Flexible and extensible agricultural and pastoral district multi-element energy consumption and carbon emission data acquisition method

Abstract

The invention discloses a flexible and extensible agricultural and pastoral district multivariate energy consumption and carbon emission data acquisition method, relates to the technical field of data acquisition, is used for solving the problem that the utilization efficiency of edge side computing resources is reduced, and aims at solving the problem that the utilization efficiency of edge side computing resources is reduced by taking edge nodes as core computing carriers and introducing an operation monitoring mechanism based on a scheduling period. The method comprises the following steps: performing quantitative evaluation on carbon emission calculation duration and result output timeliness, constructing a calculation pressure index, judging an edge node operation state through a regulation and control identification mechanism, calling park net energy flow data, analyzing a park regulation and control fluctuation state, and selecting different calculation strategies and parameter processing paths. According to the method, multi-element energy consumption parameters are sorted and classified, precision compensation or link shortening processing is carried out in combination with the calculation power margin, self-adaptive adjustment of the calculation load along with the operation state is achieved, and the utilization efficiency of edge side calculation resources is improved while the timeliness of carbon emission estimation is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, and more specifically, to a flexible and scalable method for acquiring data on diversified energy use and carbon emissions in agricultural and pastoral parks. Background Technology

[0002] With the accelerated construction of new power systems, agricultural and pastoral parks, as comprehensive energy-consuming scenarios integrating agricultural production, livestock breeding, cold chain processing, and distributed energy utilization, are showing significant characteristics of diversified, distributed, and fluctuating energy structures. Typical agricultural and pastoral parks usually connect to multiple energy forms such as electricity, natural gas, biomass energy, photovoltaic power generation, energy storage devices, and agricultural machinery fuel, and their energy flow has strong coupling and significant time-varying characteristics.

[0003] The existing technology has the following shortcomings: Currently, existing technologies mostly employ centralized cloud platforms or edge nodes based on fixed strategies for collecting diverse energy consumption data and estimating carbon emissions. These technologies generally suffer from insufficient adaptability to the computing power and scheduling cycles of edge nodes, and lack a quantitative assessment and comprehensive characterization mechanism for the computational load, carbon emission calculation time, and timeliness of result output within the scheduling cycle. This leads to reduced efficiency in utilizing edge computing resources, and when the park experiences frequent load fluctuations or participates in grid regulation, carbon emission estimation delays or computational task blockages are likely to occur. Therefore, this paper proposes a flexible and scalable method for collecting diverse energy consumption and carbon emission data in agricultural and pastoral parks.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks. This method utilizes an edge node operation monitoring and pressure index construction mechanism based on scheduling cycles, combined with a control fluctuation identification strategy driven by the park's net energy flow, and introduces a precision compensation and link adaptive adjustment mechanism for sorting and classifying multi-energy consumption parameters and sensing computing power margin to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a flexible and scalable method for collecting data on diversified energy use and carbon emissions in agricultural and pastoral parks, comprising the following steps: Step S1: When performing carbon emission estimation in the agricultural and pastoral park, access the default setting table to obtain the scheduling cycle information of the edge node to be tested and detect the window compression rate and carbon emission calculation time within the scheduling cycle. Evaluate the parameter estimation characteristics of the edge node to be tested based on the carbon emission calculation time. Step S2: Generate the computational pressure index for the current scheduling cycle by combining the comprehensive parameter estimation characteristics and window compression rate. Use the computational pressure index to determine whether to enter the control identification mechanism. If the control identification mechanism is entered, retrieve the net energy flow data of the agricultural and pastoral park. Step S3: Analyze the regulation fluctuation status of the agricultural and pastoral park based on the net energy flow data, select and mark the execution feature algorithm based on the regulation fluctuation status, obtain multi-element energy consumption parameter information and sort each multi-element energy consumption parameter before executing the marked feature algorithm; Step S4: Based on the sorting results, classify the multi-element energy parameters into core parameters and non-core parameters. After performing link shortening processing on the non-core parameters, detect the computing power margin of the core parameters, and determine whether to perform accuracy compensation on the non-core parameters based on the computing power margin.

[0007] In a preferred embodiment, in step S1, when performing the carbon emission estimation task at the edge node to be tested in the agricultural and pastoral park, a preset default setting table is accessed to obtain the scheduling cycle information corresponding to the edge node to be tested. Scheduling cycle information refers to the time limit for the edge node under test to complete one computation task and output the result during the operation process; Record the actual completion time of the carbon emission estimation task, and obtain the effective time window length based on the time interval between the actual completion time and the end time of the scheduling cycle; The window compression ratio is obtained by normalizing the effective time window length and the scheduling cycle length.

[0008] In a preferred embodiment, in step S1, the time difference between the start time and the end time of carbon emission estimation task being scheduled for execution is calculated from the start time of carbon emission estimation task to the end time of carbon emission calculation result being generated and written to local cache, and is used as the carbon emission calculation duration. The carbon emission calculation time is calculated as a ratio to the corresponding scheduling cycle length to obtain the calculation time percentage. The average carbon emission calculation time is obtained by calculating the average percentage of carbon emission calculation time recorded by the edge node under test within the historical scheduling cycle. Calculate the difference between the current calculation time percentage and the average carbon emission calculation time to generate parameter estimation features.

[0009] In a preferred embodiment, in step S2, the parameter estimation features and window compression ratio are normalized to ensure that both participate in subsequent calculations in a dimensionless form. Subsequently, the parameter estimation features and window compression rate are weighted and summed according to the preset weighting coefficients to obtain the computational pressure index of the current scheduling cycle. When the calculated pressure index is less than the pressure judgment threshold, the edge node to be tested is determined not to enter the control identification mechanism. When the calculated pressure index is greater than or equal to the pressure judgment threshold, the edge node to be tested is determined to enter the control and identification mechanism.

[0010] In a preferred embodiment, in step S2, after determining that the edge node to be tested has entered the control and identification mechanism, a data retrieval operation is triggered to obtain the net energy flow data of the scheduling cycle. Using the start and end times of the scheduling cycle as time boundaries, the energy flow data collected by each energy-consuming unit within the scheduling cycle is processed for time alignment and direction identification: Energy input is defined as positive energy flow, and energy output is defined as negative energy flow. The difference between the positive and negative energy flows of the same energy-consuming unit is calculated to obtain the net energy flow value of each energy-consuming unit within the scheduling cycle. The net energy flow value of each energy-consuming unit within a continuous scheduling cycle is used as the net energy flow data.

[0011] In a preferred embodiment, in step S3, after arranging the net energy flow data in chronological order, the difference between adjacent net energy flow data is calculated and divided by the time interval between adjacent sampling times to obtain the change in net energy flow. The maximum value of the net energy flow change is taken as the intensity of the control fluctuation. If the intensity of the regulated fluctuation is greater than or equal to the regulated fluctuation threshold, the agricultural and pastoral park is determined to be in a regulated fluctuation state; otherwise, the agricultural and pastoral park is determined to be in a non-regulated fluctuation state. The feature algorithm is selected based on the state of the regulated fluctuations. The feature algorithm includes the default feature algorithm and the sorted feature algorithm. When an agricultural and pastoral park is determined to be in a state of uncontrolled fluctuation, the default feature algorithm is selected to be executed. When the agricultural and pastoral park is in a state of regulated fluctuation, the edge node to be tested is selected to execute the sorting feature algorithm; The selected feature algorithms are labeled.

[0012] In a preferred embodiment, in step S3, when the marked feature algorithm is the default feature algorithm, the multi-electrode energy parameters are maintained in the way that the corresponding power sequence is used for calculation during the scheduling period, without triggering link shortening and accuracy compensation processing. When the labeling feature algorithm is a sorting feature algorithm, the multi-electrode energy parameter information is obtained through the local running log. The multi-electrode energy parameter information refers to the multi-electrode energy parameters that participate in the calculation of the net energy flow of the park within the same scheduling cycle. Among them, the multi-electrode energy parameters include the load power set of the energy load within the scheduling cycle and the charging and discharging power set of the energy storage unit within the scheduling cycle. Each multi-element energy parameter corresponds to a unique power sequence arranged according to the sampling time.

[0013] In a preferred embodiment, in step S3, the sorting feature algorithm specifically involves: for each multi-element energy consumption parameter, calculating the average change amplitude by performing adjacent change calculations on its corresponding power sequence within the scheduling period; The average change range of each multi-element energy parameter is used as the sorting criterion. The multi-element energy parameters are sorted from largest to smallest value to obtain the sorting result of the multi-element energy parameters.

[0014] In a preferred embodiment, in step S3, in the sorting results of the multi-electro-energy parameters, the multi-electro-energy parameter corresponding to the maximum average change value is taken as the core parameter, and the multi-electro-energy parameter corresponding to the minimum average change value is taken as the non-core parameter. For the core parameters, the edge nodes under test participate in the subsequent carbon emission estimation calculation during the scheduling cycle; For non-core parameters, link shortening is performed. The actual computation time of core parameters involved in carbon emission estimation is detected through the task timing record interface. The remaining computation time is obtained by subtracting the scheduling cycle time from the actual computation time. The ratio of the remaining computation time to the scheduling cycle time is used as the computing power margin. If the computing power margin is greater than the preset computing power margin threshold, it is determined that precision compensation processing will be performed on non-core parameters. Conversely, it is determined that no precision compensation will be performed on non-core parameters.

[0015] The technical effects and advantages of this invention are as follows: This invention utilizes edge nodes as the core computing carrier and introduces an operation monitoring mechanism based on scheduling cycles to quantitatively evaluate the carbon emission calculation time and the timeliness of result output. It constructs a computational pressure index, determines the operational status of edge nodes through a control and identification mechanism, retrieves net energy flow data from the industrial park, analyzes the park's control fluctuations, and selects different computational strategies and parameter processing paths. By sorting and classifying diverse energy consumption parameters and implementing accuracy compensation or link shortening processing based on computing power margins, the computational load is adaptively adjusted according to the operational status. This ensures the timeliness of carbon emission estimation while improving the utilization efficiency of edge-side computing resources, enhancing the real-time response capability of agricultural and pastoral parks in participating in power grid control and low-carbon assessment. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the implementation of a flexible and scalable data collection method for diversified energy use and carbon emissions in agricultural and pastoral parks according to the present invention.

[0017] Figure 2 This is a schematic diagram illustrating the steps of a flexible and scalable method for collecting data on diversified energy use and carbon emissions in agricultural and pastoral parks according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention utilizes edge nodes as the core computing carrier and introduces a scheduling cycle-based operation monitoring mechanism to quantitatively evaluate the carbon emission calculation time and the timeliness of result output. It constructs a computing pressure index, determines the operating status of edge nodes through a control and identification mechanism, retrieves net energy flow data from the industrial park, analyzes the park's control fluctuations, and selects different computing strategies and parameter processing paths. By sorting and classifying diverse energy consumption parameters and combining them with computing power margins to implement accuracy compensation or link shortening, the computing load is adaptively adjusted according to the operating status, thereby improving the utilization efficiency of edge computing resources while ensuring the timeliness of carbon emission estimation.

[0020] Example 1, such as Figures 1 to 2 As shown, a flexible and scalable method for collecting data on diversified energy use and carbon emissions in agricultural and pastoral parks includes the following steps: Step S1: When performing carbon emission estimation in the agricultural and pastoral park, access the default setting table to obtain the scheduling cycle information of the edge node to be tested and detect the window compression rate and carbon emission calculation time within the scheduling cycle. Evaluate the parameter estimation characteristics of the edge node to be tested based on the carbon emission calculation time. Step S2: Generate the computational pressure index for the current scheduling cycle by combining the comprehensive parameter estimation characteristics and window compression rate. Use the computational pressure index to determine whether to enter the control identification mechanism. If the control identification mechanism is entered, retrieve the net energy flow data of the agricultural and pastoral park. Step S3: Analyze the regulation fluctuation status of the agricultural and pastoral park based on the net energy flow data, select and mark the execution feature algorithm based on the regulation fluctuation status, obtain multi-element energy consumption parameter information and sort each multi-element energy consumption parameter before executing the marked feature algorithm; Step S4: Based on the sorting results, classify the multi-element energy parameters into core parameters and non-core parameters. After performing link shortening processing on the non-core parameters, detect the computing power margin of the core parameters, and determine whether to perform accuracy compensation on the non-core parameters based on the computing power margin.

[0021] The specific implementation is as follows: In step S1, when performing the carbon emission estimation task at the edge node to be tested in the agricultural and pastoral park, the preset default setting table is accessed to obtain the scheduling cycle information corresponding to the edge node to be tested.

[0022] Scheduling cycle information refers to the time limit for the edge node under test to complete one calculation task and output the result during operation. It is used to describe the time rhythm of the edge node under test performing periodic calculations and updating results during the processing of multi-energy consumption and carbon emission data.

[0023] It should be noted that the default settings table is a configuration data structure pre-stored in the edge computing system of the agricultural and pastoral park, used to centrally record basic parameter information related to edge node operation control and computing scheduling.

[0024] After obtaining the scheduling cycle information, the execution timing of the carbon emission estimation task within the scheduling cycle is monitored, using the start and end times of the scheduling cycle as time boundaries. Specifically, the actual completion time of the carbon emission estimation task is recorded, and based on the time interval between the actual completion time and the end time of the scheduling cycle, the effective time window length for data transmission of the carbon emission estimation results within the scheduling cycle is obtained.

[0025] The window compression ratio is obtained by normalizing the effective time window length and the scheduling cycle length.

[0026] The window compression ratio quantifies the degree of lag between the carbon emission estimation results and the output of the scheduling cycle. The larger the window compression ratio value, the more significantly the effective time window is compressed relative to the scheduling cycle, reflecting the lower the computation timeliness of the edge nodes in the current scheduling cycle.

[0027] Meanwhile, during the scheduling period, the entire process of the carbon emission estimation task executed by the edge node is timed and recorded. Specifically, the time difference between the start time and the completion time is calculated from the start time when the carbon emission estimation task is scheduled to be executed until the completion time when the carbon emission calculation result is generated and written to the local cache. This time difference is used as the carbon emission calculation duration.

[0028] Carbon emission calculation time is quantified in terms of time length, reflecting the actual computing time consumed by an edge node to complete a carbon emission calculation under the combined effect of current computing power and task scale.

[0029] By continuously monitoring the carbon emission calculation time, the changing trend of the computing load of edge nodes can be objectively characterized, providing a quantitative basis for subsequent evaluation of parameter estimation characteristics and computing pressure status.

[0030] Specifically, using the scheduling cycle length as the time benchmark, the carbon emission calculation time is normalized to obtain the proportion of time that carbon emission calculation time accounts for within the scheduling cycle. That is, the ratio of carbon emission calculation time to the corresponding scheduling cycle length is calculated to obtain the calculation time percentage. This percentage represents the proportion of time resources consumed by edge nodes to complete a single carbon emission estimation task within a complete scheduling cycle.

[0031] Based on this, the carbon emission calculation time percentage recorded by the edge node under test in historical scheduling cycles is compared and analyzed with the calculation time percentage in the current scheduling cycle. Specifically, the average carbon emission calculation time is obtained by calculating the mean of the carbon emission calculation time percentage recorded by the edge node under test in historical scheduling cycles. The difference between the current calculation time percentage and the average carbon emission calculation time is then calculated to generate parameter estimation features that characterize the load level of the edge node under test.

[0032] The parameter estimation load characteristics reflect the time intensity of computing resources occupied by the edge node under test during the process of performing multi-economy energy parameter estimation and carbon emission estimation under the current operating state. The higher the parameter estimation characteristic value, the greater the proportion of carbon emission calculation time in the scheduling cycle, indicating that the edge node under test needs to consume more computing time when performing parameter estimation, that is, the parameter estimation process occupies a higher degree of computing power of the edge node; conversely, the lower the parameter estimation characteristic, the faster the carbon emission calculation is completed within the scheduling cycle, and the lower the time occupied by computing power in the parameter estimation process.

[0033] Through the above calculation method, the carbon emission calculation time is transformed into parameter estimation characteristics that can reflect the parameter estimation behavior. This allows the parameter estimation load status of the edge node under test to be quantitatively evaluated under different scheduling cycles, different task scales, and different computing power conditions. This provides objective and comparable basic parameters for the subsequent construction of the computing pressure index and adjustment of computing strategies.

[0034] In step S2, the parameter estimation features and window compression ratio are comprehensively calculated within the current scheduling cycle to generate a computational pressure index that characterizes the overall operating load status of the edge node under test.

[0035] Specifically, the parameter estimation features and window compression ratio are normalized to ensure that both participate in subsequent calculations in a dimensionless form.

[0036] Subsequently, the parameter estimation features and window compression rate are weighted and summed according to the preset weighting coefficients to obtain the computational pressure index of the current scheduling cycle.

[0037] Among them, the weighting coefficient corresponding to the parameter estimation feature reflects the degree of influence of the computing load on the operating pressure of the edge node under test, and the weighting coefficient corresponding to the window compression rate reflects the degree of influence of the decrease in the timeliness of the result on the operating pressure of the edge node under test. The weighting coefficient is set according to historical operating statistics. Specifically, it is obtained by analyzing the operating records of the edge node under test in multiple historical scheduling cycles. The correlation between the parameter estimation feature and the window compression rate on computing timeout events, result lag events, or control command delay events in the historical scheduling cycles is statistically analyzed. The correlation index between the two and the frequency of the above-mentioned abnormal events is calculated, and the correlation index is normalized to obtain the corresponding weighting coefficient.

[0038] It should be noted that standardization refers to the process of mapping raw data of different physical quantities or different dimensions to a uniform dimension, uniform numerical range or uniform statistical distribution through a specific mathematical transformation. Standardization methods include, but are not limited to, standard linear transformation based on interval scaling, Z-Score standardization based on statistics or normalization method based on nonlinear mapping function. The application methods of standardization will not be elaborated here.

[0039] The computational pressure index comprehensively characterizes the computational complexity pressure and time constraint pressure faced by the edge node under test within the current scheduling cycle. The higher the computational pressure index value, the more computational time resources the edge node consumes and the more time the effective time window for outputting results is compressed when completing carbon emission estimation and multi-energy consumption data processing tasks within the current scheduling cycle, resulting in a higher overall computational pressure level. Conversely, the lower the computational pressure index value, the lighter the computational load of the edge node within the current scheduling cycle, the better the timeliness of result output, and the relatively relaxed overall operating status.

[0040] The calculated pressure index is compared with a preset pressure threshold to determine whether the edge node under test has entered the control and identification mechanism during the current scheduling cycle. When the computational pressure index is less than the pressure judgment threshold, it is determined that the computational pressure of the edge node under test is within a controllable range in the current scheduling cycle and will not enter the control and identification mechanism. When the computational pressure index is greater than or equal to the pressure judgment threshold, the edge node under test is determined to have significant computational pressure in the current scheduling cycle and enters the control and identification mechanism.

[0041] It should be noted that the pressure judgment threshold is set based on the operational statistics of historical scheduling cycles. Specifically, the computational pressure index of the edge node under test is collected in advance for multiple historical scheduling cycles under stable operation, and the computational pressure index is statistically analyzed to obtain its distribution range. Based on this, the critical value in the distribution range of the computational pressure index that can distinguish between normal operation and abnormal states such as computation delay, result lag, or control command timeout is selected as the pressure judgment threshold.

[0042] After determining that the edge node to be tested has entered the control and identification mechanism, a data retrieval operation is triggered to obtain the net energy flow data of the scheduling cycle.

[0043] Net energy flow data is calculated using a unified metering standard from the energy input and output of various energy-consuming units within the park. Specifically, it is the difference between the energy input and output of multiple energy sources, such as electricity, heat, and biomass energy, under the same time reference. In practice, the start and end times of the scheduling cycle are used as time boundaries. The energy flow data collected by each energy-consuming unit within the scheduling cycle is time-aligned and direction-identified. Energy input is defined as positive energy flow, and energy output is defined as negative energy flow. The difference between the positive and negative energy flows of the same energy-consuming unit is calculated to obtain the net energy flow value of each energy-consuming unit within the scheduling cycle.

[0044] The net energy flow value of each energy-consuming unit within a continuous scheduling cycle is used as the net energy flow data.

[0045] In step S3, after arranging the net energy flow data in chronological order, the difference between adjacent net energy flow data is calculated and divided by the time interval between adjacent sampling times to obtain the change in net energy flow. The maximum value of the net energy flow change is taken as the intensity of the control fluctuation. The intensity of the regulated fluctuation is compared with the preset regulated fluctuation threshold. When the intensity of the regulated fluctuation is greater than or equal to the regulated fluctuation threshold, the agricultural and pastoral park is determined to be in a regulated fluctuation state; when the intensity of the regulated fluctuation is less than the regulated fluctuation threshold, the agricultural and pastoral park is determined to be in a non-regulated fluctuation state. The feature algorithm is selected based on the control fluctuation state. The feature algorithm is a set of algorithms pre-set in the local algorithm library of the edge node to be tested, including the default feature algorithm and the sorted feature algorithm. The default feature algorithm is used for the baseline feature calculation process in the stable operation scenario of agricultural and pastoral parks. It performs feature extraction on each of the multi-energy parameters collected during the scheduling cycle and generates a complete feature vector. The ranking feature algorithm is used for the priority calculation process of multi-energy parameters in the scenario of regulation fluctuation. It generates the importance ranking result of multi-energy parameters for subsequent parameter control steps.

[0046] When it is determined that the agricultural and pastoral park is in a state of regulated fluctuation, the edge node to be tested selects to execute the sorting feature algorithm; When an agricultural and pastoral park is determined to be in a state of uncontrolled fluctuation, the default feature algorithm is selected to be executed. The selected feature algorithms are labeled; When the marking feature algorithm is the default feature algorithm, the multi-electrode energy consumption parameters are calculated according to the corresponding power sequence during the scheduling period, without triggering link shortening and accuracy compensation processing; When the labeling feature algorithm is a sorting feature algorithm, the multi-electrode energy parameter information is obtained through the local running log. The multi-electrode energy parameter information refers to the multi-electrode energy parameters that participate in the calculation of the net energy flow of the park within the same scheduling cycle. Among them, the multi-electrode energy parameters include the load power set of the energy load within the scheduling cycle and the charging and discharging power set of the energy storage unit within the scheduling cycle. Each multi-element energy parameter corresponds to a unique power sequence arranged according to the sampling time; The ranking feature algorithm is as follows: For each multi-element energy consumption parameter, the adjacent change amount is calculated for its corresponding power sequence within the current scheduling period. The adjacent change amount is calculated by taking the difference between adjacent values ​​in each power sequence and taking the absolute value. The average change amount is obtained by taking the average of the adjacent change amounts. The larger the average change amount, the more significant the impact of the multi-element energy consumption parameter on the energy balance change within the scheduling period. The average change range of each multi-element energy parameter is used as the sorting criterion. The multi-element energy parameters are sorted from largest to smallest value to obtain the sorting result of the multi-element energy parameters.

[0047] It should be noted that the preset control fluctuation threshold can be set based on the historical net energy flow change statistics of the agricultural and pastoral park; the local operation log is used to record data such as multi-dimensional energy consumption parameters and feature algorithm marking results within the scheduling cycle.

[0048] In step S4, in the sorting results of the multi-electrode energy parameters, the multi-electrode energy parameter corresponding to the maximum average change value is taken as the core parameter, and the multi-electrode energy parameter corresponding to the minimum average change value is taken as the non-core parameter. For the core parameters, the edge nodes under test participate in the subsequent carbon emission estimation calculation during the scheduling cycle; For non-core parameters, a link shortening process is performed. This means that during the scheduling cycle, real-time updates and change calculations are not performed on non-core parameters at each sampling time. Instead, the periodic average value of the power sequence corresponding to the non-core parameter in the previous scheduling cycle is used as a fixed input value to participate in the subsequent carbon emission estimation calculation. This reduces the number of calculation calls for non-core parameters during the scheduling cycle, thereby reducing the overall computational load. The periodic average value is calculated by the edge node at the end of the previous scheduling cycle based on the power sequence of this non-core parameter in the previous scheduling cycle, and stored in the local operation log.

[0049] After shortening the link for non-core parameters, the actual computation time of core parameters participating in carbon emission estimation is detected through the task timing record interface. The remaining computation time is obtained by subtracting the scheduling cycle time from the actual computation time. The ratio of the remaining computation time to the scheduling cycle time is used as the computing power margin, reflecting the remaining available computing power of the edge node under test after completing the core parameter calculation. The remaining computing power is compared with a preset computing power threshold to determine whether to perform precision compensation for non-core parameters. When the remaining computing power is greater than the preset threshold, it is determined that the edge node under test still has remaining computing power, and precision compensation processing is performed on non-core parameters. When the computing power margin is less than or equal to the computing power margin threshold, it is determined to maintain the link shortening processing state for non-core parameters and no precision compensation is performed for non-core parameters. Among them, the accuracy compensation process refers to switching the calculation method of non-core parameters from the periodic average value to the real-time power sequence recorded at the sampling time.

[0050] It should be noted that the task timing recording interface is used to record the start and end times of the carbon emission estimation calculation task; the preset computing power reserve threshold can be set according to the scheduling cycle length and historical computing load.

[0051] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0052] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0053] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0054] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0055] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A flexible scalable agri-pastoral park multi-purpose energy and carbon emission data collection method, characterized in that: Includes the following steps: ​ Step S1: When performing carbon emission estimation in the agricultural and pastoral park, access the default setting table to obtain the scheduling cycle information of the edge node to be tested and detect the window compression rate and carbon emission calculation time within the scheduling cycle. Evaluate the parameter estimation characteristics of the edge node to be tested based on the carbon emission calculation time. Step S2: Generate the computational pressure index for the current scheduling cycle by combining the comprehensive parameter estimation characteristics and window compression rate. Use the computational pressure index to determine whether to enter the control identification mechanism. If the control identification mechanism is entered, retrieve the net energy flow data of the agricultural and pastoral park. Step S3: Analyze the regulation fluctuation status of the agricultural and pastoral park based on the net energy flow data, select and mark the characteristic algorithm to be executed based on the regulation fluctuation status, obtain multi-element energy consumption parameter information and sort each multi-element energy consumption parameter before executing the marked characteristic algorithm; Step S4: Based on the sorting results, classify the multi-element energy parameters into core parameters and non-core parameters. After performing link shortening processing on the non-core parameters, detect the computing power margin of the core parameters, and determine whether to perform accuracy compensation on the non-core parameters based on the computing power margin.

2. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 1, characterized in that: In step S1, when performing the carbon emission estimation task at the edge node to be tested in the agricultural and pastoral park, the preset default setting table is accessed to obtain the scheduling cycle information corresponding to the edge node to be tested. Scheduling cycle information refers to the time limit for the edge node under test to complete one computation task and output the result during the operation process; Record the actual completion time of the carbon emission estimation task, and obtain the effective time window length based on the time interval between the actual completion time and the end time of the scheduling cycle; The window compression ratio is obtained by normalizing the effective time window length and the scheduling cycle length.

3. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 1, characterized in that: In step S1, starting from the start time when the carbon emission estimation task is scheduled to be executed, and ending at the completion time when the carbon emission calculation result is generated and written to the local cache, the time difference between the start time and the completion time is calculated as the carbon emission calculation duration. The carbon emission calculation time is calculated as a ratio to the corresponding scheduling cycle length to obtain the calculation time percentage. The average carbon emission calculation time is obtained by calculating the average percentage of carbon emission calculation time recorded by the edge node under test within the historical scheduling cycle. Calculate the difference between the current calculation time percentage and the average carbon emission calculation time to generate parameter estimation features.

4. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 3, characterized in that: In step S2, the parameter estimation features and window compression ratio are normalized to ensure that both participate in subsequent calculations in a dimensionless form. Subsequently, the parameter estimation features and window compression rate are weighted and summed according to the preset weighting coefficients to obtain the computational pressure index of the current scheduling cycle. When the calculated pressure index is less than the pressure judgment threshold, the edge node to be tested is determined not to enter the control identification mechanism. When the calculated pressure index is greater than or equal to the pressure judgment threshold, the edge node to be tested is determined to enter the control and identification mechanism.

5. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 4, characterized in that: In step S2, after determining that the edge node to be tested has entered the control and identification mechanism, a data retrieval operation is triggered to obtain the net energy flow data of the scheduling cycle. Using the start and end times of the scheduling cycle as time boundaries, the energy flow data collected by each energy-consuming unit within the scheduling cycle is processed for time alignment and direction identification: Energy input is defined as positive energy flow, and energy output is defined as negative energy flow. The difference between the positive and negative energy flows of the same energy-consuming unit is calculated to obtain the net energy flow value of each energy-consuming unit within the scheduling cycle. The net energy flow value of each energy-consuming unit within a continuous scheduling cycle is used as the net energy flow data.

6. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 1, characterized in that: In step S3, after arranging the net energy flow data in chronological order, the difference between adjacent net energy flow data is calculated and divided by the time interval between adjacent sampling times to obtain the change in net energy flow. The maximum value of the net energy flow change is taken as the intensity of the control fluctuation. If the intensity of the control fluctuation is greater than or equal to the control fluctuation threshold, the agricultural and pastoral park is determined to be in a control fluctuation state. Conversely, if the situation is as described above, the agricultural and pastoral park is determined to be in a state of uncontrolled fluctuation. The feature algorithm is selected based on the state of the regulated fluctuations. The feature algorithm includes the default feature algorithm and the sorted feature algorithm. When an agricultural and pastoral park is determined to be in a state of uncontrolled fluctuation, the default feature algorithm is selected to be executed. When the agricultural and pastoral park is in a state of regulated fluctuation, the edge node to be tested is selected to execute the sorting feature algorithm; The selected feature algorithms are labeled.

7. The flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 6, characterized in that: In step S3, when the feature algorithm is the default feature algorithm, the multi-electrode energy consumption parameters are calculated according to the corresponding power sequence during the scheduling period, without triggering link shortening and accuracy compensation processing. When the labeling feature algorithm is a sorting feature algorithm, the multi-element energy consumption parameter information is obtained through the local running log. The multi-element energy consumption parameter information refers to the multi-element energy consumption parameters that participate in the calculation of the net energy flow of the park within the same scheduling cycle. Among them, the multi-element energy consumption parameters include the load power set of the energy load within the scheduling cycle and the charging and discharging power set of the energy storage unit within the scheduling cycle. Each multi-element energy parameter corresponds to a unique power sequence arranged according to the sampling time.

8. A flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 7, characterized in that: In step S3, the sorting feature algorithm is specifically as follows: for each multi-element energy consumption parameter, the average change amplitude is obtained by performing adjacent change calculations on the corresponding power sequence within the scheduling period; The average change range of each multi-element energy parameter is used as the sorting criterion. The multi-element energy parameters are sorted from largest to smallest value to obtain the sorting result of the multi-element energy parameters.

9. A flexible and scalable method for collecting multi-energy consumption and carbon emission data in agricultural and pastoral parks according to claim 8, characterized in that: In step S3, in the sorting results of the multi-electrode energy parameters, the multi-electrode energy parameter corresponding to the maximum average change value is taken as the core parameter, and the multi-electrode energy parameter corresponding to the minimum average change value is taken as the non-core parameter. For the core parameters, the edge nodes under test participate in the subsequent carbon emission estimation calculation during the scheduling cycle; For non-core parameters, link shortening is performed. The actual computation time of core parameters involved in carbon emission estimation is detected through the task timing record interface. The remaining computation time is obtained by subtracting the scheduling cycle time from the actual computation time. The ratio of the remaining computation time to the scheduling cycle time is used as the computing power margin. If the computing power margin is greater than the preset computing power margin threshold, it is determined that precision compensation processing will be performed on non-core parameters. Conversely, it is determined that no precision compensation will be performed on non-core parameters.

Images