculate the power fluctuation rate. When extracting the time domain features of the power fluctuation curve, the periodic characteristics of the power fluctuation are determined by the autocorrelation function, and the amplitude envelope characteristics of the power fluctuation are calculated by the envelope detection algorithm. When performing frequency domain analysis on the power fluctuation, a fast Fourier transform (FFT) is performed to extract the spectral characteristics of the power fluctuation and identify the main frequency components of the power fluctuation (the typical photovoltaic frequency band is 0.5-2Hz). Frequency domain features are generated based on the main frequency components and spectral features. Finally, the power fluctuation rate, time domain features, amplitude envelope features, and frequency domain features are integrated to construct a power fluctuation feature containing a 12-dimensional feature vector, of which the frequency domain features account for 60% of the weight, ensuring accurate representation of the fluctuation characteristics of new energy.

[0161] In this embodiment, the sliding time window design takes into account both transient response speed (delay <50ms) and characteristic stability (power fluctuation calculation error ≤0.8%); frequency domain analysis effectively distinguishes the main frequency components of wind power (dominated by 1-5Hz) and load fluctuations (prominent at 0.1-0.5Hz), providing a key judgment basis for the dynamic matching of carbon emission factors.

[0162] Furthermore, the benchmark emission factor is dynamically weighted and corrected according to the real-time grid load rate and the dynamic correction coefficient to generate the real-time carbon emission factor corresponding to the current energy characteristic information, including:

[0163] Based on the energy composition ratio in the energy characteristic information, the benchmark emission factors corresponding to each energy type are extracted from the stratified carbon emission factor library;

[0164] Obtain the current grid load rate parameter and calculate the load rate correction weight, which is expressed by formula (1). Formula (1) is as follows:

[0165] W L =α·(1-e -β·L );

[0166] In formula (1), W L is the debt ratio correction weight, L is the real-time grid debt ratio, α is the first load ratio correction coefficient, β is the second debt ratio correction coefficient, and e is the base of the natural logarithm;

[0167] The initial revision of each baseline emission factor is made according to the dynamic revision factor, including:

[0168] Applying load rate correlation correction to the carbon emission factor of thermal power energy is expressed by formula (2), which is as follows:

[0169] E′ c =E c ·(1+γ·W L );

[0170] In formula (2), E′ c is the modified carbon emission factor of thermal power energy, E c is the benchmark carbon emission factor for thermal power energy, and γ is the thermal power load sensitivity coefficient;

[0171] Applying period validity correction to the carbon emission factor of renewable energy is expressed by formula (3), which is as follows:

[0172] E′ r =E r ·(1+η·A t );

[0173] In formula (3), E′ r is the carbon emission factor of renewable energy after correction, E r is the benchmark carbon emission factor of renewable energy, η is the renewable energy adjustment coefficient, A t is the time period availability rate;

[0174] The revised carbon emission factor is weighted twice using a dynamic weighting algorithm, including:

[0175] Determine the weight coefficient of each energy type according to the energy composition ratio;

[0176] Combined with the load factor correction weight to perform comprehensive weighted calculation;

[0177] The output matrix consists of the real-time carbon emission factors of each energy contribution, which is expressed by formula (4). Formula (4) is as follows:

[0178]

[0179] In formula (4), EF is the matrix of real-time carbon emission factors, E i ′ is the carbon emission factor of the i-th energy type after correction, P i is the energy composition ratio of the i-th energy type.

[0180] Furthermore, the hierarchical weighted algorithm is used to calculate the comprehensive carbon emissions, including:

[0181] The comprehensive carbon emissions within the preset period are obtained and expressed by formula (5), which is as follows:

[0182]

[0183] In formula (5), C Δt is the comprehensive carbon emissions, Δt is the preset time period, Ψ i (t) is the real-time power of energy of the i-th energy type, and t0 is the starting time of the preset time period;

[0184] The output includes a comprehensive carbon emission data set of time series, which is expressed by formula (6). Formula (6) is as follows:

[0185]

[0186] In formula (6), C Δt (t k ) is t k Comprehensive carbon emissions at the sampling time, t k is the kth sampling moment, and N is the total number of sampling points.

[0187] Furthermore, based on the energy composition change trend within the preset time window, the carbon emission intensity change rate for the next cycle is predicted to include:

[0188] Based on the energy characteristics information and the historical change rate of carbon emission intensity, perform the following steps:

[0189] Perform sliding window analysis on the energy composition ratio in the energy characteristic information to extract the temporal variation characteristics of the energy type proportion;

[0190] Modeling the correlation between temporal variation characteristics and the corresponding historical change rate of carbon emission intensity includes:

[0191] Establish a correlation coefficient matrix between changes in the proportion of each energy type and changes in carbon emission intensity;

[0192] Identify temporal cyclical patterns in the historical rate of change of carbon emission intensity;

[0193] Construct a carbon emission intensity prediction model based on the correlation coefficient matrix and time periodicity pattern;

[0194] The carbon emission intensity prediction model is used to predict the energy composition change trend of the next cycle and output the predicted value of the carbon emission intensity change rate;

[0195] The error analysis between the predicted value of the carbon emission intensity change rate and the real-time calculated carbon emission intensity change rate is carried out, and the correlation coefficient matrix and the parameters of the time periodic pattern are dynamically updated.

[0196] In this embodiment, a sliding window analysis is performed on the energy characteristic information within a preset time window to extract the time series change characteristics of the proportions of six types of energy, such as photovoltaics and wind power; a correlation coefficient matrix between the changes in the proportions of energy types and the historical change rates of carbon emission intensity is established to quantify the weights of the impact of various types of energy on carbon emissions; at the same time, spectrum analysis is used to identify the daily / weekly periodic patterns of the carbon emission intensity change rate; an ARIMA prediction model is constructed based on the correlation coefficient matrix and periodic characteristics, and after inputting real-time energy composition trend data, a predicted value of the carbon emission intensity change rate for the next period is output; finally, an error analysis is performed between the predicted value and the actual measured value, and when the threshold is exceeded, the least squares method is used to dynamically update the correlation coefficient matrix and periodic pattern parameters to ensure continuous optimization of the model.

[0197] In this embodiment, the correlation coefficient matrix accurately reflects the differentiated impact of different energy types on carbon emissions, making the correlation coefficient between the prediction results and the actual values ​​more similar; the dynamic parameter update mechanism enables the model to maintain a prediction error of ≤5.8% under the condition that the proportion of new energy fluctuates by ±15%; the periodic pattern recognition function effectively captures the differences in carbon emission patterns on weekdays / holidays, and the prediction accuracy is improved by 27%.

[0198] The power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters. Combined with the dynamic discrimination process of the energy type identification unit, the matching accuracy of the carbon emission factor is significantly improved. Through the joint analysis of the harmonic distribution characteristic spectrum and the power fluctuation characteristics, the energy type classification model can accurately identify the energy characteristic information of the current power grid; based on the dynamic matching mechanism of the hierarchical carbon emission factor library, the real-time grid load rate and dynamic correction coefficient are further used to perform weighted correction on the benchmark emission factor to generate a real-time carbon emission factor that is closer to reality. This method uses a hierarchical weighted algorithm to calculate the comprehensive carbon emissions and predict the rate of change of carbon emission intensity, realizing the refined dynamic accounting of carbon emission parameters. Finally, the complete carbon emission parameter set is uploaded to the supervision platform through the edge communication module, which solves the error problem caused by the traditional method relying on fixed emission factors, and provides high-precision and real-time data support for carbon management in industrial parks.

[0199] In addition, the functional units in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0200] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0201] The above descriptions are only some embodiments of the present invention and do not limit the scope of protection of the present invention. Any equivalent device or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A carbon emission parameter collection device, characterized in that: Applicable to a power grid, the device comprises: Power quality monitoring module, used to collect the original electrical signals of the power grid in real time and calculate the electrical signal parameters; a carbon emission calculation module, connected to the power quality monitoring module, for dynamically calculating carbon emission parameters based on electrical signal parameters and the energy type of the equipment in the power grid; the carbon emission calculation module includes an energy type identification unit and a carbon emission factor dynamic matching unit; the energy type identification unit automatically identifies the energy type by analyzing the harmonic characteristics and power fluctuation characteristics of the power grid; the carbon emission factor dynamic matching unit calls the corresponding carbon emission factor in real time according to the identified energy type for calculation; Edge communication module, which supports multi-protocol data encapsulation and transmission, and is used to upload carbon emission parameters to the regulatory platform; The local storage module is used to cache carbon emission parameters and support resumable downloads after network disconnection.

2. The carbon emission parameter collection device according to claim 1 is characterized in that: The original electrical signal includes an original voltage signal and an original current signal, and the power quality monitoring module includes: Voltage transformer, used to collect original voltage signals; Current transformer, used to collect original current signal; a signal conditioning circuit, configured to filter and amplify the collected original electrical signal to obtain a first processed signal; an analog-to-digital conversion unit, digitizing the first processed signal using synchronous sampling to obtain a second processed signal; The digital signal processor is used to perform real-time calculation on the second processed signal to obtain electrical signal parameters, which include electrical power parameters and harmonic content.

3. The carbon emission parameter collection device according to claim 1, characterized in that: The energy type identification unit includes: A harmonic characteristic analysis module is used to extract the harmonic characteristics of the power grid through FFT transformation; Power fluctuation analysis module, used to monitor power fluctuation characteristics; A machine learning classifier is configured to be trained based on historical data, and is used to classify energy types according to grid harmonic characteristics and power fluctuation characteristics.

4. A method for collecting carbon emission parameters, characterized in that: Applicable to the collection device according to any one of claims 1 to 3, the method comprising: The power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters; The energy type identification unit performs a dynamic energy type identification process, including: Perform real-time harmonic feature extraction on electrical signal parameters to obtain harmonic distribution characteristic spectrum; synchronously monitoring a power fluctuation curve and extracting power fluctuation characteristics, wherein the power fluctuation characteristics include fluctuation periodicity and amplitude characteristics; Input the harmonic distribution characteristic spectrum and power fluctuation characteristics into a pre-trained energy type classification model to output energy characteristic information of the current power grid; Inputting the energy characteristic information into the carbon emission factor dynamic matching unit to perform carbon emission factor matching, including: Obtaining a stratified carbon emission factor library, wherein the stratified carbon emission factor library includes baseline emission factors and dynamic correction factors corresponding to multiple energy types; Extracting corresponding benchmark emission factors from a stratified carbon emission factor library based on the energy characteristic information; Furthermore, the benchmark emission factor is dynamically weighted and corrected according to the real-time grid load rate and the dynamic correction coefficient to generate a real-time carbon emission factor corresponding to the current energy characteristic information; Dynamically calculate carbon emission parameters based on electrical signal parameters and matching carbon emission factors, including: A hierarchical weighted algorithm is used to calculate comprehensive carbon emissions; Based on the energy composition change trend within the preset time window, predict the carbon emission intensity change rate in the next cycle; The complete carbon emission parameter set including energy characteristic information, real-time carbon emission factors, comprehensive carbon emissions and carbon emission intensity change rate is uploaded to the supervision platform through the edge communication module.

5. The carbon emission parameter collection method according to claim 4 is characterized in that: The power quality monitoring module collects the original electrical signals of the power grid in real time and calculates the electrical signal parameters, including: The original voltage signal is collected through a voltage transformer, and the original current signal is collected through a current transformer; Performing signal conditioning processing on the original voltage signal and the original current signal includes: Perform voltage division and filtering on the original voltage signal to obtain a conditioned voltage signal; Convert and filter the original current signal to obtain a conditioned current signal; The conditioned voltage and current signals are synchronously sampled and converted into analog-to-digital signals, including: The phase-locked loop method is used to track the fundamental frequency of the power grid and realize the synchronous sampling of voltage and current signals; Converting the analog signal obtained by synchronous sampling into a digital signal, the digital signal including a digital voltage signal and a digital current signal; Perform real-time computation and processing on digital signals, including: Calculating real-time electric power parameters based on the digital voltage signal and the digital current signal, wherein the electric power parameters include active power, reactive power, and apparent power; The harmonic components of the digital voltage signal and the digital current signal are extracted by fast Fourier transform, and the harmonic content parameters are calculated. The harmonic content parameters include the total harmonic distortion rate and the content rate of each harmonic; the electrical signal parameters are generated according to the electrical power parameters and the harmonic content parameters.

6. The carbon emission parameter collection method according to claim 5 is characterized in that: Real-time harmonic feature extraction of electrical signal parameters to obtain harmonic distribution characteristic spectrum includes: Normalizing the harmonic content parameters to generate a standardized harmonic amplitude spectrum; The distribution characteristics of the characteristic harmonic group are extracted based on the standardized harmonic amplitude spectrum, including: Determine the amplitude ratio relationship of the 3rd, 5th, and 7th characteristic harmonics; Calculate the energy gradient change rate between each harmonic; Based on the harmonic content parameters of continuous sampling periods, a time-varying harmonic characteristic matrix is ​​constructed, including: Perform sliding window analysis on the normalized harmonic amplitude spectrum; Extract the time series variation characteristics of each harmonic phase angle; The distribution characteristics of the characteristic harmonic group are fused with the time-varying harmonic characteristic matrix to generate the harmonic distribution characteristic spectrum.

7. The carbon emission parameter collection method according to claim 5, characterized in that: Monitoring power fluctuation curves and extracting power fluctuation features include: Performing a sliding time window analysis on the active power to calculate the power fluctuation rate; Extract the time domain features of the power fluctuation curve, including: Determine the periodic characteristics of power fluctuations; Calculate the amplitude envelope characteristics of power fluctuations; Perform frequency domain analysis on power fluctuations to obtain frequency domain characteristics, including: Extract the spectrum characteristics of power fluctuations through Fourier transform; Identify the main frequency components of power fluctuations; Generating the frequency domain features according to the main frequency components and the spectrum features; The power fluctuation rate, time domain features, amplitude envelope features and frequency domain features are fused to generate the power fluctuation features.

8. The carbon emission parameter collection method according to claim 4 is characterized in that: The benchmark emission factor is dynamically weighted and corrected based on the real-time grid load rate and the dynamic correction coefficient to generate the real-time carbon emission factor corresponding to the current energy characteristic information, including: Extracting a benchmark emission factor corresponding to each energy type from the stratified carbon emission factor library based on the energy composition ratio in the energy characteristic information; Obtain the current grid load rate parameter and calculate the load rate correction weight, which is expressed by formula (1). Formula (1) is as follows: W L =α·(1-e -β·L ); In formula (1), W L is the debt ratio correction weight, L is the real-time grid debt ratio, α is the first load ratio correction coefficient, β is the second debt ratio correction coefficient, and e is the base of the natural logarithm; The initial revision of each baseline emission factor is made according to the dynamic revision factor, including: The load rate correlation correction is applied to the carbon emission factor of thermal power energy, which is expressed by formula (2). The formula (2) is as follows: E′ c JE c ·(1+γ·W L )4 In formula (2), E′ c is the modified carbon emission factor of thermal power energy, E c is the benchmark carbon emission factor for thermal power energy, and γ is the thermal power load sensitivity coefficient; Applying period validity correction to the carbon emission factor of renewable energy is expressed by formula (3), which is as follows: AND' r =And r ·(1+η·A t ); In formula (3), E′ r is the carbon emission factor of renewable energy after correction, E r is the benchmark carbon emission factor of renewable energy, η is the renewable energy adjustment coefficient, A t is the time period availability rate; The revised carbon emission factor is weighted twice using a dynamic weighting algorithm, including: Determine the weight coefficient of each energy type according to the energy composition ratio; Combined with the load factor correction weight to perform comprehensive weighted calculation; The output matrix includes the real-time carbon emission factors of each energy contribution, which is expressed by formula (4). The formula (4) is as follows: In formula (4), EF is the matrix of real-time carbon emission factors, E i ′ is the carbon emission factor of the i-th energy type after correction, P i is the energy composition ratio of the i-th energy type.

9. The carbon emission parameter collection method according to claim 8, characterized in that: The tiered weighted algorithm used to calculate comprehensive carbon emissions includes: The comprehensive carbon emissions within the preset period are obtained and expressed by formula (5), which is as follows: In formula (5), C Δt is the comprehensive carbon emissions, Δt is the preset time period, Ψ i (t) is the real-time power of energy of the i-th energy type, and t0 is the starting time of the preset time period; The output includes a comprehensive carbon emission data set of time series, which is expressed by formula (6), which is as follows: In formula (6), C Δt (t k ) is t k Comprehensive carbon emissions at the sampling time, t k is the kth sampling moment, and N is the total number of sampling points.

10. The carbon emission parameter collection method according to claim 4, characterized in that: Based on the energy composition change trend within the preset time window, the predicted carbon emission intensity change rate for the next cycle includes: Based on the energy characteristic information and the historical change rate of carbon emission intensity, the following steps are performed: Performing a sliding window analysis on the energy composition ratio in the energy characteristic information to extract the time series variation characteristics of the energy type ratio; Modeling the association between the temporal variation characteristics and the corresponding historical change rate of carbon emission intensity includes: Establish a correlation coefficient matrix between changes in the proportion of each energy type and changes in carbon emission intensity; Identifying a temporal periodic pattern in the historical rate of change of the carbon emission intensity; Constructing a carbon emission intensity prediction model based on the correlation coefficient matrix and the time periodic pattern; The carbon emission intensity prediction model is used to predict the energy composition change trend of the next cycle and output a predicted value of the carbon emission intensity change rate; An error analysis is performed between the predicted value of the carbon emission intensity change rate and the real-time calculated carbon emission intensity change rate, and the parameters of the correlation coefficient matrix and the time periodicity pattern are dynamically updated.

Images