A Coordinated Control Method for Oilfield Microgrids that Takes into Account Fracturing Load and Energy Storage Dynamic Constraints

Abstract

This application relates to the field of microgrid collaborative control technology, specifically to a collaborative control method for oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints. The method includes: determining the comprehensive response value for each fracturing operation cycle based on the power fluctuation characteristics and operation time of each operation period in historical fracturing operation cycles; calculating the differences in overall load change characteristics between different operation cycles; clustering all historical operation cycles based on the differences in stage response changes between different operation cycles; constructing a demand response model for each group of operation cycles; determining the corresponding demand response model based on fracturing load prediction data; calculating the feasible range for accurate fracturing demand response; and then combining this with an energy storage dynamic constraint model to achieve collaborative optimization control of the microgrid, thereby improving the control accuracy and operational stability of the oilfield microgrid.

Description

Technical Field

[0001] This application relates to the field of microgrid collaborative control technology, specifically to a collaborative control method for oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints. Background Technology

[0002] The coordinated control of oilfield microgrids is a core means to achieve green shale oil extraction, power supply stability under high energy-consuming loads, improve the absorption of new energy sources, and reduce overall application costs. Through oilfield microgrid system control technology, unified scheduling and coordinated operation of wind power, photovoltaic power, gas turbines, diesel generators, energy storage systems, and various oilfield loads can be achieved. This effectively suppresses power fluctuations, stabilizes voltage frequency, and reduces fossil energy consumption during the operation of the oilfield microgrid, meeting the power supply needs for continuous production under off-grid or weak grid conditions.

[0003] However, due to the characteristics of electric fracturing loads—high power, strong impact, intermittency, and rigid process timing—power spikes and frequent fluctuations can occur during the coordinated control of oilfield microgrids, causing control commands to deviate significantly from actual needs. Furthermore, because energy storage systems have dynamic constraints such as SOC range, charging and discharging power, and SOH (State of Health), traditional oilfield microgrid control technologies do not adjust operating boundaries based on battery status and ignore the impact of fracturing loads on control. This leads to overcharging, over-discharging, exceeding rate limits, and rapid degradation, reducing the suppression effect on fluctuations, increasing control errors, and consequently causing inaccurate power distribution, frequency and voltage exceeding limits, and accelerated equipment lifespan loss, affecting the accuracy and stability of oilfield microgrid coordinated control. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a collaborative control method for oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints, thereby resolving the existing issues.

[0005] The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints in this application adopts the following technical solution:

[0006] One embodiment of this application provides a collaborative control method for oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints. The method includes the following steps:

[0007] Obtain power data and start / stop commands for fracturing load in each historical fracturing operation cycle;

[0008] Based on the start and stop commands, obtain each operation time period in each operation cycle; analyze the power fluctuation and power distribution disorder of the fracturing load in each operation time period, construct the load change characteristic value of each operation time period, and determine the comprehensive response value of each operation cycle by combining the duration proportion of each operation time period in the operation cycle.

[0009] Obtain the load change characteristic value of each work cycle; construct the comprehensive similarity distance between any two work cycles based on the difference in comprehensive response value and the difference in load change characteristic value between any two work cycles; combine with the clustering algorithm to cluster all historical work cycles to obtain the set of each group of work cycles;

[0010] Demand response models for each set of work cycles are constructed using start / stop commands and planned durations for each work cycle. Fracturing load data for future work cycles is predicted using historical data. Based on the comprehensive similarity distance between future work cycles and historical work cycles, the demand response models that need to be input for fracturing load data in future work cycles are determined and input to obtain the feasible demand response intervals for future work cycles, which are then used for microgrid coordinated control.

[0011] In one embodiment, the process of obtaining the load change characteristic value is as follows:

[0012] Calculate the range of fracturing load power data for each operation period to determine the amplitude of load fluctuation for each operation period;

[0013] Calculate the coefficient of variation of all fracturing load power data for each operation period;

[0014] The load change characteristic value for each working time period is determined based on the amplitude value and the coefficient of variation.

[0015] In one embodiment, the amplitude of the load fluctuation during each working period is the ratio of the range to a preset rated maximum power value.

[0016] In one embodiment, the load change characteristic value is the product of the amplitude value and the coefficient of variation.

[0017] In one embodiment, the comprehensive response value is positively correlated with the load change characteristic value of each operation time period and the duration ratio.

[0018] In one embodiment, the process of obtaining the comprehensive response value is as follows:

[0019] Calculate the product of the load change characteristic value for each work period and the duration percentage; the comprehensive response value for each work cycle is the sum of the products of all work periods in each work cycle.

[0020] In one embodiment, the comprehensive similarity distance is positively correlated with the difference in the comprehensive response value and the difference in the load change characteristic value.

[0021] In one embodiment, the comprehensive similarity distance is the weighted sum of the differences in the comprehensive response values ​​and the differences in the load change characteristic values.

[0022] In one embodiment, the process of determining the input demand response model for fracturing load data in future operation cycles based on the comprehensive similarity distance between future operation cycles and historical operation cycles is as follows:

[0023] The comprehensive difference value between the future operation cycle and each set of historical operation cycles is determined based on the comprehensive similarity distance between the future operation cycle and each set of operation cycles. The demand response model of the set of operation cycles corresponding to the minimum comprehensive difference value is used as the demand response model that needs to be input into the fracturing load data of the future operation cycle.

[0024] In one embodiment, the comprehensive difference value is the mean of the comprehensive similarity distances between the future work cycle and all work cycles in each set of work cycles.

[0025] This application has at least the following beneficial effects:

[0026] This application collects power data and start / stop commands for fracturing loads from historical fracturing operation cycles; determines the comprehensive response value for each operation cycle based on power fluctuation characteristics and operation time for each time period within each operation cycle; the difference in comprehensive response values ​​between different operation cycles reflects the stage response variation differences of fracturing loads under complex operating conditions; calculates the overall load variation characteristics differences between different operation cycles, and combines the stage response variation differences to conduct a comparative analysis of the stage change response of fracturing load data for complete cycles of different historical fracturing operations, thereby determining the demand response model of fracturing load under different operating conditions under complex operating conditions. This allows for the determination of an accurate feasible range for fracturing demand response during subsequent microgrid collaborative control based on the predicted fracturing load data for different fracturing operation cycles. Furthermore, it combines this with an energy storage dynamic constraint model to achieve microgrid collaborative optimization control, improving the control accuracy and operational stability of the oilfield microgrid. This avoids the problem of large errors in microgrid collaborative control caused by traditional oilfield microgrid control technologies that fail to adjust operating boundaries based on battery status and ignore the impact of fracturing load on control. Attached Figure Description

[0027] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A flowchart of the oilfield microgrid collaborative control method that takes into account both fracturing load and energy storage dynamic constraints provided in this application;

[0029] Figure 2 This is a schematic diagram illustrating the process of obtaining characteristic values ​​of load changes. Detailed Implementation

[0030] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the oilfield microgrid coordinated control method that balances fracturing load and energy storage dynamic constraints proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0032] The following description, in conjunction with the accompanying drawings, details the specific scheme of the oilfield microgrid collaborative control method that takes into account both fracturing load and energy storage dynamic constraints provided in this application.

[0033] One embodiment of this application provides a collaborative control method for oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints.

[0034] Specifically, the following collaborative control method for oilfield microgrids, which takes into account both fracturing load and energy storage dynamic constraints, is provided. Please refer to [link / reference]. Figure 1 The method includes the following steps:

[0035] Step S1: Obtain power data and start / stop commands for fracturing load in each historical fracturing operation cycle.

[0036] To achieve oilfield microgrid system control that balances fracturing load and energy storage dynamic constraints, this application utilizes multiple types of sensors, RTUs, BMS, fracturing truck control systems, and meteorological monitoring units to synchronously acquire multi-source heterogeneous data for oilfield microgrid control. Specifically, real-time power output data of photovoltaics, real-time power output data of wind power, and ambient wind speed and solar irradiance data are acquired through the grid-connected photovoltaic inverter, wind turbine converter, and on-site meteorological station, respectively. Start and stop commands during fracturing operations are obtained through the electric fracturing truck PLC control system, the number of fracturing operation segments and planned duration are obtained through the fracturing operation monitoring host computer, and the real-time power of the fracturing load is obtained through the power transmitter. The SOC, cycle count, and SOH of the energy storage system are obtained through the energy storage battery management system (BMS), the total voltage, individual cell voltage, and charging / discharging current of the energy storage system are obtained through the energy storage converter (PCS), and the temperature of the energy storage system is obtained through the battery cluster temperature acquisition module. All the above-mentioned data are synchronously acquired according to timestamps to ensure the consistency and integrity of the data acquisition sequence.

[0037] Step S2: Obtain each operation time period in each operation cycle according to the start and stop instructions; analyze the power fluctuation and power distribution disorder of the fracturing load in each operation time period, construct the load change characteristic value of each operation time period, and determine the comprehensive response value of each operation cycle by combining the duration proportion of each operation time period in the operation cycle.

[0038] Since the electric fracturing load has the highest power and the strongest fluctuation in the oilfield microgrid, the real-time power and fluctuation characteristics of the electric fracturing load directly determine the feasibility and safety of the oilfield microgrid collaborative control strategy. Therefore, it is necessary to analyze the core characteristics that can be used for demand response based on the changing characteristics of the fracturing load, so as to realize the source-load-storage collaborative optimization of the fracturing load under complex working conditions in multiple cycles.

[0039] Specifically, due to the significant variations in fracturing load during the operation of oilfield microgrids, depending on the construction cycle, operating conditions, weather conditions, and well site platform, single-cycle, single-condition models cannot cover the full range of operational characteristics. This can easily lead to problems such as large load prediction deviations, exceeding energy storage constraints, and insufficient control accuracy. Therefore, a multi-dimensional comparison and differential analysis of the fracturing load's operating characteristics in different cycles under complex conditions is conducted. First, the real-time power, start / stop commands, number of operation segments, and planned duration of the fracturing load for 500 complete historical operating cycles are obtained. Furthermore, considering that the fracturing load must meet the requirements of the extraction process, cannot be arbitrarily interrupted, and can only be adjusted within the allowable window, the real-time power, start / stop commands, number of operation segments, and planned duration of the fracturing load for each historical cycle are analyzed. It should be noted that the number of historical operating cycles obtained can be set by the implementer according to the actual situation; this application does not impose specific restrictions.

[0040] To accurately determine the relationship between fracturing load characteristics and microgrid power balance and energy storage constraints during the operation of the oilfield microgrid, this application divides the data collected in each cycle according to the start and stop times of each operation period in the entire operation process. The number of time periods is equal to the number of operation segments. The purpose of dividing the data according to operation periods is to accurately match the actual fracturing construction rhythm, so that the load fluctuation analysis corresponds one-to-one with the process stage, avoiding large deviations in characteristic analysis caused by cross-stage mixed analysis. The difference between the maximum and minimum values ​​of the real-time power data of the fracturing load in each divided time period is calculated, and the rated maximum power value of the electric fracturing load is obtained. The ratio of the difference to the rated maximum power value is used as the amplitude value of the load fluctuation in that time period. The larger the amplitude value, the more severe the power fluctuation in that operation period, the stronger the impact on the microgrid power, and the greater the need for rapid energy storage response and constraint protection.

[0041] Furthermore, considering that fluctuation amplitude alone cannot fully reflect the dispersion and stability characteristics of staged load fluctuations under different operating conditions, and is insufficient to comprehensively reflect the impact of fracturing load changes on energy storage and microgrid control, the coefficient of variation of all real-time power data within each time period is calculated. This coefficient of variation is recorded as the response value of load fluctuations for each time period. A larger response value indicates a more discrete load power distribution and more significant random fluctuations within that time period. Since fracturing operations may be conducted under different operating conditions and different time periods during actual microgrid operation, the fluctuation response characteristics of fracturing loads in different time periods within each cycle are analyzed. Then, the fluctuation response characteristics of all time periods are comprehensively analyzed for the entire cycle, thereby achieving a comprehensive understanding of the impact of fracturing load changes on energy storage and microgrid control. The analysis focuses on the full-cycle fluctuation characteristics of load fluctuation. Specifically, for each time period, the product of the load fluctuation amplitude and the response value is calculated. This product is recorded as the load change characteristic value for each time period. The larger the load change characteristic value, the more significant the impact of the operation period on the microgrid's impact intensity and control accuracy. The ratio of the duration of each time period to the overall duration of the corresponding cycle is calculated and recorded as the effective response weight. The larger the effective response weight, the greater the impact weight of the load change within that time period relative to the overall fluctuation characteristics of the entire cycle. This effectively highlights the impact of key operation periods when analyzing the overall load characteristics and control strategies of the cycle, improving the rationality and accuracy of fluctuation assessment and control optimization.

[0042] Furthermore, based on the above analysis and processing, in order to achieve accurate analysis of the overall fluctuation intensity and control impact characteristics of the fracturing load throughout the entire cycle, the comprehensive response value for each cycle is calculated by comprehensively considering the load change response and impact characteristics of all operating time periods within each cycle. The specific calculation formula is as follows:

[0043] ,in, Indicates the first The overall response value for each cycle; Indicates the first The first period is divided into the 1st period Load variation characteristic values ​​over a time period; Indicates the first The first period is divided into the 1st period Effective response weights for each time period; This indicates the number of time periods divided. The larger the calculated comprehensive response value, the more severe the overall fluctuation of the fracturing load within that period, the stronger the impact on the microgrid power balance, and the higher the required energy storage response and control regulation intensity.

[0044] Step S3: Obtain the load change characteristic value of each work cycle; construct the comprehensive similarity distance between any two work cycles based on the difference in comprehensive response value and the difference in load change characteristic value between any two work cycles; combine with the clustering algorithm to cluster all historical work cycles to obtain the set of each group of work cycles.

[0045] Furthermore, since the impact of the phased load changes of fracturing operations under different operating conditions on the power balance of the microgrid varies significantly during microgrid operation, the influence of the phased response change characteristics of fracturing operation load under complex operating conditions on the coordinated control of the oilfield microgrid, which takes into account both fracturing load and energy storage dynamic constraints, needs to be fully considered in the process of coordinated control of the oilfield microgrid. Specifically, after the above calculation and analysis, the comprehensive response value of each complete historical operation cycle was obtained. In order to further achieve accurate division of operation cycle data under complex operating conditions, this application calculates the absolute value of the difference between the comprehensive response values ​​of any two cycles, and records the absolute value as the phased response difference between any two cycles. The larger the phased response difference, the more significant the difference between the two operation cycles in terms of load fluctuation intensity, impact characteristics and control difficulty, and the more prominent the difference in overall operating characteristics.

[0046] Furthermore, to analyze the load change characteristics of the entire complete operation cycle, the calculation method is the same as for the load change characteristic value of each time period. Real-time power data of all fracturing loads corresponding to each cycle are used as input to calculate the load change characteristic value corresponding to each cycle. The absolute value of the difference between the load change characteristic values ​​of any two cycles is then calculated and denoted as the overall response difference between any two cycles, reflecting the differences in overall load change characteristics across different cycles. Based on the stage response difference and overall response difference determined above for any two cycles, the comprehensive similarity distance between any two cycles is calculated to accurately reflect the degree of difference in stage response changes under the influence of overall fluctuation characteristics and actual working conditions in different operation cycles. Preferably, in this embodiment, the expression for the comprehensive similarity distance can be:

[0047]

[0048] In the formula, Indicates the first The cycle and the first The comprehensive similarity distance between each cycle; The first The and the first Differences in stage response between cycles; Indicates the first The and the first The overall response difference between each cycle; The weighting coefficient represents the proportion of the weight of the stage response difference in the comprehensive similarity distance calculation. In this application, we consider that the stage fluctuations of fracturing operations have a more direct impact on the power balance of the microgrid and a more significant impact on the adaptability of the control strategy. Therefore, we need to highlight the weight of the stage characteristic differences. In this embodiment, the value is set to 0.6. Implementers can determine the weighting coefficient according to the complexity of the fracturing process, the microgrid's anti-fluctuation capability, and the energy storage response speed. This application does not impose specific limitations.

[0049] Furthermore, based on the above calculation results, in order to achieve accurate classification of fracturing operation data under complex working conditions, this application uses the comprehensive similarity distance between any two operation cycles as a classification criterion, and takes the real-time power, start / stop commands, number of operation segments, and planned duration of fracturing load for all historical cycles as input. The hierarchical clustering algorithm is used to group and classify all cycle data, resulting in a set of operation cycles with similar load characteristics in each group. Each group represents a set of operation cycles in a class of actual fracturing operation scenarios with similar working conditions, consistent load fluctuation characteristics, and similar control requirements. The specific process of grouping and classifying using the hierarchical clustering algorithm is well known to those skilled in the art and will not be described in detail here.

[0050] Step S4: Construct the demand response model for each set of work cycles using the start / stop instructions and planned duration of each work cycle in each set of work cycles; predict the fracturing load data for future work cycles using historical data; determine the demand response model that needs to be input for the fracturing load data of future work cycles based on the comprehensive similarity distance between future work cycles and historical work cycles, and input it to obtain the feasible demand response interval for future work cycles, which is used for microgrid coordinated control.

[0051] Furthermore, based on the data grouping results for various fracturing operation cycles under complex operating conditions, a demand response model for fracturing load is constructed to achieve accurate analysis of the demand response behavior of fracturing load under different scenarios. Specifically, firstly, based on the grouping results, the operable time window and minimum continuous operation duration for each cycle within each operation cycle set are determined according to the start / stop instructions and planned duration of fracturing operations in each cycle. Using the determined operable time windows and minimum continuous operation durations for all cycles in the set, along with the corresponding real-time power data of the fracturing load, as input, the Gurobi software is used to construct the corresponding demand response model for each operation cycle set. The demand response model for fracturing load under operating conditions takes the historical power trajectory within each set of operating cycles as input, uses the limitations of the fracturing load operation time shift range and the upper and lower limits of the charging and discharging power of the energy storage system as boundary constraints, and takes minimizing the total power offset of the system within the historical cycle as the objective function for simulation analysis. By traversing different shift limits under historical operating conditions, the maximum fracturing load offset that the microgrid resources can cover under such operating conditions is calculated, and this offset is defined as the feasible demand response interval corresponding to the set of operating conditions. The specific construction process of the demand response model for fracturing load is well known to those skilled in the art and will not be described in detail here.

[0052] Furthermore, in the actual operation of microgrids, in addition to the impact of fracturing load on microgrid operation and control, the dynamic constraints of energy storage also need to be considered due to the decay of battery state of health (SOH), real-time changes in state of charge (SOC), temperature, and cycle count. Therefore, a dynamic constraint model for energy storage is constructed based on the real-time operating status and aging characteristics of energy storage to achieve dynamic updates of the safe operating boundary of energy storage and precise constraints on charging and discharging behavior. Specifically, the SOC, total voltage, individual cell voltage, charging and discharging current, temperature, cycle count, and SOH of the energy storage system for all historical operating cycles are used as inputs. A three-layer fully connected neural network constructed using TensorFlow is used to train the data of the historical energy storage system to obtain the dynamic constraint model of energy storage. After training, the real-time state parameters of energy storage are used as inputs to the trained model, and the outputs are the current energy storage SOH, remaining lifetime (RUL), and dynamic SOC boundary.

[0053] To adapt to the practical engineering challenges of limited computing resources at the edge of oilfield microgrids, which prevent the execution of complex battery aging models, this application employs a knowledge distillation algorithm to lightweight the energy storage dynamic constraint model. Specifically, the aforementioned energy storage dynamic constraint model is used as a high-precision, complex teacher network to learn the degradation characteristics and state mapping patterns of energy storage throughout its entire lifecycle. The output of the teacher network is used as the training label for the student network, training a lightweight student network and transferring knowledge from the teacher network. During the training process, a distillation loss function guides the student network to approximate the knowledge distribution of the teacher network. The lightweight student network obtained after knowledge distillation is deployed on the field controller of the oilfield microgrid to achieve real-time and rapid estimation of energy storage SOH, RUL, and dynamic SOC boundaries. This addresses the mismatch between the high-precision model and edge computing capabilities, providing lightweight and real-time dynamic constraints for collaborative control. The specific construction of the energy storage dynamic constraint model and the process of lightweighting using knowledge distillation are well-known to those skilled in the art and will not be elaborated further.

[0054] In the collaborative control of oilfield microgrids, due to the strong randomness of wind and solar power output and the strong impact and uncertainty of fracturing load, traditional prediction methods cannot meet the requirements of high-precision power balance. Therefore, it is necessary to construct a source-load uncertainty prediction model to achieve accurate prediction of wind and solar power output and fracturing load in future periods. Specifically, historical environmental wind speed and light intensity data, wind and solar power output data, and fracturing load data are divided into training and testing sets in a 7:3 ratio. The prediction model is trained using a Transformer-LSTM hybrid deep learning model, where the loss function is a weighted combination of normalized root mean square error (RMSE) and mean absolute percentage error (MAPE), with RMSE and MAPE weights of 0.7 and 0.3, respectively. The optimizer is the Adam optimizer. The currently collected environmental wind speed and light intensity data, wind and solar power output data, and fracturing load data are used as inputs. The trained prediction model is then used to obtain the predicted wind power output data, predicted solar power output data, and fracturing load data for future fracturing operation cycles. The training process of the above prediction model is a well-known technique, and the specific process will not be described in detail here.

[0055] Furthermore, based on the start / stop instructions and planned duration of the current planned fracturing operation cycle, the fracturing load prediction data for future operation cycles is divided. Similar to the calculation method for the comprehensive similarity distance between different historical cycles, the comprehensive similarity distance between the predicted load data for future operation cycles and each cycle in each set of operation cycles is calculated. The average of all such comprehensive similarity distances is recorded as the comprehensive difference value between the future operation cycle and that set of operation cycles. The set of operation cycles corresponding to the minimum value among all such comprehensive difference values ​​is obtained; that is, the fracturing load data for the future operation cycle within the planned duration more closely matches the set of operation cycles. Based on the load fluctuation and operational characteristics under the given operating conditions, the predicted fracturing load data for future operating cycles is input into the demand response model of the fracturing load under the corresponding operating conditions for this set of operating cycles. This allows for the acquisition of the feasible range for the fracturing load to participate in the demand response for future operating cycles. It should be noted that the reason for performing the above matching process is that there are significant differences in the fluctuation characteristics, adjustability, and feasible range of fracturing load under different operating condition categories. Matching the predicted load to the most similar scenario allows for accurate invocation of the demand response model under the corresponding operating condition, improving the accuracy of the feasible range calculation and avoiding low control precision due to large deviations in the demand response analysis.

[0056] Furthermore, using the wind power forecast output, photovoltaic forecast output, fracturing load forecast data, the feasible interval of fracturing demand response corresponding to the currently predicted fracturing load data, the shiftable time window, the minimum continuous operating time, and the energy storage dynamic constraints output by Transformer-LSTM as model inputs, a multi-objective rolling time-domain optimization model is constructed with the optimization objectives of minimizing the comprehensive operating cost, energy storage aging cost, and wind and solar curtailment penalty cost. The objective function is: In the formula: For unit fuel and operation and maintenance costs, To address the dynamic constraints and aging penalty costs of energy storage, The model incorporates constraints to mitigate the cost of wind and solar power curtailment. These constraints include power balance constraints, upper and lower limits for unit output, fracturing load demand response constraints, energy storage dynamic constraints, and charging / discharging power constraints. A rolling time-domain optimization solution using the Gurobi solver is employed to output the optimal fracturing operation period, conventional unit output plan, energy storage charging / discharging plan, and microgrid power allocation instructions. Priority is given to scheduling fracturing loads during periods of peak wind and solar power output, achieving proactive source-load matching and peak shaving. Coordinated control is implemented based on microgrid power allocation instructions to reduce fluctuation impacts from a scheduling perspective. The specific process of using the Gurobi solver for multi-objective rolling time-domain optimization is well-known to those skilled in the art and will not be elaborated further.

[0057] A schematic diagram of the process for obtaining load change characteristic values ​​is shown below. Figure 2 As shown.

[0058] In summary, this application's embodiments collect power data and start / stop commands for fracturing loads during historical fracturing operation cycles; determine the comprehensive response value for each operation cycle based on power fluctuation characteristics and operation time for each time period within each operation cycle; the difference in comprehensive response values ​​between different operation cycles reflects the stage response variation differences of fracturing loads under complex operating conditions; calculate the overall load variation characteristics differences between different operation cycles, and combine them with stage response variation differences to conduct a comparative analysis of the stage change response of fracturing load data for complete cycles of different historical fracturing operations, thereby determining the demand response model of fracturing load under different operating conditions under complex operating conditions. This allows for the determination of an accurate feasible range for fracturing demand response during subsequent microgrid collaborative control based on predicted fracturing load data for different fracturing operation cycles. Furthermore, by combining this with an energy storage dynamic constraint model, microgrid collaborative optimization control is achieved, improving the control accuracy and operational stability of the oilfield microgrid. This avoids the problem of large errors in microgrid collaborative control caused by traditional oilfield microgrid control technologies that fail to adjust operating boundaries based on battery status and ignore the impact of fracturing load on control.

[0059] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0060] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0061] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for coordinated control of oilfield microgrids that takes into account both fracturing load and energy storage dynamic constraints, characterized in that: The method includes the following steps: Obtain power data and start / stop commands for fracturing load in each historical fracturing operation cycle; Based on the start and stop commands, obtain each operation time period in each operation cycle; analyze the power fluctuation and power distribution disorder of the fracturing load in each operation time period, construct the load change characteristic value of each operation time period, and determine the comprehensive response value of each operation cycle by combining the duration proportion of each operation time period in the operation cycle. Obtain the load change characteristic value of each work cycle; construct the comprehensive similarity distance between any two work cycles based on the difference in comprehensive response value and the difference in load change characteristic value between any two work cycles; combine with the clustering algorithm to cluster all historical work cycles to obtain the set of each group of work cycles; Demand response models for each set of work cycles are constructed using start / stop commands and planned durations for each work cycle. Fracturing load data for future work cycles is predicted using historical data. Based on the comprehensive similarity distance between future work cycles and historical work cycles, the demand response models that need to be input for fracturing load data in future work cycles are determined and input to obtain the feasible demand response intervals for future work cycles, which are then used for microgrid coordinated control.

2. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 1, characterized in that, The process for obtaining the load change characteristic values ​​is as follows: Calculate the range of fracturing load power data for each operation period to determine the amplitude of load fluctuation for each operation period; Calculate the coefficient of variation of all fracturing load power data for each operation period; The load change characteristic value for each working time period is determined based on the amplitude value and the coefficient of variation.

3. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 2, characterized in that, The load fluctuation amplitude for each working period is the ratio of the range to the preset rated maximum power value.

4. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 2, characterized in that, The characteristic value of the load change is the product of the amplitude value and the coefficient of variation.

5. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 1, characterized in that, The comprehensive response value is positively correlated with the load change characteristic value of each working time period and the duration ratio.

6. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 5, characterized in that, The process for obtaining the comprehensive response value is as follows: Calculate the product of the load change characteristic value for each work period and the duration percentage; the comprehensive response value for each work cycle is the sum of the products of all work periods in each work cycle.

7. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 1, characterized in that, The comprehensive similarity distance is positively correlated with the difference in the comprehensive response value and the difference in the characteristic value of the load change, respectively.

8. The oilfield microgrid coordinated control method that takes into account both fracturing load and energy storage dynamic constraints as described in claim 7, characterized in that, The comprehensive similarity distance is the weighted sum of the differences in the comprehensive response values ​​and the differences in the load change characteristic values.

9. The oilfield microgrid coordinated control method considering both fracturing load and energy storage dynamic constraints as described in claim 1, characterized in that, The process of determining the input demand response model for fracturing load data in future operation cycles based on the comprehensive similarity distance between future operation cycles and historical operation cycles is as follows: The comprehensive difference value between the future operation cycle and each set of historical operation cycles is determined based on the comprehensive similarity distance between the future operation cycle and each set of operation cycles. The demand response model of the set of operation cycles corresponding to the minimum comprehensive difference value is used as the demand response model that needs to be input into the fracturing load data of the future operation cycle.

10. The oilfield microgrid coordinated control method considering both fracturing load and energy storage dynamic constraints as described in claim 9, characterized in that, The comprehensive difference value is the average of the comprehensive similarity distances between the future work cycle and all work cycles in each set of work cycles.

Images