A carbon emission comparison and evaluation method based on coal-fired power supply unit energy storage type

Abstract

The application provides a coal-fired power supply unit energy storage type carbon emission comparison and evaluation method, comprising: aligning the thermal steam pressure temperature and the flue gas carbon flow in the same time window according to the real-time monitoring data set, using a timestamp matching method to process the heat supply extraction steam flow enthalpy value, and forming time-aligned monitoring data; triggering carbon accounting sampling with the backwater temperature inflection point position, extracting the thermal steam pressure temperature, heat supply extraction steam flow enthalpy value and flue gas carbon flow in the time window before and after the inflection point from the time-aligned monitoring data, and obtaining a triggered sampling data set; and evaluating the steam turbine power generation load from the thermal steam pressure and temperature in the triggered sampling data set, calculating the instantaneous carbon emission intensity by dividing the power generation load by the flue gas carbon flow, and forming an instantaneous carbon emission intensity record.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a comparative assessment method for carbon emissions based on energy storage in coal-fired power generation units. Background Technology

[0002] In the energy sector, the flexibility retrofitting of coal-fired power units is crucial for promoting low-carbon development and optimizing the energy structure. The power system faces the dual pressures of peak-shaving capacity and heating demand, especially during the peak winter heating season, when units need to frequently switch from high load to low load to respond to fluctuations in renewable energy while maintaining stable heating supply from the regional heating network. Research in this area not only concerns improving energy efficiency but also directly impacts the achievement of environmental protection goals.

[0003] However, current carbon emission analysis methods for flexible retrofitting of coal-fired power units have significant shortcomings. Many methods often overlook the dynamic interaction of various factors during operation, especially in scenarios involving low-load peak shaving and heating coupling, failing to fully consider the impact of heating network operating characteristics on carbon emissions. The heating network system itself has a large heat storage capacity, and its core characteristic parameter—the return water temperature—does not respond instantly to changes in unit load, but rather exhibits a significant lag. This lag causes a temporal misalignment between the release of heat storage in the heating network and boiler coal consumption, thus interfering with the assessment of carbon emission intensity for power generation, leading to discrepancies between the analysis results and actual operating conditions, and making it difficult to accurately reflect the true effect of the retrofitting scheme. This temporal misalignment makes traditional carbon emission accounting methods based on power generation load unsuitable for complex operating scenarios, especially when heating steam extraction and boiler coal consumption cannot be synchronized, the trend of carbon emission intensity changes may exhibit unexpected fluctuations or stagnation.

[0004] Specifically, when coal-fired units enter a low-load peak-shaving state, the amount of steam extracted for heating decreases due to the slow change in the return water temperature of the heating network. However, the amount of coal burned in the boiler may not decrease synchronously, causing carbon emission intensity to fail to rise as expected in the short term, and instead may remain unchanged or even decrease temporarily. This inconsistency between operating parameters and carbon emissions poses a significant challenge to accurately assessing the effectiveness of retrofit programs. Therefore, accurately capturing the impact of dynamic changes in the return water temperature of the heating network on carbon emission intensity during the heat storage and release phase has become a key issue that this study urgently needs to address. Summary of the Invention

[0005] This invention provides a comparative assessment method for carbon emissions from coal-fired power generation units with energy storage. The method is applied to coal-fired power generation systems equipped with thermal energy storage devices, and includes: By connecting the power grid dispatch and control system of the power supply unit to the coal-fired power generation and power supply system, the original records of hot steam pressure and temperature, heating extraction steam flow and enthalpy, flue gas carbon flow and heating network return water temperature are collected to obtain a real-time monitoring dataset that is synchronized with time. Based on the real-time monitoring dataset, the hot steam pressure and temperature and flue gas carbon flow rate within the same time window are aligned, and the enthalpy value of the heating extraction steam flow rate is processed to form monitoring data aligned with the grid dispatch timing. By analyzing monitoring data aligned with the grid dispatch timing, the peak power supply load of the power supply unit is assessed. The peak power supply load reflects the real-time power distribution output of the power supply unit to the power system. The instantaneous carbon emission intensity is obtained by dividing the flue gas carbon flow by the peak power supply load, thus forming an instantaneous carbon emission intensity record. By comparing the instantaneous carbon emission intensity record with the enthalpy value of the heating extraction steam flow rate, the range of decrease in the enthalpy value of the heating extraction steam flow rate is identified. The range of decrease corresponds to the process in which the thermal energy storage device in the power storage system releases heat energy to compensate for the power output. The plateau or swing pattern of the intensity record within the range of decrease is analyzed to determine the stage identifier of the energy storage and thermal energy release. The energy storage and thermal storage release stages are divided into energy storage release sub-stages and non-release sub-stages based on the energy storage and thermal storage release stage identifiers. The instantaneous carbon emission intensity of each flexibility transformation group under the grid peak-shaving power supply condition is compared, and the difference in the impact of each transformation scheme on the carbon emission intensity of the power supply side of the power system is output. The comparative evaluation conclusion of the power supply carbon emission of each transformation scheme is obtained.

[0006] Preferably, by connecting the power grid dispatch control system of the power supply unit to the coal-fired power generation and supply system, and collecting raw records of hot steam pressure and temperature, heating extraction steam flow rate and enthalpy, flue gas carbon flow rate, and heating network return water temperature, a time-synchronized real-time monitoring dataset is obtained, including: A continuous flue gas monitoring device is installed in the flue gas emission channel. Pressure transmitters and temperature transmitters are respectively installed in the main steam pipeline and the reheat steam pipeline. A vortex flow meter and temperature and pressure compensation measuring points are installed in the heating extraction steam branch line. A platinum resistance thermometer is arranged in the return water header of the heating network. A unified clock reference is sent to all monitoring devices using the Network Time Protocol to align the local clocks of each device with the main clock of the power grid dispatch and control system of the power supply units. Data from each channel is sent up at the same sampling interval. The cleaned records from each channel are merged into the same storage table with a unified timestamp to obtain the real-time monitoring dataset that is synchronized with the time.

[0007] Preferably, the step of aligning the hot steam pressure and temperature with the flue gas carbon flow rate within the same time window based on the real-time monitoring dataset, processing the enthalpy value of the heating extraction steam flow rate, and forming monitoring data aligned with the grid dispatch timing includes: Using the master clock of the power grid dispatch and control system of the power supply unit as the alignment reference, a continuous and non-overlapping time window is defined according to a preset duration. The sampling points of hot steam pressure and temperature and flue gas carbon flow rate that fall within the same time window are grouped into the same row according to the timestamp to obtain the initial aligned record; For each target timestamp in the initial alignment record, the sampling point with the smallest difference from the target timestamp in the heating steam extraction flow enthalpy channel is found. The heating steam extraction flow field and enthalpy field are written into the heating steam extraction flow enthalpy channel with a difference not exceeding the preset tolerance threshold. For the difference exceeding the threshold, the empty fields are filled with linear interpolation to obtain the monitoring data aligned with the power grid scheduling time sequence.

[0008] Preferably, the method further includes: Extract the return water temperature segment of the heating network from the monitoring data aligned with the power grid dispatch time sequence, and identify the inflection point of the return water temperature when the temperature change rate is continuously lower than the preset threshold and exceeds the preset period by calculating the temperature change rate at continuous time points.

[0009] Preferably, the step of extracting the return water temperature segment from monitoring data aligned with the power grid dispatch time sequence, and identifying the inflection point of the return water temperature where the temperature change rate is continuously lower than a preset threshold and exceeds a preset period by calculating the temperature change rate at consecutive time points, includes: Based on the start and end timestamps of the power supply units entering the grid under low-load peak-shaving power supply and heating coupling conditions, the original records of the heat network return water temperature covering the complete peak-shaving power supply process are extracted to obtain the return water temperature sub-segment. The return water temperature segment is smoothed by sliding median filtering with an odd window length to obtain a smoothed return water temperature segment. The temperature rate sequence is obtained by dividing the temperature difference between adjacent sampling points in the smoothed return water temperature segment by the timestamp interval. A windowed temperature rate sequence is obtained by taking the absolute value of the temperature rate sequence and calculating its arithmetic mean using a sliding window of preset width. In the windowed temperature rate sequence, all intervals with values ​​lower than a preset rate threshold and a duration not less than a preset duration threshold are retained, and their starting timestamps are taken as the inflection point of the return water temperature. The inflection point of the return water temperature marks the start time of energy release of the thermal energy storage device under the peak power supply condition of the power grid.

[0010] Preferably, the method further includes: Carbon sampling is triggered by the inflection point of the return water temperature. The hot steam pressure and temperature, heating extraction steam flow rate and enthalpy, and flue gas carbon flow rate within the time window before and after the inflection point are extracted to obtain the triggered sampling dataset.

[0011] Preferably, carbon accounting sampling is triggered at the inflection point of the return water temperature, and the hot steam pressure and temperature, heating extraction steam flow rate and enthalpy, and flue gas carbon flow rate within the time window before and after the inflection point are extracted to obtain the triggered sampling dataset, which specifically includes: Based on the inflection point of the return water temperature, the left boundary of the window is obtained by extending it before the inflection point by a preset pre-time, and the right boundary of the window is obtained by extending it after the inflection point by a preset post-time. The carbon accounting sampling time window is defined by the left boundary and the right boundary of the window. The trigger sampling dataset is obtained by retrieving all record rows whose timestamps fall within the sampling time window from the monitoring data aligned with the power grid dispatch timing and sorting them in ascending order of timestamps.

[0012] Preferably, the step of evaluating the peak-shaving power supply load of the power generation unit by analyzing monitoring data aligned with the grid dispatch timing, wherein the peak-shaving power supply load reflects the real-time power distribution output of the power generation unit to the power system, and obtaining the instantaneous carbon emission intensity by dividing the flue gas carbon flow by the peak-shaving power supply load, thereby forming an instantaneous carbon emission intensity record, includes: The main steam pressure and temperature are extracted from the real-time monitoring dataset row by row according to the timestamp. The pre-established correspondence curve between the steam inlet parameters of the power supply unit and the power supply load is retrieved. The main steam pressure and temperature are interpolated bilinearly to obtain the main steam power supply load component. The reheat steam power supply load component is obtained by using the same interpolation process for reheat steam pressure and temperature at the same timestamp. The instantaneous peak-shaving power supply load is obtained by summing the two components; The instantaneous carbon emission intensity record is obtained by dividing the flue gas carbon flow rate by the instantaneous peak power supply load and arranging them in ascending order of timestamps.

[0013] Preferably, the step of identifying the decreasing range of the enthalpy value of the heating extraction steam flow rate by comparing the instantaneous carbon emission intensity record with the enthalpy value of the heating extraction steam flow rate, wherein the decreasing range corresponds to the process of the thermal energy storage device in the energy storage system releasing heat energy to compensate for the power supply output, and analyzing the plateau or swing pattern exhibited by the intensity record within the decreasing range to determine the energy storage and thermal energy release stage identifier, includes: The instantaneous carbon emission intensity record is aligned with the enthalpy value of the heating extraction steam flow rate in the triggered sampling dataset by timestamp; For each timestamp, the product of the steam extraction flow rate and enthalpy is taken, and the sequence of the product of the steam extraction flow rate and enthalpy is obtained by arranging them in ascending order of timestamp. Dividing the product difference between adjacent sampling points by the timestamp interval yields the product change rate sequence. In the product change rate sequence, if the rate of continuous intervals are all negative and the absolute value is not lower than the preset decrease rate threshold, and the duration is not less than the preset duration threshold, it is marked as the heating steam extraction flow enthalpy value reduction interval. This reduction interval corresponds to the process in which the thermal energy storage device in the electric energy storage system releases stored thermal energy to the heating network to maintain the peak power output of the power supply unit. The instantaneous carbon emission intensity record is determined to be a plateau pattern if the range of the subsequences within the reduction range does not exceed a preset fluctuation amplitude threshold. If a reverse reversal feature appears in the subsequence, the reversal amplitude is determined by the reverse reversal feature. If the reversal amplitude exceeds a preset reversal amplitude threshold, it is determined to be a swing pattern. For the reduction range that presents a platform shape or a swing shape, the energy storage and heat release stage identifier is determined.

[0014] Preferably, the process involves dividing the energy storage and thermal storage into energy release sub-stages and non-release sub-stages based on the energy storage and thermal storage release stage identifier, comparing the instantaneous carbon emission intensity of each flexibility modification group under the grid peak-shaving power supply condition, outputting the difference in the impact of each modification scheme on the carbon emission intensity of the power supply side of the power system, and obtaining the comparative evaluation conclusion of the power supply carbon emission of each modification scheme, including: The entire process of power grid low-load peak power supply and heating is divided on the time axis according to the start and end timestamps in the energy storage and thermal release stage identifier. The time interval covered by the energy storage and thermal release stage identifier is marked as the energy storage release sub-stage, and the remaining time intervals not covered are marked as non-release sub-stages. For each sub-stage, the intensity values ​​are retrieved from the instantaneous carbon emission intensity records and the arithmetic mean is calculated to obtain the average instantaneous carbon emission intensity of the sub-stage. The difference and ratio of the average instantaneous carbon emission intensity between the energy storage release sub-stage and the non-release sub-stage of each flexibility modification group are calculated, and the power supply carbon emission comparison evaluation conclusion of each modification scheme is obtained by summarizing them.

[0015] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a comparative assessment method for carbon emissions from coal-fired power generation units based on energy storage. Addressing the issue of varying carbon emission intensity differences in coal-fired units during low-load peak shaving and heating processes, it innovatively utilizes real-time monitoring dataset collection and time alignment, combined with sampling triggered by return water temperature inflection points, to accurately assess instantaneous carbon emission intensity and identify the heat storage release phase. Through comprehensive analysis of main and reheat steam pressure and temperature, heating extraction steam flow rate and enthalpy, and flue gas carbon flow rate, this invention divides the process into heat storage release and non-release sub-phases, extracts the average carbon emission intensity of each sub-phase, and finally compares the carbon emission effects of different flexibility retrofit schemes. The core of this invention lies in its refined carbon emission analysis method based on return water temperature inflection points and heat storage release phase identification. This method achieves precise quantification of the carbon emission impact of flexible retrofit schemes for coal-fired units, providing a scientific basis for optimizing low-load peak shaving and heating strategies, and significantly improving energy utilization efficiency and emission reduction effects. Attached Figure Description

[0016] Figure 1 This is a flowchart of a carbon emission comparison and assessment method based on energy storage of coal-fired power generation units according to the present invention.

[0017] Figure 2 This is a schematic diagram of a carbon emission comparison and evaluation method based on energy storage of coal-fired power generation units according to the present invention.

[0018] Figure 3 This is another schematic diagram of a carbon emission comparison and evaluation method based on energy storage of coal-fired power generation units according to the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figures 1 to 3 As shown in the figure, the carbon emission comparison and assessment method based on coal-fired power generation unit energy storage in this embodiment may specifically include: In the scenario of energy storage-type flexibility retrofitting of coal-fired power generation units, the power generation unit is equipped with a thermal energy storage device. The thermal energy storage device can be a molten salt thermal storage tank, a solid electric thermal storage body, or a hot water thermal storage tank connected in parallel with the heating extraction steam branch line. It stores thermal energy under low load peak-shaving power supply conditions and releases thermal energy to the heating network to compensate for the reduction in heating steam extraction when the peak-shaving depth increases. The thermal energy storage device and the heating network together constitute an electrical energy storage system. The unit control system, as the field execution layer of the power grid dispatch control system, is connected to the power system power supply and distribution monitoring network, receives peak-shaving power output instructions issued by the dispatch side, and transmits back real-time operating parameters of the unit. In this embodiment, the "unit control system" mentioned later refers to the field execution layer of the aforementioned power grid dispatch control system, and the "time-aligned monitoring data" mentioned later refers to the monitoring data aligned with the power grid dispatch timing.

[0021] S101. By deploying a continuous flue gas monitoring device and a steam extraction flow meter through the unit control system, the original records of main and reheat steam pressure and temperature, heating extraction steam flow enthalpy, flue gas carbon flow, and heating network return water temperature are collected to obtain a real-time monitoring dataset that is synchronized with the time.

[0022] A continuous flue gas monitoring device is installed in the flue gas emission channel of the unit control system. Pressure transmitters and temperature transmitters are configured for the main steam pipeline and reheat steam pipeline, respectively. Vortex flow meters and temperature and pressure compensation measuring points are installed on the heating extraction steam branch line. Platinum resistance thermometers are arranged on the return water header of the heating network. The original records of flue gas carbon flow are obtained through the continuous flue gas monitoring device. The original records of main and reheat steam pressure and temperature are obtained through the pressure transmitter and temperature transmitter. The original records of heating extraction steam flow are obtained through the vortex flow meter. The original records of heating extraction steam enthalpy are obtained by looking up the steam property table based on the temperature and pressure measured by the temperature and pressure compensation measuring points. The original records of heating network return water temperature are obtained through the platinum resistance thermometer, resulting in multi-source original monitoring records. For the aforementioned multi-source raw monitoring records, a unified clock reference is sent to each monitoring device using a network time protocol to align the local clocks of the flue gas continuous monitoring device, pressure transmitter, temperature transmitter, vortex flow meter, and platinum resistance thermometer with the main clock of the unit control system. The sampling periods for the main and reheat steam pressure and temperature, heating extraction steam flow and enthalpy, flue gas carbon flow, and heating network return water temperature are uniformly set, and data transmission from each channel is triggered at the same sampling interval to obtain multi-channel monitoring records with a unified timestamp. For the multi-channel monitoring records with a unified timestamp, distortion identification is performed on the measuring points of each channel by range over-limit judgment and change rate threshold judgment. Abnormal jumps in the enthalpy of the heating network return water temperature and heating extraction steam flow are eliminated. The zero drift offset of the flue gas carbon flow is corrected by baseline correction. The instantaneous peaks of the main and reheat steam pressure and temperature are filtered by sliding median. The cleaned records of each channel are merged into the same storage table according to the unified timestamp. Linear interpolation is used to fill in the short-term data gaps between adjacent sampling points to obtain a time-synchronized real-time monitoring dataset.

[0023] In the low-load peak-shaving and heating coupling conditions of coal-fired power unit flexibility retrofits, the synchronization and accuracy of monitoring data are fundamental to conducting carbon emission assessments. The unit control system, acting as a distributed control layer, is responsible for coordinating the operating parameters of main and auxiliary equipment, and also undertakes the acquisition and aggregation of signals from multiple monitoring points. To support subsequent identification of return water temperature inflection points and judgment of the heat storage release stage, all monitoring devices must have complete spatial coverage and strict temporal alignment. A continuous flue gas monitoring device is installed at the sampling section of the desulfurization and denitrification outlet section of the tail flue, including an infrared or non-dispersive infrared principle carbon dioxide concentration measurement unit, a Pitot tube or ultrasonic flow element, and outputs the converted flue gas carbon flow rate. Pressure transmitters and temperature transmitters are respectively installed at the pressure measuring sections and temperature sleeves of the main steam pipeline and the high-pressure reheat steam pipeline. A vortex flowmeter with temperature and pressure compensation measuring points is installed on the straight pipe section of the heating extraction steam branch line near the regulating valve, and a platinum resistance thermometer is installed before the heat exchanger at the first station of the heating network enters the return water header. The instantaneous temperature and pressure at the output section of the temperature and pressure compensation measuring point are used by the unit control system to look up the pre-stored steam property table based on the temperature and pressure, and use bilinear interpolation to obtain the corresponding superheated steam specific enthalpy. This specific enthalpy is stored in pairs with the heating extraction steam flow rate output by the vortex flow meter at the same timestamp, forming two parallel fields of the original record of the heating extraction steam flow enthalpy value. This avoids the ambiguity of mistakenly treating the enthalpy value as a direct measurement quantity.

[0024] The multi-source monitoring devices are distributed across different fieldbus segments and control cabinets, resulting in cumulative drift of local crystal oscillator frequencies ranging from milliseconds to seconds. The unit control system's master clock is bound to a satellite time source, serving as a time reference server. It periodically sends synchronization messages to the field-level nodes containing the continuous flue gas monitoring device, pressure transmitter, temperature transmitter, vortex flow meter, and platinum resistance thermometer. The nodes calculate the deviation between the timestamp carried in the message and their local receiving time, and progressively correct their local clocks. The sampling period for each parameter is uniformly set to the same whole-second interval. The unit control system triggers the data upload of each channel at a unified pace, resulting in multi-channel monitoring records with uniform timestamps.

[0025] During on-site monitoring, it is difficult to avoid instances of exceeding the upper and lower limits of the measurement range, sensor jitter, and noise spikes. The range exceedance judgment sets thresholds for each channel based on the upper and lower limits of the rated range on the equipment nameplate; instantaneous values ​​exceeding these thresholds are marked as suspicious points. The rate of change threshold judgment targets the difference between two adjacent sampling points. If the change in the return water temperature of the heating network exceeds a preset rate threshold within a sampling period, it is judged as an abnormal jump. This preset rate threshold is significantly higher than the rate of change of the return water temperature under normal operating conditions of the heating network, and is set based on the upper limit of the allowable physical change rate of the platinum resistance thermometer, set at 0.3℃ / s, to eliminate sudden jumps caused by instantaneous electrical interference from the sensor. Abnormal jumps in the enthalpy value of the heating steam extraction flow rate are also identified and eliminated using the above rate of change criterion, leaving a vacancy for subsequent interpolation processing. For the flue gas carbon flow rate, zero drift offset manifests as the sensor's zero point slowly deviating from the true zero position after prolonged operation.

[0026] The average value during unit shutdown or low-level stable operation is selected as the zero-point reference. This reference offset is uniformly subtracted from the online measurements to obtain the baseline-corrected flue gas carbon flow rate. For the instantaneous peaks in the main and reheat steam pressure and temperature, a sliding median filter with an odd-length window is used. Each sampling point is replaced with the median of all samples within the window centered on that point, thus suppressing single-point noise while preserving trend changes. Finally, the cleaned records from each channel are merged into the same relational storage table with a unified timestamp. Each row corresponds to a timestamp, and the column fields cover all the aforementioned parameters. Short-term data gaps between adjacent sampling points are filled using linear interpolation, resulting in a time-synchronized real-time monitoring dataset, providing complete basic data support for subsequent processing.

[0027] S102. Based on the real-time monitoring dataset, align the pressure and temperature of the main and reheat steam and the carbon flow rate of the flue gas within the same time window, and use the timestamp matching method to process the enthalpy value of the heating extraction steam flow rate to form monitoring data aligned with the grid dispatching time sequence.

[0028] Based on the real-time monitoring dataset synchronized with the time, and using the master clock of the unit control system as the alignment reference, continuous non-overlapping time windows are defined according to a preset duration. For the two channels of main and reheat steam pressure and temperature and flue gas carbon flow, all sampling points falling into the same time window are extracted, arranged in ascending order of timestamp, and the two records at the same sampling time are grouped into the same row. For individual sampling points that are misaligned due to transmission delay, they are grouped into the time window with the smallest difference based on the timestamp, thus obtaining the initial alignment record of main and reheat steam pressure and temperature and flue gas carbon flow aligned in the same window. For the initial alignment record, all sampling points of the heating steam extraction flow and enthalpy channel are retrieved. Using a timestamp matching method, for each row of the initial alignment record with a target timestamp, the sampling point with the smallest difference from the target timestamp is found in the heating steam extraction flow and enthalpy channel. If the difference does not exceed a preset time tolerance threshold, the heating steam extraction flow and enthalpy fields of that sampling point are written to that row. If the difference exceeds the preset time tolerance threshold, the heating steam extraction flow and enthalpy fields of that row are marked as gaps to be filled, thus obtaining an intermediate alignment record with the steam extraction parameters filled in. For the rows marked as having gaps to be filled in the intermediate alignment record, retrieve the two most recent valid heating extraction steam flow and enthalpy sampling points before and after its timestamp, and use linear interpolation to fill in the gap fields; sort the filled intermediate alignment record in ascending order according to a unified timestamp, and normalize the order of the fields of main and reheat steam pressure and temperature, flue gas carbon flow, heating extraction steam flow and enthalpy, to obtain monitoring data aligned with the power grid dispatching time sequence.

[0029] Under the low-load peak-shaving and heating coupling conditions of flexible retrofitting of coal-fired units, the three types of parameters—main and reheat steam pressure and temperature, flue gas carbon flow rate, and heating extraction steam flow rate and enthalpy—although originating from the same unit, belong to different fieldbus nodes and acquisition cards. Even after alignment with a unified clock reference, millisecond- to sub-second time misalignments will still occur due to differences in transmission path length, signal conditioning delay, and internal buffer depth of the sampling card. The goal of the alignment process is to eliminate these misalignments at the data level, providing consistent parallel records for subsequent return water temperature inflection point identification and instantaneous carbon emission intensity calculation.

[0030] The unit control system's master clock is bound to a satellite timing source, outputting an alignment pulse at every whole second. Continuous, non-overlapping time windows are defined according to a preset duration. The left boundary of each time window strictly coincides with the whole second pulse of the master clock, and the right boundary is the left boundary plus the preset duration. Each time window corresponds to a target timestamp, which is taken as the midpoint or left boundary of the time window, serving as the alignment anchor point for all records within that window.

[0031] The initial alignment record is formed as follows: The original sampling sequences of the main and reheat steam pressure and temperature channels and the flue gas carbon flow channel are retrieved from the time-synchronized real-time monitoring dataset and read in ascending order of timestamps. For each sampling record, it is determined which time window its timestamp falls into; two records falling into the same time window are grouped into the same row. When a record's timestamp is near the boundary of two adjacent time windows due to transmission delay, its distance from the midpoint of the left and right windows is compared, and it is grouped into the side with the smaller distance. After this processing, the main and reheat steam pressure and temperature and the flue gas carbon flow are arranged row by row according to the target timestamps, forming the initial alignment record.

[0032] The processing method for the heating steam extraction flow and enthalpy channel differs from the aforementioned window alignment. The acquisition triggering cycle of the heating steam extraction flow and enthalpy channel has an inherent phase difference with the main, reheat steam, and flue gas channels, requiring a timestamp matching method. This timestamp matching method is implemented as follows: For each row in the initial alignment record, its target timestamp is denoted as t0. All sampling points of the heating steam extraction flow and enthalpy channel are traversed, and the absolute difference between the timestamp of each sampling point and t0 is calculated. The sampling point with the smallest difference is selected. If the difference does not exceed a preset time tolerance threshold, the heating steam extraction flow and enthalpy fields of that sampling point are written to that row; if the difference exceeds the preset time tolerance threshold, the heating steam extraction flow and enthalpy fields of that row are marked as needing to be filled.

[0033] The preset time tolerance threshold is set to half of a sampling period. In this embodiment, the sampling period is uniformly set to 1 second according to the data acquisition procedure of the unit control system. Therefore, the preset time tolerance threshold is 0.5 seconds, thereby ensuring that the matched sampling point is closest to t0 in time. After this processing, the intermediate alignment record is obtained, which contains complete main and reheat steam pressure and temperature and flue gas carbon flow fields, as well as extraction steam flow enthalpy value fields that are partially written or partially marked as empty.

[0034] For rows marked as having missing spaces in the intermediate alignment records, backtrack to the most recent valid sampling point for heating extraction steam flow and enthalpy values ​​with a timestamp less than t0, and record its timestamp as t1, flow value as f1, and enthalpy value as h1; then search forward to the most recent valid sampling point with a timestamp greater than t0, and record its timestamp as t2, flow value as f2, and enthalpy value as h2. Linearly interpolate the missing fields according to the timestamps. Sort the completed intermediate alignment records in ascending order according to a unified timestamp. Standardize the fields of main and reheat steam pressure and temperature, flue gas carbon flow, heating extraction steam flow, and enthalpy values ​​according to a pre-set column order to obtain monitoring data aligned with the grid dispatch timeline. Each row in the monitoring data aligned with the grid dispatch timeline corresponds to a target timestamp, and the values ​​of each field in each row strictly correspond to the unit operating status at the same time.

[0035] The monitoring data, aligned with the grid dispatch time sequence, is presented as a two-dimensional table structure with the target timestamp as the primary key. The column fields cover the flue gas carbon flow rate on the boiler combustion side, the main and reheat steam pressure and temperature on the turbine inlet side, and the flow rate and enthalpy on the heating extraction side. This provides a time-consistent operating parameter basis for subsequently extracting the heating network return water temperature segment from this data, identifying the inflection point of the return water temperature, and triggering carbon accounting sampling.

[0036] S103. Extract the return water temperature segment of the heating network from the monitoring data aligned with the power grid dispatch time sequence. By calculating the temperature change rate at continuous time points, analyze the time period when the temperature change rate is continuously lower than the preset temperature change rate threshold and exceeds the preset duration period, and identify the inflection point of the return water temperature.

[0037] Based on the monitoring data aligned with the grid dispatch timing, the return water temperature sequence covering the entire peak shaving process is extracted according to the start and end timestamps of the units entering the low-load peak shaving and heating coupling operation, resulting in a return water temperature segment. For the return water temperature segment, the sampling uniformity is confirmed based on the timestamp interval of each sampling point within the segment. For local measuring points with slight jumps, a sliding median filter with an odd window length is used for smoothing. The smoothed sequence of the return water temperature segment is then rearranged in ascending order of timestamps, and incomplete sampling points at the beginning and end of the transition segment are removed to obtain the smoothed return water temperature segment. For the smoothed return water temperature segment, the temperature difference between two adjacent sampling points is divided by the interval between two adjacent timestamps to obtain the temperature rate of that adjacent sampling point pair. The temperature rates of all adjacent sampling point pairs in the smoothed return water temperature segment are arranged in ascending order of timestamps to obtain a temperature rate sequence. For the temperature rate sequence, a sliding window of a preset width is used to advance point by point. The absolute value of the temperature rate within each sliding window is taken and the arithmetic mean is calculated. The arithmetic mean is used as the windowed rate value at the center time of the sliding window to obtain a windowed temperature rate sequence. For each sampling point in the windowed temperature rate sequence, it is compared with a preset rate threshold. If the values ​​of all sampling points in a certain continuous interval of the windowed temperature rate sequence are lower than the preset rate threshold, then the continuous interval is marked as a candidate stable period. For all candidate stable periods, their duration is compared with a preset duration threshold. If the duration is not less than the preset duration threshold, then the candidate stable period is retained. For the retained candidate stable periods, their starting timestamp is taken as the inflection point position of the return water temperature.

[0038] The turning point when the return water temperature of the heating network changes from rapid change to slow stabilization is called the return water temperature inflection point. It corresponds to the critical moment when the heating network switches from the overall cooling stage to the energy release stage of the thermal storage device, and serves as the time benchmark for dividing the boundary of the subsequent thermal storage and release stage.

[0039] The boundary of the return water temperature segment is defined by the start and end timestamps of the unit entering the low-load peak-shaving and heating coupling operation. The start timestamp is taken from the first sampling moment after the unit load command switches from the rated section to the peak-shaving section, and the end timestamp is taken from the last sampling moment before the unit load recovers to the rated section. From the monitoring data aligned with the grid dispatch timing, all sampling points of the heating network return water temperature channel are extracted according to the continuous interval between the above two timestamps to form the return water temperature segment.

[0040] Although the original return water temperature channel has undergone preliminary cleaning and linear interpolation by the monitoring device, slight fluctuations in local measuring points due to pipe wall adhesion and pump / valve switching disturbances still exist along long-distance heating network pipelines. The return water temperature segment is smoothed using a sliding median filter with an odd-numbered window length. The sliding median filter window length is selected to cover 5 to 9 consecutive sampling points. After processing, several sampling points at the beginning and end of the return water temperature segment due to incomplete windows are removed. The remaining sampling points are then rearranged in ascending order of timestamps to form the smoothed return water temperature segment. The temperature rate is calculated for adjacent sampling point pairs within the smoothed return water temperature segment.

[0041] Single-point values ​​in a temperature rate sequence are susceptible to residual noise interference, necessitating windowing to extract trend changes. A sliding window of preset width is used to window the temperature rate sequence, covering 15 to 30 consecutive rate sampling points within a preset sampling period. After traversing the temperature rate sequence, a windowed temperature rate sequence slightly shorter than the original sequence is obtained. For each sampling point in the windowed temperature rate sequence, a preset rate threshold is compared. This preset rate threshold is determined based on the 10th percentile of the rate distribution during the normal cooling phase in the historical operating data of the heating network in the unit's area, and is set to 0.05℃ / min, reflecting the critical level at which the return water temperature transitions from rapid to slow change. If a continuous interval exists in the windowed temperature rate sequence where all sampling points are below the preset rate threshold, this continuous interval is marked as a candidate stable period.

[0042] Candidate stable periods may be falsely triggered by occasional local rate drops, necessitating the introduction of a duration criterion for filtering. For all candidate stable periods, the duration between their first and last timestamps is calculated. This duration is then compared to a preset duration threshold, which is determined to be 2 to 3 times the heat transfer time constant of the heat exchanger at the first station of the heating network. In this embodiment, the heat transfer time constant is calibrated to 10 minutes based on unit thermal performance test data; therefore, the preset duration threshold is 25 minutes to exclude short-term disturbances. If the duration is not less than the preset duration threshold, the candidate stable period is retained; otherwise, it is discarded. Finally, for the retained candidate stable periods, their starting timestamp is taken as the inflection point of the return water temperature. The inflection point of the return water temperature corresponds to the turning point on the time axis when the heating network switches from the overall cooling stage to the slow stabilization stage. The time misalignment between this point and the time when the unit boiler coal consumption decreases and the time when the heating steam extraction decreases is the lagging characteristic of the thermal energy storage device releasing energy through the heating network buffer, providing a time reference for the boundary division of the subsequent thermal energy release stage.

[0043] S104. Carbon accounting sampling is triggered by the inflection point of return water temperature. The main and reheat steam pressure and temperature, heating extraction steam flow enthalpy and flue gas carbon flow within the time window before and after the inflection point are extracted from the monitoring data aligned with the grid dispatch time sequence to obtain the triggered sampling dataset.

[0044] Using the inflection point of the return water temperature as a reference, the left boundary of the window is obtained by extending it before the inflection point by a preset pre-time interval, and the right boundary of the window is obtained by extending it after the inflection point by a preset post-time interval. The carbon accounting sampling time window is defined by the left and right boundaries of the window. All record rows whose timestamps fall within the carbon accounting sampling time window interval are retrieved from the monitoring data aligned with the grid dispatch timing. For each record row falling within the interval, the main and reheat steam pressure and temperature, the heating extraction steam flow enthalpy value, and the flue gas carbon flow fields are extracted and sorted in ascending order of timestamp to obtain the triggered sampling dataset.

[0045] Under the coupled operation of low-load peak shaving and heating of coal-fired units, the inflection point of the return water temperature corresponds to the transition moment on the time axis when the heating network switches from the overall cooling stage to the energy release stage of the thermal storage device. The operating parameters within a certain period before and after this transition moment cover the complete transition process of unit load reduction, steam extraction reduction, and boiler coal combustion changes, and are the core data source for calculating the instantaneous carbon emission intensity during the thermal storage release stage.

[0046] The left and right boundaries of the carbon accounting sampling time window are determined by adding two offsets before and after the inflection point of the return water temperature. The left boundary of the window is obtained by subtracting a preset pre-set time from the timestamp of the inflection point of the return water temperature, and the right boundary of the window is obtained by adding a preset post-set time to the timestamp of the inflection point of the return water temperature. The carbon accounting sampling time window is defined by the left and right boundaries of the window. The preset pre-set time is selected based on the typical duration of the cooling phase of the heating network, and is determined by three times the heat transfer time constant of the heat exchanger at the first station of the heating network in the unit's area, taking 30 minutes. The preset post-set time is selected based on the typical duration of the heat storage and release phase, and is determined by five times the heat transfer time constant, taking 50 minutes.

[0047] For each row of monitoring data aligned with the grid dispatch time sequence, its timestamp t is taken and compared with the left boundary tL and the right boundary tR of the window: if tL≤t≤tR, the row falls within the carbon accounting sampling time window and is included in the extraction set; otherwise, the row is skipped. For each row in the extraction set, the values ​​of the three fields—main and reheat steam pressure and temperature, heating extraction steam flow enthalpy, and flue gas carbon flow—are read according to a pre-agreed field order and written into the corresponding row of the trigger sampling dataset along with the timestamp. The trigger sampling dataset is presented as a two-dimensional table structure with the timestamp as the primary key. Each row corresponds to a sampling moment within the carbon accounting sampling time window. The column fields cover the operating parameters of three sections: the turbine inlet side, the heating extraction steam side, and the boiler combustion side, arranged in ascending order of timestamp, providing operating parameter slices aligned with inflection points for subsequent calculation of instantaneous carbon emission intensity.

[0048] S105. The peak power supply load of the power supply unit is evaluated by the main and reheat steam pressure and temperature in the triggered sampling dataset. The instantaneous carbon emission intensity is calculated by dividing the flue gas carbon flow by the peak power supply load, and an instantaneous carbon emission intensity record is formed.

[0049] The main steam pressure and temperature are extracted row by row from the triggered sampling dataset in time-stamp order. For each time-stamp, the pre-established correspondence curve between the steam inlet parameters of the power supply unit and the power supply load is retrieved. Bilinear interpolation is applied to the main steam pressure and temperature to obtain the power supply load component of the main steam portion at that time-stamp. For the reheat steam pressure and temperature at the same time-stamp in the triggered sampling dataset, the same bilinear interpolation process is applied according to the correspondence curve to obtain the power supply load component of the reheat steam portion at that time-stamp. For each time-stamp, the power supply load component of the main steam portion and the power supply load component of the reheat steam portion are summed to obtain the instantaneous peak-shaving power supply load of the power supply unit at that time-stamp. All time-stamps in the triggered sampling dataset are traversed, and the instantaneous peak-shaving power supply loads at each time-stamp are arranged in ascending order of time-stamp to obtain the instantaneous peak-shaving power supply load sequence corresponding one-to-one with the time-stamps of the triggered sampling dataset. For each row of flue gas carbon flow rate in the triggered sampling dataset in timestamp order, it is paired one-to-one with the instantaneous peak-shaving power supply load at the same timestamp in the instantaneous peak-shaving power supply load sequence; for each timestamp, the flue gas carbon flow rate is divided by the instantaneous peak-shaving power supply load to obtain the instantaneous carbon emission intensity at that timestamp; the instantaneous carbon emission intensities at each timestamp are arranged in ascending order of timestamp to obtain the instantaneous carbon emission intensity record.

[0050] Under the coupled operation of low-load peak shaving and heating of coal-fired units, the triggered sampling dataset contains complete operating parameters within a time window before and after the inflection point of the heating network return water temperature. To measure the carbon dioxide emissions corresponding to a unit of power supply within this window, the instantaneous peak shaving power supply load of the power supply unit must first be reconstructed from the state parameters of the main and reheat steam, and then the instantaneous carbon emission intensity is obtained by dividing the flue gas carbon flow rate at the same timestamp by this load.

[0051] The correlation curve between the steam inlet parameters of the power supply unit and the power supply load is established in advance before the system is put into operation. Based on the unit's thermodynamic performance test data and historical normal operating condition data, the power generation output is converted according to the net power supply output (i.e., after deducting plant power consumption). The power supply load values ​​are gridded and calibrated using the main steam pressure and temperature as coordinate axes to form a two-dimensional lookup table for the main steam side. A corresponding two-dimensional lookup table is established for the reheat steam side in the same way. The two lookup tables together constitute the correlation curve. The grid step size for the main steam pressure is calibrated based on the unit's thermodynamic performance test data, divided at 2% of the unit's rated pressure, and 0.5 MPa is used for units with a rated main steam pressure of 25 MPa. The grid step size for the main steam temperature is divided according to the deviation range of the unit's rated temperature, and is taken as 5℃.

[0052] The power load component on the main steam side reflects the work done by the high-pressure cylinder, while the power load component on the reheat steam side reflects the work done by the intermediate and low-pressure cylinders. Decoupling these two components and processing them separately avoids errors caused by estimations based on a single parameter. For each time stamp in the triggered sampling dataset, the main steam pressure P and temperature T at that time stamp are extracted. The grid sub-rectangle into which the pressure and temperature fall is located in the two-dimensional lookup table on the main steam side. The power load values ​​at the intersections of the four grids of the sub-rectangle are then subjected to bilinear interpolation to obtain the power load component of the main steam portion at that time stamp. For the reheat steam pressure and temperature at the same time stamp in the triggered sampling dataset, the same bilinear interpolation process is used based on the two-dimensional lookup table on the reheat steam side in the corresponding relationship curve to obtain the power load component of the reheat steam portion at that time stamp.

[0053] The power supply load components of the main steam section and the reheat steam section at each timestamp are summed to obtain the instantaneous peak-shaving power supply load of the power unit at that timestamp. All timestamps in the triggered sampling dataset are traversed, and the instantaneous peak-shaving power supply loads at each timestamp are arranged in ascending order to form a sequence of instantaneous peak-shaving power supply loads that corresponds one-to-one with the timestamps of the triggered sampling dataset. Each row of flue gas carbon flow rate in the triggered sampling dataset is paired with the instantaneous peak-shaving power supply load at the same timestamp in the instantaneous peak-shaving power supply load sequence. The instantaneous carbon emission intensity at that timestamp is obtained by dividing the flue gas carbon flow rate by the instantaneous peak-shaving power supply load, which physically represents the mass of carbon dioxide corresponding to a unit of power supply.

[0054] The instantaneous carbon emission intensity record is presented as a one-dimensional sequence with timestamp as the primary key. Each item corresponds to a sampling moment within the carbon accounting sampling time window. The value reflects the instantaneous conversion relationship between the power supply unit's emissions on the coal-fired side and the power supply output at that moment. This provides a time-by-time comparison basis for identifying the plateau or swing pattern of the instantaneous carbon emission intensity within the range of the decrease in the enthalpy value of the heating steam extraction flow rate.

[0055] S106. By comparing the instantaneous carbon emission intensity record with the enthalpy value of the heating extraction steam flow, identify the range of decreasing enthalpy value of the heating extraction steam flow, analyze the plateau or swing pattern of the intensity record within the range of decreasing enthalpy value of the heating extraction steam flow, and determine the energy storage and heat storage release stage marker.

[0056] For the instantaneous carbon emission intensity record and the heating extraction steam flow rate enthalpy value in the triggered sampling dataset, they are aligned one-to-one by timestamp; for each timestamp, the heating extraction steam flow rate and the enthalpy value are multiplied to obtain the heating extraction steam flow rate enthalpy value product at that timestamp; all timestamps in the triggered sampling dataset are traversed, and the heating extraction steam flow rate enthalpy value products at each timestamp are sorted in ascending order by timestamp to obtain the heating extraction steam flow rate enthalpy value product sequence corresponding one-to-one with the timestamp of the instantaneous carbon emission intensity record. For the product sequence of the enthalpy values ​​of the heating extraction steam flow rate, the product difference between two adjacent sampling points is divided by the interval between the two timestamps to obtain the product change rate of the adjacent sampling point pair; the product change rates of all adjacent sampling point pairs are arranged in ascending order of timestamps to obtain the product change rate sequence; for a continuous interval in the product change rate sequence, if all the product change rates in the continuous interval are negative and their absolute values ​​are not lower than a preset decrease rate threshold, and the duration of the continuous interval is not less than a preset duration threshold, then the continuous interval is marked as a heating extraction steam flow rate enthalpy reduction interval. For the subsequence corresponding to the instantaneous carbon emission intensity recorded within the range of the enthalpy reduction of the heating extraction steam flow rate, the range of all values ​​within the subsequence is calculated as the fluctuation amplitude. If the fluctuation amplitude does not exceed a preset fluctuation amplitude threshold, the subsequence is determined to be a plateau pattern. If the subsequence exhibits a reverse refractory feature of first rising and then falling or first falling and then rising, the difference between the extreme value before the turning point and the return point after the turning point in the reverse refractory feature is used as the refractory amplitude, and if the refractory amplitude exceeds a preset refractory amplitude threshold, the subsequence is determined to be a swing pattern. For the range of the enthalpy reduction of the heating extraction steam flow rate exhibiting a plateau pattern or a swing pattern, an energy storage and heat storage release stage identifier is determined.

[0057] Under the coupled operation of low-load peak shaving and heating of coal-fired units, the instantaneous carbon emission intensity record deviates from the normal relationship between boiler coal consumption and power supply load during the heat storage release phase, exhibiting a trend characteristic that is completely different from that of the load reduction phase alone. The reason for this deviation is that the heat storage device in the electrical energy storage system releases stored heat energy into the heating network under peak power supply conditions. After being buffered by the heat storage capacity of the heating network, the return water temperature lags behind the unit load decrease, causing the heating steam extraction to be passively reduced for a period of time, while the boiler coal consumption is not reduced synchronously. As a result, the instantaneous carbon emission intensity record shows a plateau or swing pattern.

[0058] The product of the heating extraction steam flow rate and enthalpy corresponds, in physical terms, to the heat power carried out of the turbine by the extraction branch per unit time. Its decrease signifies a reduction in the heat taken from the turbine extraction side of the heating network. The heating extraction steam flow rate and enthalpy in the instantaneous carbon emission intensity record and the triggered sampling dataset are aligned one-to-one by timestamp. The product of the heating extraction steam flow rate and enthalpy at each timestamp is taken as the heating extraction steam flow rate and enthalpy product at that timestamp. All timestamps within the carbon accounting sampling time window are traversed, and the products W at each timestamp are arranged in ascending order, forming a heating extraction steam flow rate and enthalpy product sequence corresponding one-to-one with the timestamps of the instantaneous carbon emission intensity record. The ratio of the product difference to the timestamp interval is calculated for two adjacent sampling point pairs in the heating extraction steam flow rate and enthalpy product sequence to obtain the rate of change of the product for that adjacent sampling point pair. A negative rate of change indicates a decrease in the product, and the absolute value reflects the severity of the decrease.

[0059] The determination of the interval for the decrease in enthalpy of the heating extraction steam flow rate adopts a dual-threshold filtering method. All continuous intervals in the product change rate sequence are traversed, and each continuous interval is checked point by point: if all product change rates within the interval are negative, and the absolute value of the rate at each point is not lower than a preset decrease rate threshold, then the continuous interval passes the decrease magnitude criterion; the duration between the start and end timestamps of the continuous interval is then calculated and compared with a preset duration threshold. If the duration is not less than the preset duration threshold, then the continuous interval passes the duration criterion. Continuous intervals that pass both criteria are marked as intervals for the decrease in enthalpy of the heating extraction steam flow rate. The preset decrease rate threshold is selected based on the 10th percentile of the rate of decrease in extraction steam power during normal load reduction, and is statistically calibrated to 50 kW / s based on historical operating data of the unit; the preset duration threshold is selected as twice the heat transfer time constant of the heat exchanger at the first station of the heating network, taking 20 minutes. The two are combined to eliminate occasional disturbances.

[0060] Within the decreasing range of the enthalpy value of the heating steam extraction flow rate, the values ​​corresponding to the timestamps of the instantaneous carbon emission intensity records are extracted and arranged in ascending order of timestamps to form an intensity subsequence. The plateau shape determination of the intensity subsequence is implemented as follows: the maximum value e of all values ​​within the intensity subsequence is taken. max With minimum value e min The difference ΔE between the two is calculated as the fluctuation amplitude, and compared with a preset fluctuation amplitude threshold. If ΔE does not exceed the preset fluctuation amplitude threshold, the intensity subsequence is determined to be a plateau pattern. Physically, the plateau pattern corresponds to a stable relative ratio between boiler coal consumption and instantaneous power supply load, reflecting good coordination between unit flue gas carbon flow and power output during the heat storage release period. The preset fluctuation amplitude threshold is statistically calibrated to 20 g / (kW·h) based on short-term fluctuations in instantaneous carbon emission intensity during peak-shaving periods in the unit's historical operating data.

[0061] The determination of the swing pattern of the intensity subsequence is implemented as follows: For the intensity subsequence, the first local extreme point and the immediately following reverse extreme point are located. If the intensity subsequence exhibits a reverse refractory characteristic of first rising and then falling or first falling and then rising, the absolute value of the difference between the extreme value before the turning point and the return point after the turning point is taken as the refractory amplitude. The preset refractory amplitude threshold is selected as a multiple of the typical short-term fluctuation range of instantaneous carbon emission intensity of the unit in the peak-shaving section, and is statistically calibrated to 40g / (kW·h) based on the historical operating data of the unit. If the refractory amplitude exceeds the preset refractory amplitude threshold, the intensity subsequence is determined to be a swing pattern. The swing pattern physically corresponds to the short-term maintenance of boiler coal consumption while the extraction steam power decreases rapidly. The mismatch between the two rhythms leads to a reversal in instantaneous carbon emission intensity. Finally, for the heating extraction steam flow enthalpy reduction range exhibiting a platform pattern or a swing pattern, its start and end timestamps and pattern category are written into the energy storage and thermal release stage identifier. The energy storage and thermal energy release stage identifier defines the active period on the timeline during which the thermal energy storage device releases stored energy to the heating load through the heating network, and its morphology category field depicts the changes in instantaneous carbon emission intensity during this active period.

[0062] S107. Based on the energy storage and heat storage release stage identifier, the entire process of power supply and heating during low load peak regulation of the power grid is divided into energy storage release sub-stage and non-release sub-stage. The average instantaneous carbon emission intensity of each sub-stage is extracted. The instantaneous carbon emission intensity of the energy storage release sub-stage and the non-release sub-stage is compared for multiple sets of monitoring data aligned with the power grid dispatching time sequence. The comparative evaluation conclusion of power supply carbon emission of each flexibility transformation group is obtained.

[0063] Based on the start and end timestamps in the energy storage and thermal energy release stage identifiers, the entire process of low-load peak power supply and heating in the power grid is divided on the time axis. The time intervals covered by the energy storage and thermal energy release stage identifiers are marked as energy storage release sub-stages; the remaining time intervals of the entire process not covered by the energy storage and thermal energy release stage identifiers are marked as non-release sub-stages. The energy storage release sub-stages and non-release sub-stages are arranged in ascending order of timestamps to obtain the sub-stage segmentation results. For each energy storage release sub-stage in the sub-stage segmentation results, all intensity values ​​within the time interval of that sub-stage are retrieved from the instantaneous carbon emission intensity record, and their arithmetic mean is calculated to obtain the average instantaneous carbon emission intensity of that energy storage release sub-stage. For each non-release sub-stage in the sub-stage segmentation results, all intensity values ​​within the time interval of that sub-stage are retrieved from the instantaneous carbon emission intensity record, and their arithmetic mean is calculated to obtain the average instantaneous carbon emission intensity of that non-release sub-stage. The average instantaneous carbon emission intensities of all sub-stages are collected to obtain the sub-stage intensity mean set. For multiple sets of monitoring data aligned with grid dispatch timing from different flexibility transformation groups, the average instantaneous carbon emission intensity of the energy storage release sub-stage and the average instantaneous carbon emission intensity of the non-release sub-stage are extracted from the average intensity set of the sub-stages corresponding to each group. For each flexibility transformation group, the difference and ratio between the average instantaneous carbon emission intensity of the energy storage release sub-stage and the average instantaneous carbon emission intensity of the non-release sub-stage are calculated. The difference and ratio of each flexibility transformation group are summarized to obtain the comparative evaluation conclusion of power supply carbon emissions of each flexibility transformation group.

[0064] In the scenario of evaluating the effectiveness of flexibility retrofits for coal-fired power generation units, simply comparing the total carbon emissions of the units before and after the retrofit during low-load peak-shaving power supply from the grid is insufficient to reveal the specific impact of the thermal energy storage release phase on the intensity of carbon emissions from power supply. The thermal energy storage release phase marker defines the active period on the timeline during which the thermal energy storage device releases stored energy to the heating load. Separating this period from the remaining time periods without energy storage release allows for the separate characterization of carbon emissions on the power supply side under both types of periods, thereby supporting targeted evaluation of the retrofit plan.

[0065] The sub-stage segmentation results are constructed as follows: The energy storage and thermal energy release stage identifier may contain several intervals. The start timestamp of each interval is denoted as ts, and the end timestamp as te. Each interval [ts, te] is marked as an energy storage release sub-stage on the time axis of the entire process of power grid low-load peak-shaving power supply and heating. The total time axis of the entire process, after deducting the time intervals occupied by all energy storage release sub-stages, contains several discrete time periods, which are then marked as non-release sub-stages. The energy storage release sub-stages and the non-release sub-stages alternate, and are arranged in ascending order of timestamps to form the sub-stage segmentation results.

[0066] The calculation of the average intensity of the thermal energy storage and release sub-stages depends on the values ​​taken from the corresponding time periods in the instantaneous carbon emission intensity records. For each energy storage and release sub-stage, the start and end positions are located from the instantaneous carbon emission intensity records by timestamp. All instantaneous carbon emission intensity values ​​within that position interval are extracted, and the number of values ​​is recorded as N. The sum of all values ​​is divided by N to obtain the average instantaneous carbon emission intensity of that energy storage and release sub-stage. The same value taking and arithmetic averaging operation is performed on each non-release sub-stage to obtain the average instantaneous carbon emission intensity of that non-release sub-stage. The average instantaneous carbon emission intensities of all energy storage and release sub-stages and all non-release sub-stages throughout the entire process are collected, categorized by sub-stage type, and constitute a set of sub-stage intensity averages.

[0067] The method of setting up flexibility modification groups directly affects the interpretability of the comparative conclusions. Multiple sets of monitoring data aligned with the grid dispatching time sequence can be derived from multiple low-load peak-shaving power supply and heating test data of the same unit under different flexibility modification stages, or from the concurrent operation data of multiple units with different modification schemes under the same regional heating network. The different flexibility modification schemes include at least one of the following combinations: (a) the original unit without thermal energy storage devices; (b) the scheme of connecting a molten salt thermal storage tank in parallel on the heating extraction steam branch; (c) the scheme of connecting an electrode boiler and a water thermal storage tank on the boiler feedwater side; (d) the scheme of arranging a solid electric thermal storage body before the first station of the heating network; each modification scheme corresponds to a flexibility modification group. Each modification group collects a round of monitoring data covering the complete peak-shaving power supply process under similar ambient temperatures, initial values ​​of heating network return water, and load commands. After processing such as time synchronization, alignment with grid dispatching time sequence, return water temperature inflection point identification, trigger sampling, instantaneous carbon emission intensity calculation, and thermal energy storage release stage identification, the average intensity set of each sub-stage is obtained. For each flexibility modification group, the difference Δe is calculated between the average instantaneous carbon emission intensity eh of the energy storage release sub-stage and the average instantaneous carbon emission intensity en of the non-release sub-stage, where Δe = eh The ratio r = eh / en, where Δe reflects the absolute deviation of the energy storage release period relative to the non-release period, and r reflects the relative deviation ratio.

[0068] The Δe and r values ​​of each flexibility modification group are summarized into a comparison result. This comparison result is recorded group by group, with each group's record including the average instantaneous carbon emission intensity during the energy storage release phase, the average instantaneous carbon emission intensity during the non-release phase, the difference Δe, and the ratio r. The difference and ratio of each flexibility modification group in the comparison result are used to horizontally evaluate the deviation of different modification schemes from the power supply carbon emission intensity during the energy storage release period. Schemes with smaller absolute values ​​of Δe and r closer to 1 indicate a smaller impact on the power supply carbon emission intensity during the thermal energy storage release period, and thus better power supply carbon emission performance. Finally, the comparison result is output as the power supply carbon emission comparison assessment conclusion for each flexibility modification group. This conclusion presents the differences in instantaneous carbon emission intensity of multiple modification schemes during the energy storage release and non-release periods in a horizontally comparable manner, providing an objective basis based on operational data for further selection and optimization of energy storage-type flexibility modification routes for coal-fired power units, as well as for grid load dispatching and optimized configuration of energy storage systems.

[0069] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A comparative assessment method for carbon emissions based on energy storage in coal-fired power generation units, characterized in that, The method is applied to a coal-fired power generation system equipped with thermal energy storage devices, including: By connecting the power grid dispatch and control system of the power supply unit to the coal-fired power generation and power supply system, the original records of hot steam pressure and temperature, heating extraction steam flow and enthalpy, flue gas carbon flow and heating network return water temperature are collected to obtain a real-time monitoring dataset that is synchronized with time. Based on the real-time monitoring dataset, the hot steam pressure and temperature and flue gas carbon flow rate within the same time window are aligned, and the enthalpy value of the heating extraction steam flow rate is processed to form monitoring data aligned with the grid dispatch timing. By analyzing monitoring data aligned with the grid dispatch timing, the peak power supply load of the power supply unit is assessed. The peak power supply load reflects the real-time power distribution output of the power supply unit to the power system. The instantaneous carbon emission intensity is obtained by dividing the flue gas carbon flow by the peak power supply load, thus forming an instantaneous carbon emission intensity record. By comparing the instantaneous carbon emission intensity record with the enthalpy value of the heating extraction steam flow rate, the range of decrease in the enthalpy value of the heating extraction steam flow rate is identified. The range of decrease corresponds to the process in which the thermal energy storage device in the power storage system releases heat energy to compensate for the power output. The plateau or swing pattern of the intensity record within the range of decrease is analyzed to determine the stage identifier of the energy storage and thermal energy release. The energy storage and thermal storage release stages are divided into energy storage release sub-stages and non-release sub-stages based on the energy storage and thermal storage release stage identifiers. The instantaneous carbon emission intensity of each flexibility transformation group under the grid peak-shaving power supply condition is compared, and the difference in the impact of each transformation scheme on the carbon emission intensity of the power supply side of the power system is output. The comparative evaluation conclusion of the power supply carbon emission of each transformation scheme is obtained.

2. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 1, characterized in that, The process involves connecting the power grid dispatch control system of the power supply unit to the coal-fired power generation power supply system, collecting raw records of hot steam pressure and temperature, heating extraction steam flow and enthalpy, flue gas carbon flow, and heating network return water temperature to obtain a time-synchronized real-time monitoring dataset, including: A continuous flue gas monitoring device is installed in the flue gas emission channel. Pressure transmitters and temperature transmitters are respectively installed in the main steam pipeline and the reheat steam pipeline. A vortex flow meter and temperature and pressure compensation measuring points are installed in the heating extraction steam branch line. A platinum resistance thermometer is arranged in the return water header of the heating network. A unified clock reference is sent to all monitoring devices using the Network Time Protocol to align the local clocks of each device with the main clock of the power grid dispatch and control system of the power supply units. Data from each channel is sent up at the same sampling interval. The cleaned records from each channel are merged into the same storage table with a unified timestamp to obtain the real-time monitoring dataset that is synchronized with the time.

3. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage as described in claim 1, characterized in that, The process of aligning the hot steam pressure and temperature with the flue gas carbon flow rate within the same time window based on the real-time monitoring dataset, processing the enthalpy value of the heating extraction steam flow rate, and forming monitoring data aligned with the grid dispatch timing includes: Using the master clock of the power grid dispatch and control system of the power supply unit as the alignment reference, a continuous and non-overlapping time window is defined according to a preset duration. The sampling points of hot steam pressure and temperature and flue gas carbon flow rate that fall within the same time window are grouped into the same row according to the timestamp to obtain the initial aligned record; For each target timestamp in the initial alignment record, the sampling point with the smallest difference from the target timestamp in the heating steam extraction flow enthalpy channel is found. The heating steam extraction flow field and enthalpy field are written into the heating steam extraction flow enthalpy channel with a difference not exceeding the preset tolerance threshold. For the difference exceeding the threshold, the empty fields are filled with linear interpolation to obtain the monitoring data aligned with the power grid scheduling time sequence.

4. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage as described in claim 1, characterized in that, The method further includes: Extract the return water temperature segment of the heating network from the monitoring data aligned with the power grid dispatch time sequence, and identify the inflection point of the return water temperature when the temperature change rate is continuously lower than the preset threshold and exceeds the preset period by calculating the temperature change rate at continuous time points.

5. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 4, characterized in that, The step of extracting the return water temperature segment from monitoring data aligned with the power grid dispatch time sequence, and identifying the inflection point of return water temperature where the temperature change rate is continuously lower than a preset threshold and exceeds a preset period by calculating the temperature change rate at continuous time points, includes: Based on the start and end timestamps of the power supply units entering the grid under low-load peak-shaving power supply and heating coupling conditions, the original records of the heat network return water temperature covering the complete peak-shaving power supply process are extracted to obtain the return water temperature sub-segment. The return water temperature segment is smoothed by sliding median filtering with an odd window length to obtain a smoothed return water temperature segment. The temperature rate sequence is obtained by dividing the temperature difference between adjacent sampling points in the smoothed return water temperature segment by the timestamp interval. A windowed temperature rate sequence is obtained by taking the absolute value of the temperature rate sequence and calculating its arithmetic mean using a sliding window of preset width. In the windowed temperature rate sequence, all intervals with values ​​lower than a preset rate threshold and a duration not less than a preset duration threshold are retained, and their starting timestamps are taken as the inflection point of the return water temperature. The inflection point of the return water temperature marks the start time of energy release of the thermal energy storage device under the peak power supply condition of the power grid.

6. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 1, characterized in that, The method further includes: Carbon sampling is triggered by the inflection point of the return water temperature. The hot steam pressure and temperature, heating extraction steam flow rate and enthalpy, and flue gas carbon flow rate within the time window before and after the inflection point are extracted to obtain the triggered sampling dataset.

7. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 6, characterized in that, Carbon accounting sampling is triggered by the inflection point of the return water temperature. The hot steam pressure and temperature, heating extraction steam flow rate and enthalpy, and flue gas carbon flow rate within the time window before and after the inflection point are extracted to obtain the triggered sampling dataset, which specifically includes: Based on the inflection point of the return water temperature, the left boundary of the window is obtained by extending it before the inflection point by a preset pre-time, and the right boundary of the window is obtained by extending it after the inflection point by a preset post-time. The carbon accounting sampling time window is defined by the left boundary and the right boundary of the window. The trigger sampling dataset is obtained by retrieving all record rows whose timestamps fall within the sampling time window from the monitoring data aligned with the power grid dispatch timing and sorting them in ascending order of timestamps.

8. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 6, characterized in that, The method involves analyzing monitoring data aligned with grid dispatch timing to assess the peak-shaving load of power supply units. This peak-shaving load reflects the real-time power output of the power supply unit to the power system. Instantaneous carbon emission intensity is obtained by dividing flue gas carbon flow by the peak-shaving load, forming an instantaneous carbon emission intensity record, including: The main steam pressure and temperature are extracted from the real-time monitoring dataset row by row according to the timestamp. The pre-established correspondence curve between the steam inlet parameters of the power supply unit and the power supply load is retrieved. The main steam pressure and temperature are interpolated bilinearly to obtain the main steam power supply load component. The reheat steam power supply load component is obtained by using the same interpolation process for reheat steam pressure and temperature at the same timestamp. The instantaneous peak-shaving power supply load is obtained by summing the two components; The instantaneous carbon emission intensity record is obtained by dividing the flue gas carbon flow rate by the instantaneous peak power supply load and arranging them in ascending order of timestamps.

9. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 1, characterized in that, The method involves comparing instantaneous carbon emission intensity records with the enthalpy of heating steam extraction flow rate to identify the reduction range of heating steam extraction flow rate enthalpy. This reduction range corresponds to the process in which thermal energy storage devices in the energy storage system release heat energy to compensate for power output. The method analyzes the plateau or oscillation pattern exhibited by the intensity records within this reduction range to determine the energy storage and thermal energy release stage identifier, including: The instantaneous carbon emission intensity record is aligned with the enthalpy value of the heating extraction steam flow rate in the triggered sampling dataset by timestamp; For each timestamp, the product of the steam extraction flow rate and enthalpy is taken, and the sequence of the product of the steam extraction flow rate and enthalpy is obtained by arranging them in ascending order of timestamp. Dividing the product difference between adjacent sampling points by the timestamp interval yields the product change rate sequence. In the product change rate sequence, if the rate of continuous intervals are all negative and the absolute value is not lower than the preset decrease rate threshold, and the duration is not less than the preset duration threshold, it is marked as the heating steam extraction flow enthalpy value reduction interval. This reduction interval corresponds to the process in which the thermal energy storage device in the electric energy storage system releases stored thermal energy to the heating network to maintain the peak power output of the power supply unit. The instantaneous carbon emission intensity record is determined to be a plateau pattern if the range of the subsequences within the reduction range does not exceed a preset fluctuation amplitude threshold. If a reverse reversal feature appears in the subsequence, the reversal amplitude is determined by the reverse reversal feature. If the reversal amplitude exceeds a preset reversal amplitude threshold, it is determined to be a swing pattern. For the reduction range that presents a platform shape or a swing shape, the energy storage and heat release stage identifier is determined.

10. The carbon emission comparison and assessment method based on coal-fired power generation unit energy storage according to claim 1, characterized in that, The process involves dividing the energy storage and thermal storage release phases into energy release sub-phases and non-release sub-phases based on energy storage and thermal storage release stage identifiers. It then compares the instantaneous carbon emission intensity of each flexibility upgrade group under grid peak-shaving power supply conditions, outputs the difference in impact of each upgrade scheme on the power system's supply side carbon emission intensity, and obtains comparative evaluation conclusions on the power supply carbon emissions of each upgrade scheme, including: The entire process of power grid low-load peak power supply and heating is divided on the time axis according to the start and end timestamps in the energy storage and thermal release stage identifier. The time interval covered by the energy storage and thermal release stage identifier is marked as the energy storage release sub-stage, and the remaining time intervals not covered are marked as non-release sub-stages. For each sub-stage, the intensity values ​​are retrieved from the instantaneous carbon emission intensity records and the arithmetic mean is calculated to obtain the average instantaneous carbon emission intensity of the sub-stage. The difference and ratio of the average instantaneous carbon emission intensity between the energy storage release sub-stage and the non-release sub-stage of each flexibility modification group are calculated, and the power supply carbon emission comparison evaluation conclusion of each modification scheme is obtained by summarizing them.

Images