Well site mobile energy storage equipment, system and multi-time scale collaborative optimization scheduling method thereof

Abstract

The application discloses a kind of well site mobile energy storage equipment, system and its multi-time scale collaborative optimization scheduling method, belong to oilfield well site power supply and digital management and control technical field.System uses including battery cabin, power conversion cabin and control cabin The three-cabin structure of container type, and sets up power supply, communication, grounding and fire-fighting quick plug interface.Edge controller collects beam diagram of beam-pumping unit, well site load and energy storage state data, predicts load based on beam diagram physical model and long short-term memory network LSTM, identifies operating scene and calculates target urgency, executes hour-level charge-discharge plan, minute-level dynamic weighted rolling optimization, second-level stroke phase synchronous compensation, non-periodic impact load fast compensation and SOC safety check in turn;When multiple machines are connected in parallel, power distribution and SOC equalization are performed.System realizes local autonomy, remote monitoring and model updating through edge-cloud-end collaboration, and is suitable for single well, well group and temporary operation scene.

Description

Technical Field

[0001] This invention belongs to the field of oilfield well site power supply and digital management and control technology, specifically involving a mobile energy storage equipment and system for well sites and its multi-timescale collaborative optimization scheduling method. Background Technology

[0002] Currently, existing technologies for addressing the power supply problem at oilfield well sites mainly fall into the following categories: In existing technologies, one type of oilfield well site microgrid technology integrates photovoltaic energy storage, multi-machine drive, and intelligent well site control into a single cabinet to achieve coordinated operation of the well site's power generation, grid, load, and storage, thereby reducing the land area required. Another type of oilfield microgrid operation control method divides operating conditions and allocates power output according to the state of charge and net load power. Yet another method studies the determination of the access capacity for grid-connected oilfield microgrids with energy storage. In addition, there are well site energy dispatching methods based on game theory, as well as equipment solutions such as explosion-proof sodium-ion battery energy storage cabinets.

[0003] The aforementioned existing technologies involve well site microgrid equipment integration, energy storage capacity configuration, and energy dispatch, which can achieve a certain degree of coordinated operation of well site source, grid, load, and storage. However, the relevant solutions are mainly geared towards fixed deployment scenarios and still have shortcomings in terms of rapid transfer of energy storage equipment, rapid on-site access, and collaborative expansion of multiple mobile energy storage devices; their control strategies mainly adjust based on net load, grid power, or state of charge, and have not fully utilized the displacement, load, and stroke phase information in the pumping unit dynamometer diagram to implement predictive power compensation.

[0004] In existing technologies, mobile energy storage equipment is usually in the form of energy storage containers or mobile energy storage vehicles, with energy storage compartments and cable reel compartments, or a combination of mobile energy storage vehicles and deployment systems.

[0005] The aforementioned existing technologies mainly address the issues of transportation, deployment, and power output of mobile energy storage devices. However, their structures and control methods are primarily geared towards general power consumption scenarios and are not specifically adapted to the explosion-proof, corrosion-resistant, wide-temperature-range, and weak communication environments of oil well sites. Furthermore, they lack integrated designs for well site applications in terms of isolation and protection between the battery compartment, power conversion compartment, and control compartment, cross-compartment fault linkage, and rapid access to the well site power supply and distribution system.

[0006] In existing technologies, one type of flexible regulation strategy integrating source, grid, load, and storage in oilfield well sites can achieve coordinated control of source, grid, load, and storage; another type of dynamic access and multi-mode operation system method for autonomous mobile energy storage vehicles can achieve dynamic access of mobile energy storage.

[0007] Although the aforementioned existing technologies involve well site source-grid-load-storage coordinated control and dynamic access of mobile energy storage, their control strategies mainly rely on bus voltage, state of charge, or task commands for power regulation. They have not yet coupled the displacement, load, and stroke phase characteristics in the pumping unit dynamometer diagram with energy storage control, nor have they involved predicting pumping unit load changes based on the dynamometer diagram and implementing stroke phase synchronous compensation. Furthermore, they have not formed a collaborative control scheme that combines the state of charge balancing of multiple mobile energy storage devices with energy storage operating condition mapping.

[0008] In the existing technology, there are single-sided digital twin solutions for energy storage or oil pumping units, which are mainly used for energy storage equipment status monitoring, battery health assessment or oil pumping unit operating condition analysis.

[0009] The aforementioned existing technologies typically establish state models for a single object such as energy storage equipment or pumping unit, and do not adequately consider the correlation between the characteristics of the pumping unit's dynamometer diagram, the electrical load of the well site, and the operating status of energy storage. They also lack a collaborative mechanism to generate energy storage power demand based on changes in pumping unit operating conditions and to use the actual response results of energy storage to provide feedback correction to the load prediction and control model.

[0010] Existing technologies include energy storage or converter thermal management solutions that employ liquid cooling, air cooling, PCM phase change materials, etc.

[0011] The aforementioned existing technologies use methods such as air cooling, liquid cooling, or phase change materials to regulate the temperature of energy storage devices, which can meet the heat dissipation requirements in general scenarios. However, for well site environments where low-temperature start-up, normal operation heat dissipation, and short-term high-power charging and discharging thermal shock coexist, there is still room for improvement in the integration of thermal management components and mobile energy storage compartment structures, as well as in the coordinated control between preheating, air cooling, liquid cooling, and power-limited operation.

[0012] In summary, existing well site energy storage and mobile energy storage technologies have achieved a certain degree of power regulation, equipment mobility access, and operational monitoring. However, they still lack a coordinated solution that integrates the dynamometer characteristics of pumping units with mobile energy storage power control. This makes it difficult to predict periodic loads in advance based on changes in pumping unit displacement, load, and stroke phase, and to control the energy storage converter to implement phase-matching power compensation. Furthermore, there is still room for improvement in mobile energy storage equipment regarding rapid well site access, multi-unit coordinated operation, and the correlation between energy storage operating status and pumping unit operating conditions. Summary of the Invention

[0013] This invention provides a mobile energy storage equipment and system for well sites, as well as a multi-timescale collaborative optimization scheduling method. It predicts load change trends based on pumping unit displacement, load, and stroke phase, and implements impact power compensation, operation strategy optimization, state of charge adjustment, and power coordination among multiple energy storage devices at different time scales, thereby improving the well site load smoothing capability and energy storage utilization efficiency.

[0014] This invention is achieved through the following technical solution: A multi-timescale collaborative optimization scheduling method for a mobile energy storage system at a well site, the method comprising the following steps: Step S1: Collect pumping unit dynamometer data, well site load power data, energy storage system status data, time-of-use electricity price information, well site electrical equipment operation status data, and well site power supply and backup power status data. Perform time synchronization, abnormal data identification and processing, and feature extraction on the collected data to form a well site operation dataset. Step S2: Establish a load prediction model combining the physical model of the pumping unit's dynamometer diagram and a Long Short-Term Memory (LSTM) network. Input the physical load characteristics obtained from the physical model of the pumping unit's dynamometer diagram, as well as the pumping unit's displacement, load, stroke phase, and historical load time series data, into the LSTM network to obtain the pumping unit load prediction results. Based on the operating status, current variation characteristics, power variation trend, and start-up / shutdown information of aperiodic or impulsive loads, predict the start-up / shutdown status, power mutation time, and load variation trend of the aperiodic or impulsive loads. Superimpose the predicted power of each well site's electrical equipment at the same prediction time to form the overall predicted load curve for the well site. Step S3: Calculate the predicted average load power, predicted peak power, predicted load fluctuation variance and predicted load peak-to-average ratio based on the total predicted load curve of the well site. Combined with the well site power grid status, backup power status, energy storage charging status, demand control limits and critical load power supply status, identify the well site operation scenario and calculate the urgency parameters corresponding to the four objectives of economic operation, demand control, load impact mitigation and power supply reliability assurance. Step S4: Generate an hourly basic charge and discharge plan based on the load forecast results, well site operation scenario and target urgency parameters, and determine the bidirectional adjustment power margin and adjustment energy margin for second-level rapid compensation, as well as the energy storage backup power requirement and energy storage backup energy requirement undertaken by the energy storage system. Step S5: Based on the hourly basic charge and discharge plan, real-time well site load, latest load prediction results, energy storage system status, and the target urgency parameter most recently updated by step S3, calculate the dynamic weights corresponding to the four objectives, and perform dynamic weighted multi-objective rolling optimization within the minute-level feasible domain after reserving adjustment power and adjustment amount for second-level rapid compensation, to obtain minute-level energy storage power instructions; Step S6: Within a second-level control cycle, determine the stroke phase corresponding to the load peak and the stroke phase corresponding to the load trough based on the pumping unit indicator diagram data and real-time load power data within several consecutive strokes. Generate pumping unit stroke phase synchronous compensation power based on the real-time stroke phase of the pumping unit and the deviation of the real-time load power relative to the average load power of the corresponding pumping unit stroke. Generate rapid compensation power for aperiodic impact loads based on the predicted start-stop status of the non-periodic or impact loads, the predicted power change time, the load change trend, and the deviation of the real-time load power relative to the short-term average load power. The pumping unit stroke phase synchronization compensation power and non-periodic impact load rapid compensation power are superimposed, and the superposition result is limited within the bidirectional adjustment power margin. Then, it is superimposed with the minute-level energy storage power command to form the second-level energy storage target power. The lower limit of dynamic state of charge operation is determined based on the predicted peak-to-average load ratio. When the well site is in a power supply guarantee scenario and the risk of power failure of the critical load is 1, the fixed basic safety lower limit preset by the battery management system is determined as the current discharge state lower limit. In other cases, the lower limit of dynamic state of charge operation is determined as the current discharge state lower limit. Based on the discharge state lower limit, the fixed safety upper limit preset by the battery management system, the charging and discharging power boundary of the energy storage converter, and the equipment availability status, the second-level energy storage target power is safely verified to form the final energy storage power command. Step S7: When a single mobile energy storage device is running, the final energy storage power command is sent to the corresponding energy storage converter for execution; when multiple mobile energy storage devices are running in parallel, the final energy storage power command is used as the final total energy storage power command. Based on the state of charge, available charge and discharge power, battery temperature and equipment availability of each mobile energy storage device, basic power allocation and state of charge balancing adjustment are performed to form the individual charge and discharge power command of each mobile energy storage device, and the individual charge and discharge power command is sent to the corresponding energy storage converter for execution.

[0015] A digital management and control system for mobile energy storage at well sites based on edge-cloud collaboration is disclosed. The system utilizes the aforementioned mobile energy storage method based on dynamometer card coupling. The system includes end-side equipment, an edge control layer, a cloud management layer, and a communication network connecting the end-side equipment, the edge control layer, and the cloud management layer. The end-side equipment includes mobile energy storage equipment, a pumping unit dynamometer card acquisition device, and well site power supply and distribution equipment. The pumping unit indicator diagram acquisition device is used to acquire displacement, load, and stroke phase data of the pumping unit. The well site power supply and distribution equipment is used to provide data on well site load, power grid operation status, backup power status, and critical load power supply status. The mobile energy storage equipment is used to collect energy storage state of charge, energy storage charging and discharging power, battery temperature and equipment operating status, and to perform energy storage charging, energy storage discharging, load impact compensation and power regulation according to the control commands issued by the edge control layer. The edge control layer is used to perform time synchronization, abnormal data processing, feature extraction and local storage of the collected data, and to perform load prediction, well site operation scenario identification, hourly basic charge and discharge plan generation, minute-level dynamic weighted multi-objective rolling optimization, second-level stroke phase synchronization compensation of pumping unit, rapid compensation of non-periodic impact load and state of charge safety verification, and to perform single-unit power allocation and state of charge balancing control when multiple mobile energy storage devices are running in parallel. The cloud management layer includes a data management module, an energy storage operating condition digital twin module, a model training and parameter management module, a remote monitoring module, and a remote control module; The communication network is used to enable data interaction between the end-side equipment, the edge control layer, and the cloud management layer.

[0016] A mobile energy storage device for well sites includes a container body that can be transported and hoisted as a whole. The mobile energy storage device is connected as an end-side device to a digital control system for mobile energy storage in well sites based on edge-cloud collaboration, and performs charging, discharging and power regulation according to the energy storage power command generated by the multi-timescale collaborative optimization scheduling method of the mobile energy storage system in well sites. The container is equipped with a battery compartment, a power conversion compartment, and a control compartment that are isolated from each other. The battery compartment is equipped with battery clusters, a battery management system, a thermal management device, and a fire-fighting device. The power conversion compartment is equipped with a bidirectional energy storage converter, AC / DC power distribution equipment, and grid-connected / off-grid switching equipment. The control cabin is equipped with an energy management controller, an edge controller, a communication gateway, and a human-machine interaction device; The battery compartment, power conversion compartment and control compartment are connected by power lines, control lines and communication bus, and are equipped with fireproof isolation devices, pressure relief devices, fire-fighting linkage devices and fault isolation devices; The outer side of the container body is provided with one or more of the following interfaces: AC interface, DC interface, diesel generator interface, photovoltaic interface, well site bus interface, communication interface, grounding interface, and fire linkage quick-connect interface. The interface is connected to the power conversion compartment through a circuit breaker, disconnect switch, or contactor, and is equipped with anti-misinsertion device and electrical interlock device. The edge controller is used to perform load prediction, well site operation scenario identification, hourly basic charge and discharge plan generation, minute-level dynamic weighted rolling optimization, second-level stroke phase synchronization compensation and state of charge safety verification based on pumping unit dynamometer data, well site load power data and energy storage system status data, generate energy storage power commands, and control the operation of the bidirectional energy storage converter according to the energy storage power commands. The edge controller and / or energy management controller are also used to control the grid-connected switching device to complete the connection of the well site power supply and distribution system after completing grounding confirmation, communication confirmation, insulation detection, and voltage, frequency and phase sequence verification. When over-temperature, over-current, insulation abnormality, smoke or communication failure is detected, alarm, power limitation, fault branch disconnection or fault compartment isolation are performed according to the fault level.

[0017] The beneficial effects of this invention are: This invention employs an integrated three-compartment structure, isolating the battery compartment, power conversion compartment, and control compartment. It is equipped with quick-connect interfaces for AC, DC, diesel generator, photovoltaic, well site busbar, communication, grounding, and fire-fighting linkage. This allows mobile energy storage equipment to be transferred, configured, and quickly connected according to the load scale and power supply requirements of a single well, well group, or temporary operation scenario, reducing the dependence of fixed energy storage facilities on well site location and construction conditions. Simultaneously, through fireproof isolation, pressure relief, fire-fighting linkage, electrical interlocking, and fault isolation, as well as dustproof, waterproof, corrosion-resistant, explosion-proof, and wide-temperature-range thermal management measures, the safety and adaptability of mobile energy storage equipment in well site environments are improved.

[0018] This invention combines the displacement, load, stroke phase, and mechanical operating characteristics reflected in the pumping unit dynamometer diagram with historical load time-series data. It uses a physical model of the pumping unit dynamometer diagram and a Long Short-Term Memory (LSTM) network to jointly predict pumping unit load changes. Furthermore, it predicts sudden load changes based on the operating status, current variations, power trends, and start-up / shutdown information of water injection pumps, workover rigs, mud pumps, and other non-periodic or impact loads. This allows for the simultaneous utilization of the pumping unit's stroke operating mechanism and the real-time operating characteristics of well site electrical equipment, providing a more accurate predictive basis for operational scenario identification, peak value prediction, and energy storage scheduling.

[0019] This invention identifies high-frequency pulses, continuous production, demand-constrained conditions, power supply security, and complex operating scenarios based on the predicted load's average power, peak power, load fluctuation variance, and peak-to-average power ratio, combined with the grid, backup power supply, and critical load status. It automatically calculates the urgency and dynamic weights of four objectives: economic operation, demand control, load surge mitigation, and power supply reliability assurance. Furthermore, through hourly-level basic charge-discharge planning, minute-level dynamic weighted rolling optimization, and second-level rapid compensation, and by reserving bidirectional power margin, energy margin, and power reserve capacity according to the operating scenario, the energy storage operation strategy can automatically adjust with changes in well site conditions. This reduces the amount of electricity purchased from the grid during periods of high electricity prices while simultaneously addressing demand control, load fluctuation mitigation, and power supply security for critical loads.

[0020] This invention controls the energy storage equipment to perform discharge compensation during load rise and peak phases, and absorb power during load trough phases, based on the real-time stroke phase of the pumping unit, the stroke phase corresponding to load peak, the stroke phase corresponding to load trough, and the deviation between real-time load power and average load power. For non-periodic or impulsive loads without stable stroke phase characteristics, second-level rapid compensation is performed based on predicted start-up and shutdown states, predicted power change times, and real-time power deviation. By limiting the total second-level compensation power within a reserved bidirectional adjustment power margin, and combining it with the dynamic SOC operating lower limit, fixed safety upper limit, charge and discharge power boundaries, and equipment availability status for final safety verification, it can reduce instantaneous power fluctuations and peak-to-average power ratio on the well site grid side, while avoiding overcharging, over-discharging, or over-power operation of the energy storage equipment.

[0021] This invention allocates the final total energy storage power command based on the State of Charge (SOC), available charge / discharge power, battery temperature, and equipment availability of each mobile energy storage device. It then uses SOC balancing power adjustment to gradually bring the state of charge of different devices closer to uniformity. When a single device reaches its power boundary or SOC safety boundary, its power command is limited, and the remaining power demand is redistributed to other functioning devices. When a device malfunctions or resumes operation, the set of devices participating in coordinated operation is redefined, and SOC balancing control is initialized. This reduces the risk of some devices operating at excessive power for extended periods or reaching their operational boundaries prematurely, and improves the stability, equipment availability, and continuous adjustment capability of multi-device parallel operation.

[0022] This invention utilizes end-side equipment, an edge control layer, a cloud management layer, and a communication network to achieve well site operation data acquisition, local load prediction and scheduling control, remote status monitoring, digital twin analysis, fault backtracking, and model parameter updates. The edge control layer verifies the permissions, integrity, equipment status, and security constraints of the operating modes, scheduling parameters, model parameters, and remote control commands issued from the cloud. Even in the event of cloud communication anomalies or interruptions, it continues to independently execute real-time scheduling and safety control based on locally stored prediction models, control parameters, and the most recently valid configuration. This reduces the impact of weak communication or communication interruptions on well site operation, improving the system's operational status monitorability, fault handling capabilities, model adaptability, and remote operation and maintenance efficiency. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall architecture of the well site mobile energy storage system of the present invention.

[0024] Figure 2 This is a schematic diagram of the three-compartment integrated mobile energy storage equipment of the present invention.

[0025] Figure 3 A schematic diagram illustrating the rapid access and electrical connection between the mobile energy storage equipment and the well site power supply and distribution system of this invention.

[0026] Figure 4 This is a flowchart of the multi-time-scale collaborative optimization scheduling method based on dynamometer diagram coupling of the present invention.

[0027] Figure 5 This is a schematic diagram illustrating the principle of pumping unit stroke phase synchronization compensation and SOC safety constraint in this invention.

[0028] Figure 6 This is a schematic diagram of the coordinated control and SOC balancing of multiple mobile energy storage devices according to the present invention.

[0029] Figure 7 This is a schematic diagram of the edge-cloud-device collaborative digital management and control and model feedback update architecture of the present invention. Detailed Implementation

[0030] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0031] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0032] It should also be understood that the terminology used in this application specification is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this application specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0033] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0034] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0035] Implementation Method 1 This implementation provides a multi-timescale collaborative optimization scheduling method for mobile energy storage at well sites. It predicts well site load changes based on the characteristics of the pumping unit dynamometer diagram and sequentially performs hourly economic scheduling, minute-level rolling optimization, second-level impact compensation, and power allocation for multiple mobile energy storage devices, including the following steps.

[0036] Step S1: Multi-source operating condition data and historical load acquisition: The edge gateway collects pumping unit dynamometer data, well site load power data, energy storage system status data, time-of-use electricity price information, well site electrical equipment operating status data, and well site power supply and backup power status data at a preset sampling frequency via industrial communication buses such as RS485 and Ethernet.

[0037] The pumping unit dynamometer data includes pumping unit displacement, load, and stroke phase; the well site load power data includes real-time and historical power at each node of the well site; the energy storage system status data includes state of charge, charging and discharging power, battery temperature, and equipment availability; the well site electrical equipment operating status data includes current, power, and start / stop status of pumping units, water injection pumps, workover rigs, and other well site electrical equipment. The well site power supply and backup power status data includes grid voltage, grid frequency, grid connection / disconnection status, start / stop status and current available output of diesel generators and other backup power supplies, as well as power and power supply status of critical production loads, communication loads, and safety loads.

[0038] The collected data is synchronized in time, anomaly data is removed, and features are extracted to form a well site operation dataset for load forecasting, scheduling optimization, and energy storage control.

[0039] Step S2: Load forecasting based on a combination of physical model and LSTM: A load prediction model was established by combining the physical model of the pumping unit's indicator diagram with a long short-term memory (LSTM) network.

[0040] The physical model of the pumping unit's indicator diagram is represented as follows:

[0041] in, For the first The pumping unit of the well is The load at any given moment; For static loads, The displacement-load linearity coefficient, For the displacement of the oil pumping unit, The amplitude of the periodic dynamic load. The load fluctuation frequency corresponding to the stroke cycle of the pumping unit. This is the initial phase.

[0042] By identifying and fitting the parameters of the physical model of the dynamometer diagram, physical load characteristics reflecting the mechanical operating state of the pumping unit are obtained. These physical load characteristics, along with measured displacement, load, stroke phase, and historical load time-series data of the pumping unit, are input into an LSTM model to predict the load change trend within a preset time domain.

[0043] in, For the first The load power of the well at the predicted time; For the trained LSTM model, These are model parameters; It is a physical feature vector formed by the physical model of the dynamometer diagram and the measured dynamometer diagram data; This is a historical time-series vector containing displacement, load, stroke phase, and load power. For water injection pumps, workover rigs, mud pumps, and other non-periodic or impact loads, the load status is identified based on their operating conditions, current variation characteristics, power variation trends, and start-stop information, and their start-stop status, power abrupt change times, and load variation trends are predicted.

[0044] Based on the prediction results of the pumping unit and other electrical equipment at the well site, the predicted power of each electrical device at the same prediction time is superimposed to form the total predicted load curve of the well site, and the predicted peak, average load and the expected time of peak occurrence are obtained, providing input for subsequent operation scenario identification and multi-time scale scheduling; among them, the predicted start and stop status of non-periodic or impact load, the predicted power change time and load change trend are used as the prediction basis for step S6 to determine whether to perform second-level fast compensation.

[0045] Step S3: Well site operation scenario prediction and identification and scheduling target urgency calculation: The edge controller calculates the average power, peak power, load fluctuation variance and peak-to-average power ratio based on the load prediction sequence output in step S2, and predicts and identifies the well site operation scenario in conjunction with the power grid, backup power supply and key load status; on this basis, it automatically calculates the urgency parameters corresponding to the four objectives of economic operation, demand control, load shock smoothing and power supply reliability assurance.

[0046] 1. Calculation of predicted load characteristics Let the predicted load sequence output in step S2 be:

[0047] in, To predict the number of prediction data points included in the time domain, To predict the first in the time domain The predicted load power at each time point.

[0048] The average load power in the predicted time domain is:

[0049] The maximum instantaneous peak power in the predicted time domain is:

[0050] The predicted load fluctuation variance is:

[0051] The predicted peak-to-average load ratio (PAR) is calculated using the following formula:

[0052] When the average load power in the prediction time domain To avoid the denominator of the peak-to-average ratio calculation being 0, let Furthermore, it does not trigger the high-frequency pulse scenario determination based on peak-to-average power ratio; demand-limited scenarios and power supply guarantee scenarios are still identified according to the corresponding criteria.

[0053] 2. Well Site Operation Scenarios Prediction and Recognition The edge controller predicts and identifies future operating scenarios in the forecast time domain based on the peak-to-average ratio of the predicted load, the variance of load fluctuations, the characteristics of periodic changes and instantaneous changes, and in combination with the well site power grid, backup power supply and critical load status.

[0054] (1) High-frequency pulse scenario When the predicted load meets:

[0055] Furthermore, the operational scenario in the future prediction time domain will be identified as a high-frequency pulse scenario if at least one of the following conditions is met simultaneously: The predicted load fluctuation variance exceeds the upper limit of the variance reference for continuous production, i.e.:

[0056] The predicted load sequence exhibits recurring peak-valley variations corresponding to the pumping unit stroke cycle; for water injection pumps, workover rigs, mud pumps, and other non-periodic or impact loads, when the load power change between adjacent prediction times exceeds a preset abrupt change threshold, it is not affected. Due to limitations, the operating scenario in the future prediction time domain is directly identified as a high-frequency pulse scenario.

[0057] in, The peak-to-average power ratio (PAPR) threshold is used to determine the performance of high-frequency pulse scenarios. This serves as a reference upper limit for the allowable load fluctuation variance under continuous production conditions. Both are pre-calibrated based on historical stable operation data from the well site and field trial operation data, and stored in the edge controller.

[0058] (2) Continuous production scenario When the predicted load simultaneously meets:

[0059] as well as:

[0060] Furthermore, when the predicted load sequence does not have obvious periodic shocks, instantaneous power surges, grid anomalies, or critical load power supply risks, the operating scenarios in the future prediction time domain will be identified as continuous production scenarios.

[0061] The continuous production scenario indicates that the overall load on the well site will be relatively stable in the future, and the energy storage system will mainly operate according to the hourly economic scheduling plan and the minute-level rolling optimization results.

[0062] (3) Demand-limited scenarios When the predicted peak power reaches or exceeds the well site demand warning value:

[0063] The operating status within the future prediction time domain is identified as a demand-constrained scenario.

[0064] in, The well site demand warning value is set in advance according to the well site contract demand, power supply and distribution capacity, or maximum allowable power.

[0065] (4) Power supply guarantee scenario The current operating state will be identified as a power supply protection scenario when any of the following conditions are detected: The grid voltage or frequency has reached the warning limit; The power grid is experiencing a power outage or there is a significant risk of power loss. The well site has entered islanded operation or there is a need for islanded switching. The available backup power capacity is insufficient; The available backup capacity of energy storage is lower than the preset safety value; There is a risk of power outage for critical production loads, communication loads, or safety loads.

[0066] In scenarios where power supply security is guaranteed, control priority is higher than economic operation requirements. Subsequent dispatching prioritizes ensuring power supply to critical loads and necessary energy storage backup capacity.

[0067] (5) Composite operation scenarios When two or more scenario determination conditions are met simultaneously, the well site operation status is identified as a composite operation scenario.

[0068] For example, when the predicted load exhibits high-frequency pulse characteristics and is close to the demand warning value, it is identified as a high-frequency pulse-demand-limited composite scenario; when the predicted load exhibits high-frequency pulse characteristics and the power grid is operating abnormally, it is identified as a high-frequency pulse-power supply guarantee composite scenario.

[0069] 3. Automatic calculation of target urgency parameters To ensure that operational indicators of different units and orders of magnitude can be uniformly used for dynamic weight calculation, a limiting function is defined:

[0070] Among them, when hour, ;when hour, ;when hour, .

[0071] The edge controller automatically calculates the urgency of four objectives: economic operation, demand control, surge mitigation, and power supply reliability assurance, forming an objective urgency vector.

[0072] And satisfy:

[0073] The closer the urgency level is to 1, the more important the corresponding target is within the current scheduling cycle.

[0074] (1) Urgency of economic operation

[0075] The edge controller calculates the urgency of charging at low prices and the urgency of discharging at high prices based on the current time-of-use electricity price and the available charging / discharging capacity of the energy storage. When At that time, without calculating the urgency of low-cost charging, let ;when At that time, without calculating the urgency fraction of high-valence discharge, let .when When, it means there is no peak-valley price difference within the current time-of-use electricity pricing cycle, let .

[0076] The urgency of low-cost charging is:

[0077] The urgency of high-priced discharge is:

[0078] The urgency of economic operation is as follows:

[0079] in, The electricity price for the current period; , and These are the lowest, highest, and average electricity prices within the current time-of-use pricing cycle; This represents the current state of charge of the energy storage. and These are the fixed upper security limit and fixed basic security lower limit preset by the BMS, respectively.

[0080] The electricity price reference value is read directly from the time-of-use electricity price table, and the SOC boundary is read directly from the BMS, without being filled in by the operator on the spot.

[0081] This step uses a fixed basic safety lower limit. The dynamic SOC lower limit calculated in step S6 is not used.

[0082] (2) Urgency of demand control

[0083] The urgency of demand control is automatically calculated using the following formula:

[0084] in, For the well site contract demand, power supply and distribution capacity, or the pre-set maximum allowable power; This is the early warning value for well site demand.

[0085] The demand warning value is determined according to the following formula:

[0086] in, The demand warning coefficient is preset based on the well site demand safety margin, and its value ranges from 0.80 to 0.90.

[0087] When the predicted peak demand does not exceed the demand warning value When the predicted peak power reaches or exceeds the maximum allowable power, .

[0088] (3) Urgency of load shock mitigation

[0089] The urgency of load shock mitigation is determined by both the predicted peak-to-average load ratio and the load fluctuation variance.

[0090] in, This serves as a reference for the peak-to-average power ratio (PAR) corresponding to the high-frequency pulse state. This serves as a high-value reference for the load fluctuation variance corresponding to high-frequency pulse states; Let be the combined coefficient of peak-to-average power ratio and load fluctuation variance, and satisfy:

[0091] and Obtained based on statistical analysis of historical valid data under continuous production conditions; and Obtained based on confirmed high-frequency pulse operation data or field impact test data; Pre-calibrate based on historical control effects or on-site trial operation results.

[0092] To ensure that the denominator is positive in the calculation of the load peak-to-average ratio and load fluctuation variance normalization, the high-value reference of the high-frequency pulse state satisfies:

[0093] When water injection pumps, workover rigs, mud pumps, and other non-periodic or impact loads are directly identified as high-frequency pulse scenarios due to load power changes exceeding a preset abrupt change threshold between adjacent predicted times, the urgency of load impact mitigation is adjusted. =1; otherwise, calculate according to the aforementioned formula combining peak-to-average power ratio and load fluctuation variance. .

[0094] The above reference values ​​are all pre-stored in the edge controller and are automatically called by the system during operation.

[0095] (4) Urgency of ensuring power supply reliability

[0096] The urgency of ensuring power supply reliability is automatically determined based on voltage risk, frequency risk, insufficient reserve capacity risk, and risk of power failure of critical loads.

[0097] in, Risk of voltage exceeding limits; Risk of exceeding frequency limits; Risk of insufficient backup capacity; Risk of power loss for critical loads.

[0098] Voltage over-limit risk is calculated by normalizing the deviation of the grid voltage from the rated voltage, as well as the voltage warning limit and protection limit; frequency over-limit risk is calculated by normalizing the deviation of the current frequency from the rated frequency, as well as the frequency warning limit and protection limit.

[0099] Both the voltage and frequency over-limit risks are piecewise linearly normalized using the aforementioned limiting function C(x): when the corresponding voltage or frequency deviation does not reach the warning limit, the risk value is 0; when it reaches or exceeds the protection limit, the risk value is 1; when it is between the warning limit and the protection limit, it is linearly mapped to 0 to 1 according to the position of the actual deviation between the warning limit and the protection limit.

[0100] The risk of insufficient reserve capacity is calculated based on the difference between the total system reserve power requirement for critical loads during the current scheduling cycle and the total reserve power available from various backup power sources at the well site.

[0101] in, The total system reserve power required to ensure critical production loads, communication loads, and security loads during the current scheduling cycle; The total standby power currently available for energy storage, diesel generators, and other backup power sources.

[0102] When the total backup power provided by various backup power sources at the well site reaches or exceeds the total backup power requirement of the system for critical loads, the risk of insufficient backup capacity is 0; as the total backup power deficit increases, the risk of insufficient backup capacity increases accordingly.

[0103] When the total system reserve power requirement for critical loads in the current scheduling cycle is 0, it means there is no reserve power requirement in this cycle, and the risk of insufficient reserve capacity is 0.

[0104] The risk of power failure for critical loads is determined based on the real-time power supply status of critical production loads, communication loads, and safety loads.

[0105] The above reference limits are predetermined based on the well site power supply and distribution design parameters, protection settings, and critical load capacity, and are not selected by operators on an ad hoc basis.

[0106] Step S4: Generation of hourly economic scheduling and basic charging and discharging plan: Based on the future load forecast results obtained in step S2, the well site operation scenario and target urgency vector output in step S3, and combined with time-of-use electricity price, current state of charge of energy storage, available energy storage capacity and well site power supply reliability requirements, the edge controller automatically determines the charging, discharging or standby status of each hourly scheduling period within the future preset scheduling cycle, calculates the corresponding basic charging and discharging power and target state of charge, and forms an hourly basic charging and discharging plan.

[0107] During off-peak electricity price periods, the edge controller automatically generates the base power for energy storage charging; during peak electricity price periods, it automatically generates the base power for energy storage discharging to reduce the amount of electricity purchased from the grid during high-price periods. When the predicted peak load approaches the well site demand control limit, the base discharge power is increased during the corresponding period.

[0108] When generating an hourly basic charge and discharge plan, the edge controller determines the corresponding energy storage regulation margin and power supply backup requirements based on the operating scenario identified in step S3.

[0109] In high-frequency pulse scenarios, a bidirectional adjustable power margin is reserved for second-level impact compensation in step S6. and adjust battery margin The bidirectional power margin is used to limit the occupation of the maximum charge and discharge power of energy storage in step S5, and the power margin is used to limit the occupation of the available range of energy storage SOC in step S5. In demand-constrained scenarios, by increasing the target state of charge before the predicted peak period and setting the corresponding hourly basic discharge power during the predicted peak period, the discharge capacity required for the predicted peak period is reserved. In power supply assurance scenarios, based on the critical load power, preset assurance duration, and available output of diesel generators and other backup power sources, and after deducting the backup power capacity that diesel generators and other non-energy storage backup power sources can provide, the energy storage backup power requirement for the current time period is determined to be borne by the energy storage system. and energy storage backup power requirements .in, and These are all backup requirements for the energy storage system and do not represent the total backup demand of the system shared by all backup power sources at the well site. The energy storage backup power requirement and energy storage backup quantity requirement are both not less than 0; when the backup power supply capacity that diesel generators and other non-energy storage backup power sources can undertake has already covered the backup demand of critical loads, the corresponding energy storage backup power requirement or energy storage backup quantity requirement is taken as 0.

[0110] In complex operating scenarios, the regulation margin and power reserve requirements for each scenario are determined simultaneously. When the current available power or available electricity of the energy storage is insufficient to meet both simultaneously, the reserve power and reserve electricity required for critical load power supply are reserved first, in accordance with the principle of prioritizing power supply security. The bidirectional regulation power margin is then adjusted accordingly based on the remaining available power and available electricity. and adjust battery margin The bidirectional adjustable power margin The adjustment capacity margin is determined based on the difference between the predicted peak power and the average load power, the expected number of loads participating in compensation simultaneously, and the currently available charging and discharging power of the energy storage. The bidirectional adjustment power margin, the expected compensation duration, and the current available energy storage capacity are determined based on the bidirectional adjustment power margin, the expected compensation duration, and the current available energy storage capacity. To avoid the energy storage charge / discharge power boundary and the SOC operating boundary intersecting after deducting the adjustment margin in step S5, the bidirectional adjustment power margin and the adjustment capacity margin satisfy the following:

[0111]

[0112] This avoids the intersection of the energy storage charge / discharge power boundary or the SOC operating boundary after deducting the adjustment margin.

[0113] The edge controller will include the hourly basic charge / discharge plan, basic charge / discharge power, target state of charge, and the bidirectional adjustment power margin required for step S6. and adjust battery margin The current energy storage backup power requirement and energy storage backup power requirement are sent to step S5 as the basic instruction for minute-level rolling optimization. The hour-level basic power is not directly used as the final execution instruction. After minute-level correction in step S5 and second-level compensation and safety verification in step S6, the final energy storage power instruction is formed. When multiple devices are running in parallel, the power allocation for each device is performed in step S7.

[0114] Step S5: Dynamically weighted multi-objective rolling optimization at the minute level: Based on the hourly basic charge and discharge plan generated in step S4, and combined with the latest load forecast results, real-time well site load, energy storage system status, and target urgency parameters output in step S3, the hourly basic charge and discharge power is rolled over within a minute-level control cycle.

[0115] Step S5 directly calls the urgency of the four objectives—economic operation, demand control, load surge mitigation, and power supply reliability assurance—calculated in step S3, without repeating scenario identification and urgency calculation. Step S5 evaluates the operational performance of the current minute-level energy storage power through the four objective functions and determines the current minute-level energy storage charging and discharging power by combining dynamic weights.

[0116] Regulation Indicates energy storage and discharge. This indicates energy storage charging.

[0117] 1. Dynamic weight calculation The dynamic weights corresponding to the four objectives are calculated according to the following formula:

[0118] in, The first one calculated in step S3 Project urgency For the first The basic weight of the project target, These are the corresponding weight adjustment coefficients. The basic weights and weight adjustment coefficients are pre-calibrated and stored in the edge controller based on historical scheduling results or field trial operation results, and are automatically called by the system during operation. , And at least one basic weight is greater than 0. Each dynamic weight satisfies:

[0119] 2. Minute-level economic operation targets

[0120] The minute-level economic operation target is used to limit the excessive deviation of minute-level energy storage power from the hour-level base charge and discharge power in step S4, so that minute-level adjustments maintain the overall trend of hour-level economic dispatch and reduce unnecessary additional charge and discharge adjustments.

[0121] Its expression is:

[0122] in, This represents the objective function value for economic operation at the minute level. The current minute-level energy storage charging and discharging power to be determined in step S5; The hourly baseline charge / discharge power for the current time period determined in step S4; For the normalization reference value of energy storage power, the maximum allowable charge and discharge power of mobile energy storage equipment can be used.

[0123] when:

[0124] hour:

[0125] 3. Load demand control target

[0126] First, calculate the corresponding grid-side power of the well site after the current minute-level energy storage power adjustment:

[0127] in, This represents the grid-side power at the well site after energy storage regulation. This represents the current total load power at the well site; The load demand control target is expressed as:

[0128] in, The objective function value for load demand control; The contracted power demand, power supply capacity, or maximum allowable power for the well site shall be specified, and the following conditions shall be met:

[0129] when:

[0130] hour:

[0131] This means that the current energy storage capacity is already sufficient to keep the grid-side power below the demand warning value.

[0132] when:

[0133] At that time, as the grid-side power approaches As the demand control objective function value increases, minute-level control tends to increase the energy storage discharge power to reduce peak loads.

[0134] 4. Load shock mitigation target

[0135] The load shock mitigation target is used to limit drastic changes in grid-side power between adjacent minute-level control cycles, enabling the energy storage system to smoothly adjust to minute-scale load changes.

[0136] Its expression is:

[0137] in, The objective function value for mitigating minute-level load shocks; This represents the grid-side power corresponding to the current minute-level control cycle; The control cycle is on the order of minutes. This is a normalized reference value for minute-level power variation, pre-determined based on historical stable operating data from the well site or the allowable minute-level power variation range. During the first minute-level control cycle after system startup, Take the most recent valid power value of the well site grid side collected and stored in step S1.

[0138] The smaller the power change between adjacent minutes:

[0139] The more drastic the power change:

[0140] This allows the energy storage system to actively adjust its charging and discharging, reducing power fluctuations at the well site on a minute-by-minute scale.

[0141] 5. Power supply reliability assurance target

[0142] The power supply reliability assurance target is used to evaluate whether the remaining reserve power and reserve capacity after minute-level energy storage regulation meet the power supply reserve requirements determined in step S4, preventing excessive use of the energy storage capacity reserved for critical load power supply and abnormal operating conditions by minute-level energy storage regulation. First, the upward adjustment power margin that can be further provided under the current minute-level energy storage power is calculated:

[0143] in, After deducting the bidirectional adjustment power margin reserved in step S4 for step S6, the current operating point has a power margin that can be further used for critical load power supply backup in the discharge direction. This represents the maximum allowable discharge power under the current energy storage state. Based on the current SOC, current minute-level energy storage power, minute-level control cycle, rated energy storage capacity, and charge / discharge efficiency, the expected state of charge after executing the current minute-level power is calculated and denoted as:

[0144] When the stored energy is discharged, that is hour:

[0145] When energy storage is charged, that is hour:

[0146] in, The current state of charge; To predict the state of charge one control cycle after executing the current minute-level power; For minute-level control cycles expressed in hours, for example, a minute-level control cycle of 1 minute. ; This represents the rated available capacity of the energy storage. For discharge efficiency; For charging efficiency.

[0147] Based on the reserve power requirement determined in step S4, calculate the energy storage capacity currently expected to be available for backup power supply:

[0148] in, The energy storage capacity currently expected to be available for backup power supply of critical loads is after deducting the adjustment capacity reserved for step S6 in step S4. A fixed basic safety lower limit preset for BMS.

[0149] The power supply reliability guarantee target is expressed as:

[0150] in, To ensure the objective function value for power supply reliability; The current time period reserve power requirement is determined in step S4; The current time period reserve power requirement is determined in step S4; The current available upward adjustment power margin; This represents the amount of energy currently expected to be available for backup power supply.

[0151] When the current time period energy storage backup power requirement determined in step S4 is 0, the corresponding backup power insufficiency evaluation item in the power supply reliability guarantee target is set to 0; when the energy storage backup power requirement is 0, the corresponding backup power insufficiency evaluation item is set to 0, so as to avoid the denominator of the normalization calculation being 0.

[0152] 6. Comprehensive Optimization Objectives After dimensionless processing of the above four objectives, a minute-level comprehensive objective function is established:

[0153] Step S5, based on minimizing the comprehensive objective function, solves for the current minute-level energy storage charge / discharge power within the minute-level feasible region formed by the energy storage charge / discharge power boundary and the SOC operating boundary:

[0154] in, The candidate minute-level energy storage charging and discharging power is defined as follows: ΩS5(t) represents the feasible region corresponding to the current minute-level control cycle, which is jointly determined by the energy storage charging and discharging power boundary after deducting the bidirectional adjustment power margin determined in step S4, and the SOC operating boundary after deducting the adjustment power margin; the energy storage reserve power requirement and energy storage reserve power requirement determined in step S4 are still based on the power supply reliability guarantee target. To conduct an evaluation; Indicates candidate power Substituting into the aforementioned comprehensive objective function The objective function value is then obtained.

[0155] Therefore, by comprehensively coordinating four objectives—hourly-level basic plan tracking, demand control, minute-level load fluctuation smoothing, and maintaining power supply reserve capacity—the current minute-level energy storage charging and discharging power is determined. .

[0156] 7. Energy storage operation constraints The minute-level energy storage power determined in step S5 satisfies:

[0157] Simultaneously satisfy:

[0158] in, and These are the current maximum allowed charging power and maximum discharging power, respectively; The bidirectional adjustable power margin determined in step S4 for second-level impact compensation in step S6; and These are the fixed basic security lower limit and fixed security upper limit preset by the BMS, respectively; The adjustment power margin determined in step S4 for the second-level impact compensation in step S6. The rated available capacity for energy storage. Through the above power boundary and SOC operating boundary constraints, step S5 retains the adjustment capability required by step S6 in the charging direction, discharging direction, and both sides of the upper and lower limits of SOC, so as to avoid minute-level adjustment occupying the bidirectional adjustment power margin and adjustment capacity margin reserved in step S4.

[0159] 8. Minute-level rolling updates and output At the start of each minute-level control cycle, the edge controller reacquires: the latest well site load; the latest load forecast results; the current energy storage SOC; the current equipment status; and the latest target urgency output from step S3.

[0160] Recalculate the dynamic weights and the four objective function values, and determine the current minute-level energy storage charging and discharging power. ; After entering the next minute-level control cycle, the latest operating data is used to recalculate, thus forming a minute-level rolling correction.

[0161] The minute-level energy storage power output in step S5 Instead of being directly used as the final execution power of the energy storage converter, it is sent to step S6 as the basic power command for second-level stroke phase synchronization compensation and non-periodic impact load rapid compensation.

[0162] Step S6: Second-level rapid compensation and SOC safety constraint: Based on the minute-level energy storage power command output in step S5, within the second-level control cycle, the compensation power and compensation time of the energy storage converter are rapidly adjusted according to the real-time stroke phase of the pumping unit, the real-time load power and its deviation from the average load. This ensures that the energy storage discharge compensation matches the pumping unit load rise and peak phases, and absorbs power according to the energy storage operating status when the load drops or falls below the average load, thereby reducing the well site power fluctuations caused by the cyclic operation of the pumping unit.

[0163] Regarding the first Based on the dynamometer data and real-time load power data over several consecutive strokes, the first well is determined. Peak load of wellhead corresponds to stroke phase Stroke phase corresponding to load trough The edge controller operates based on the real-time stroke phase. Determine the current stroke stage of the pumping unit and predict the peak load arrival time to keep the energy storage compensation command synchronized with the pumping unit stroke cycle.

[0164] To clarify the correspondence between real-time stroke phase and peak and trough load phases, the first... The stroke phase of the wellhead is uniformly represented in the interval Within this context, the annular phase deviation between the real-time stroke phase and the stroke phase corresponding to the load peak and the stroke phase corresponding to the load trough are calculated respectively:

[0165]

[0166] in, The annular phase deviation between the real-time stroke phase and the stroke phase corresponding to the load peak; The annular phase deviation between the real-time stroke phase and the stroke phase corresponding to the load trough; For the first The preset phase window for each well is determined in advance based on the well's historical dynamometer data, stroke cycle, and on-site compensation effect, and meets the following requirements: .

[0167] No. The second-level energy storage compensation power corresponding to each well is determined according to the following formula:

[0168] in, In order to target the The second-level energy storage compensation power generated by the wellhead; The impact mitigation coefficient ranges from 0.5 to 0.9. For the first Koujing at all times Real-time load power; The average load power is calculated based on the current stroke or the most recent complete strokes; and These are the permissions allowed for the current running state. The maximum discharge power and maximum charging power of the wellhead impact compensation.

[0169] When the real-time stroke phase enters the phase window corresponding to the load peak and the real-time load power is higher than the average load power, the energy storage converter performs discharge compensation. When the real-time stroke phase enters the phase window corresponding to the load trough and the real-time load power is lower than the average load power, the energy storage converter absorbs power under the condition of satisfying the SOC and charging power boundaries. In other stroke stages, the second-level compensation power for this well is zero. Among them, the stroke phase and the preset phase window are used to determine the trigger time of the compensation power, and the deviation between the real-time load power and the average load power is used to determine the direction and amplitude of the compensation power, thereby achieving synchronous matching between the energy storage compensation power and the peak and trough stages of the pumping unit stroke load.

[0170] For water injection pumps, workover rigs, mud pumps, and other non-periodic or impact loads that do not have stable stroke phase characteristics, stroke phase window judgment is not performed. Step S6 determines whether to trigger second-level rapid compensation based on the predicted start-stop status, predicted power change time, and load change trend output in step S2, combined with the real-time power change of the load.

[0171] When step S2 predicts that the load will start, stop, or experience a significant power surge, or when step S6 detects that the absolute value of the deviation between the real-time load power and the short-time average load power exceeds a preset power surge threshold, a second-level rapid compensation power is generated for the r-th non-periodic or impulsive load.

[0172] in, In order to target the A second-level rapid compensation power generated by a non-periodic or impulsive load; For rapid compensation coefficient; For the first The load at time Real-time load power; This is the short-time average load power calculated based on the real-time load power within the most recent preset second-level time window; and These are the maximum discharge power and maximum charging power allowed for rapid load compensation under the current operating conditions, respectively. The rapid compensation coefficient... The preset power surge threshold is pre-calibrated based on the historical start-stop data, power surge test data or field trial operation results of the corresponding load, and stored in the edge controller.

[0173] When the real-time load power is higher than the short-time average load power, the compensation power is positive, and the energy storage converter performs discharge compensation; when the real-time load power is lower than the short-time average load power, the compensation power is negative, and the energy storage converter absorbs power under the condition of satisfying the SOC and charging power boundaries; when step S2 does not predict the start-up, stop, or significant power change of the load, and the absolute value of the deviation between the real-time load power and the short-time average load power detected in step S6 does not exceed the preset power change threshold, the second-level rapid compensation power for the load is set to 0.

[0174] For well sites with multiple pumping units and non-periodic or shock loads, the stroke phase synchronous compensation power corresponding to each pumping unit is superimposed with the second-level rapid compensation power corresponding to each non-periodic or shock load:

[0175] in, The number of pumping units participating in second-level impact compensation. The number of non-periodic or impact loads participating in second-level rapid compensation. The bidirectional adjustment power margin is determined in step S4 and reserved in step S5. This limiting ensures that the total second-level compensation power does not exceed the bidirectional adjustment power margin reserved in step S4.

[0176] The second-level compensation power is superimposed on the minute-level energy storage power command output in step S5 to obtain the second-level energy storage target power:

[0177] in, This refers to the minute-level energy storage power command output in step S5. The target energy storage power is determined by superimposing phase synchronization compensation at the second-level stroke and rapid compensation for non-periodic impact loads. A positive value for energy storage power indicates discharge, and a negative value indicates charging.

[0178] Dynamic SOC operating lower limit To prevent excessive consumption of energy storage capacity due to continuous load surges and to reserve necessary available power for subsequent periodic load surges and sudden load changes, step S6 uses the predicted peak-to-average load ratio calculated in step S3. Dynamically adjust the lower limit of the SOC operation of the energy storage system:

[0179] in, This is the dynamically adjusted lower limit for SOC operation. The highest allowed operating limit; A fixed basic safety lower limit preset for BMS; Impact sensitivity coefficient; This represents the peak-to-average load ratio within the current forecast time domain. The peak-to-average ratio (PAPR) reference value that triggers the adjustment of the lower limit of SOC operation, and satisfies:

[0180] when:

[0181] hour:

[0182] That is, the lower limit of SOC operation is maintained at the fixed basic safety lower limit of BMS.

[0183] when:

[0184] At the same time, as the load impact increases, the lower limit of dynamic SOC operation is raised accordingly, thereby reserving more available power for subsequent continuous impact compensation and sudden load changes.

[0185] Final security check Step S6 performs a safety verification of the second-level energy storage target power based on the dynamic SOC operating lower limit, the BMS fixed safety upper limit, the maximum charging and discharging power of the energy storage converter, and the equipment availability status. This is especially important when the well site is in a power supply guarantee scenario and there is a risk of power loss to critical loads. In order to prioritize power supply to critical production loads, communication loads, and safety loads, the discharge SOC lower limit used in the final safety verification in step S6 is switched to the fixed basic safety lower limit preset by the BMS. In other cases, the dynamic SOC lower limit will continue to be used. .

[0186] Right now:

[0187] First, limit the charging and discharging power:

[0188] in, The second-level energy storage power after the charging and discharging power is limited by the energy storage converter; and These represent the maximum charging power and maximum discharging power allowed under the current device conditions, respectively.

[0189] The final energy storage power command is determined according to the following formula:

[0190] in, A fixed security limit preset for the BMS.

[0191] When the actual SOC reaches or falls below the currently adopted lower limit of discharge SOC, the energy storage is restricted or stopped from continuing to discharge, but charging is allowed if other safety conditions are met; when the actual SOC reaches or exceeds the fixed upper limit of safety, the energy storage is restricted or stopped from continuing to charge, but discharging is allowed if other safety conditions are met.

[0192] This results in a final energy storage power command that satisfies SOC, power boundary, and device safety constraints.

[0193] When a single mobile energy storage device is in operation, the final energy storage power command is directly sent to the energy storage converter for execution.

[0194] When multiple mobile energy storage devices are operating in parallel, the aforementioned The final total energy storage power command is sent to step S7, where step S7 performs single-unit power allocation and SOC balancing control based on the SOC, available charge / discharge power, battery temperature, and equipment availability of each mobile energy storage device.

[0195] Through the above control, the basic operating trend determined by hourly and minute-level scheduling remains unchanged. At the same time, second-level control is used to quickly correct the periodic load changes of the pumping unit and the non-periodic or impact load changes of water injection pumps, workover rigs, mud pumps, etc., so as to achieve coordinated control of "minute-level basic power + second-level stroke synchronous compensation and non-periodic impact compensation + SOC safety constraints".

[0196] Step S7: Execution of energy storage power commands and coordinated control of multiple mobile energy storage devices: When multiple mobile energy storage devices are running in parallel, step S7 receives the final total energy storage power command generated in step S6 that satisfies the SOC and device safety constraints, and obtains the state of charge, available charge and discharge power, battery temperature and device availability of each mobile energy storage device.

[0197] Step S7, taking into account the SOC operating boundary, available charge / discharge power, battery temperature, and equipment availability of each mobile energy storage device, performs basic allocation of the final total energy storage power command generated in Step S6 based on the currently available charging or discharging power of each normal mobile energy storage device.

[0198] Assume the number of normally operating mobile energy storage devices participating in the collaborative operation is... When the number of mobile energy storage devices participating in the coordinated operation is M=0, the basic power allocation and SOC equalization control are stopped, the total achievable energy storage power is set to 0, and the edge controller generates an alarm indicating that all mobile energy storage devices are unavailable. The diesel generator or other backup power supply is then coordinated to be put into operation according to the power supply status of the well site. When M>0, the subsequent basic power allocation and SOC equalization control are executed.

[0199] When the final total energy storage power command generated in step S6 When the sum of the current available discharge power of each normal mobile energy storage device is not less than the final total energy storage power command, it indicates that the energy storage system is discharging, and the basic allocated power of each normal mobile energy storage device is determined according to the proportion of the current available discharge power; when When the sum of the currently available charging power of each normal mobile energy storage device is not less than the absolute value of the final total energy storage power command, it indicates that the energy storage system is performing charging, and the basic allocated power of each normal mobile energy storage device is determined according to the proportion of the currently available charging power. The basic allocated power of the mobile energy storage equipment is:

[0200] in, For the first The basic allocated power of the mobile energy storage equipment; and According to the first The current available discharge power and current available charging power are determined by the SOC, battery temperature, power boundary, and equipment availability status of the mobile energy storage equipment.

[0201] when However, the sum of the currently available discharge power of each normal mobile energy storage device is less than [a certain value]. At this time, no proportional distribution of discharge power is implemented, and each normal mobile energy storage device outputs according to its current available discharge power, that is:

[0202] when However, the sum of the currently available charging power of each normal mobile energy storage device is less than [a certain value]. At this time, no proportional allocation of charging power is implemented, and each normal mobile energy storage device outputs according to its currently available charging power, that is:

[0203] When the total available power in the corresponding direction is sufficient to fulfill the final total energy storage power command, the sum of the base allocated power equals When the total available power is insufficient, the sum of the base allocated power equals the total energy storage power that can be realized at present.

[0204] Based on the initial power allocation, step S7 further adjusts the power allocation of each mobile energy storage device using a SOC self-balancing control strategy, gradually aligning the state of charge (SOC) of different devices. The average SOC of the normally operating mobile energy storage devices participating in the coordinated operation is as follows:

[0205] No. The power adjustment amount of a mobile energy storage device is calculated according to the following formula:

[0206] in, For the first The power adjustment amount of the mobile energy storage equipment; The average state of charge of normally operating mobile energy storage equipment participating in collaborative operation; For the first The current state of charge of the mobile energy storage equipment; and These are the proportional adjustment coefficient and the integral adjustment coefficient, respectively. The most recent determination or reinitialization time of the mobile energy storage equipment set currently participating in collaborative operation; each mobile energy storage equipment in the same collaborative operation set adopts the same integration start time. .

[0207] Under the sign convention that energy storage discharge power is positive and charging power is negative, the first... The single-unit power command of the mobile energy storage equipment after SOC equalization adjustment is as follows:

[0208] When the When the state of charge of the mobile energy storage device is lower than the average state of charge... This reduces the discharge power of the equipment or increases its charging power; when the first When the state of charge of a mobile energy storage device is higher than its average state of charge... This increases the discharge power of the equipment or reduces its charging power, thereby reducing the state of charge deviation between the mobile energy storage devices.

[0209] When the mobile energy storage equipment participating in the collaborative operation is located within the current integral range If the internal state remains unchanged, none of the mobile energy storage devices have reached the SOC boundary and the charge / discharge power boundary, and the total available charge / discharge power of all normal mobile energy storage devices is sufficient to complete the final total energy storage power command generated in step S6, then since all devices use the same average state of charge and integration start time, the sum of the SOC deviation adjustments of each individual device is 0. Combined with the fact that the sum of the base allocated power equals the final total energy storage power command generated in step S6, we can conclude that:

[0210] Among them, the basic power allocation satisfies The SOC equalization power adjustment amount meets the requirements. Therefore, SOC equalization adjustment will not change the total power command of the energy storage system.

[0211] During power allocation, step S7 verifies the SOC operating boundary, available charge / discharge power, battery temperature, and equipment availability of each mobile energy storage device, and ensures that the individual power command meets the following requirements:

[0212] When a mobile energy storage device reaches its charge / discharge power boundary or SOC safety boundary, the individual power command of that device is limited, and the remaining power demand generated after the limitation is redistributed to other normal mobile energy storage devices with adjustment capabilities. When the total available charge / discharge power of the remaining normal mobile energy storage devices is sufficient to meet the remaining power demand, the sum of the redistributed individual power commands is kept equal to the final total energy storage power command generated in step S6; when the total available charge / discharge power of the remaining normal mobile energy storage devices is insufficient, each device outputs according to the current available power boundary, and the insufficient energy storage output processing described later is executed.

[0213] When the single-unit power command of a mobile energy storage device is continuously limited due to the charging / discharging power boundary or the SOC safety boundary, the SOC equalization integral term of the device is subjected to anti-saturation processing by pausing integration or limiting integration; when the limiting condition is lifted, the integration update of the device is restored.

[0214] When some mobile energy storage devices experience over-temperature, over-current, insulation abnormalities, communication failures, or other unavailability, the abnormal devices are removed from the collaborative power allocation, and the set of normal mobile energy storage devices participating in collaborative operation and its quantity M are re-determined. When the re-determined M=0, the aforementioned handling method for all mobile energy storage devices being unavailable is executed, and the current step S7 control cycle ends; when the re-determined M>0, the average state of charge and basic allocated power are recalculated, and the remaining power demand is redistributed among the remaining normal mobile energy storage devices; simultaneously, the current moment is taken as the new integration start moment. The SOC equalization integral term for each mobile energy storage device currently participating in the collaborative operation is reinitialized to 0. When a mobile energy storage device that has exited the collaborative operation returns to normal and rejoins the collaborative power allocation, or when the number and set of devices participating in the collaborative operation change for other reasons, the set of normal mobile energy storage devices participating in the collaborative operation and its quantity M are redefined. When the redefined M=0, the aforementioned handling method for all mobile energy storage devices being unavailable is followed. When the redefined M>0, the calculation is recalculated. The power allocation for each piece of equipment is re-allocated according to the current final total energy storage power command, and the SOC equalization integral term of all participating equipment is re-initialized according to the new participating set.

[0215] During any basic power allocation, SOC equalization adjustment, or residual power reallocation process, when the total available charging and discharging power of the normally operating mobile energy storage equipment participating in the coordinated operation is sufficient to meet the final total energy storage power command generated in step S6, the aforementioned basic power allocation and SOC equalization control methods are executed. When the total available power in the corresponding charging or discharging direction is insufficient, each normally operating mobile energy storage equipment outputs its currently achievable power according to its available power boundary. The edge controller generates an energy storage output insufficiency alarm based on the difference between the actual achievable total power and the final total energy storage power command, and coordinates the operation of diesel generators or other backup power sources according to the well site power supply status. During periods of insufficient total available power, SOC equalization adjustments that may further reduce the current achievable total power are not performed; SOC equalization control is resumed after the total available power is restored.

[0216] Step S7 sends the allocated or reallocated single-unit charge / discharge power commands to the corresponding energy storage converters for execution. When the total available charge / discharge power of the normally mobile energy storage equipment participating in the coordinated operation is sufficient, each piece of equipment jointly completes the final total energy storage power command generated in step S6; when the total available charge / discharge power is insufficient, each piece of equipment jointly outputs the currently achievable total energy storage power. Simultaneously, SOC equalization control improves the stability of multi-unit parallel operation and the continuity of power supply to the well site.

[0217] This embodiment is applied to a remote single-well site with a pumping unit. The electrical equipment at the well site includes pumping units, auxiliary pumps, heating equipment, lighting equipment, and communication equipment, and a diesel generator is configured as a backup power source. A three-compartment integrated mobile energy storage device is deployed at the well site. The mobile energy storage device includes an isolated battery compartment, a power conversion compartment, and a control compartment. Its rated power and rated capacity are determined according to the well site load and backup power supply requirements.

[0218] After the mobile energy storage equipment arrives at the well site, it connects to the well site's power supply and distribution system through the well site bus interface, diesel generator interface, communication interface, grounding interface, and fire alarm linkage interface. After completing grounding confirmation, communication confirmation, insulation testing, and voltage, frequency, and phase sequence verification, the control and off-grid switching device completes the connection, enabling the mobile energy storage equipment to enter islanded operation or coordinate with the diesel generator.

[0219] The edge controller collects data on pumping unit displacement, load, stroke phase, well site load power, energy storage state of charge, battery temperature, equipment availability, and diesel generator output according to step S1, and performs time synchronization, anomaly processing, and feature extraction on the collected data. According to step S2, the physical characteristics of the pumping unit dynamometer diagram and historical load time-series data are input into a Long Short-Term Memory (LSTM) network to form the overall predicted load curve for the well site.

[0220] The edge controller identifies the well site operation scenario and calculates the urgency of the four objectives according to step S3; it generates an hourly basic charge and discharge plan according to step S4, and reserves the bidirectional adjustment power margin and adjustment energy margin required for step S6. When there is a power supply guarantee requirement, the energy storage backup power requirement and energy storage backup energy requirement are determined based on the critical load power, preset guarantee duration, and available output of the diesel generator.

[0221] According to step S5, the edge controller performs minute-level rolling correction on the hourly base charge / discharge power within the power boundary and state-of-charge operating boundary after deducting the adjustment margin. According to step S6, the stroke phase synchronization compensation power is generated based on the real-time stroke phase of the pumping unit and the deviation between the real-time load power and the average load power, and the second-level compensation power is limited to the bidirectional adjustment power margin reserved in step S4.

[0222] The edge controller performs safety verification on the target power of energy storage based on the dynamic state of charge lower limit, fixed safety upper limit, energy storage converter charging and discharging power boundaries, and equipment availability. In power supply guarantee scenarios where critical loads are at risk of power loss, the fixed basic safety lower limit preset by the battery management system is used as the discharge state lower limit. When the available power or available energy of the energy storage is insufficient, an insufficient energy storage output alarm is generated, and the diesel generator is coordinated to start operation or increase its output.

[0223] Since this embodiment uses a single mobile energy storage unit, the final energy storage power command is directly sent to the bidirectional energy storage converter for execution. The cloud management layer is used to receive operating data, scheduling results, and fault alarms; when cloud communication is abnormal or interrupted, the edge controller continues to execute well site energy storage control based on the local model and the most recent valid configuration.

[0224] The above methods enable the prediction of periodic loads and stroke phase synchronization compensation for single-port pumping units, and improve the load mitigation capability and power supply continuity of remote well sites through energy storage backup management and local autonomous control.

[0225] Example 2: Parallel operation of multiple units in a 3–5 well group (Scenario B): This embodiment applies to a well group consisting of 3 to 5 pumping units, water injection pumps, and auxiliary electrical equipment. The well site has access to the public power grid and a time-of-use pricing mechanism, and a contracted demand or maximum allowable power is set. When multiple pumping units are running at similar stroke stages, their peak loads may overlap; the start-up, shutdown, or switching of the water injection pumps may also cause rapid changes in the power on the power grid side of the well site.

[0226] Based on the total load of the well group, demand control requirements, and backup power supply time, multiple integrated mobile energy storage devices with three compartments are deployed at the well site. As an example, four mobile energy storage devices with a rated power of 200 kW and a rated available capacity of 500 kWh can be deployed at the well site, forming an energy storage system with a total rated power of 800 kW and a total rated available capacity of 2 MWh. The number of devices, rated power, and rated capacity can be adjusted according to the load scale of the well group. Each mobile energy storage device is connected in parallel via the well site's AC or DC bus, and the edge controller uniformly acquires the state of charge, available charge / discharge power, battery temperature, and equipment availability of each device.

[0227] The edge controller collects the displacement, load, and stroke phase of each pumping unit according to step S1, as well as the operating status of the water injection pump, the well group load power, the time-of-use electricity price, the energy storage system status, the grid status, and the backup power status. According to step S2, it predicts the load power of each pumping unit, and predicts its start-up and shutdown status, power change time, and load change trend based on the operating status, current change, power change trend, and start-up and shutdown information of the water injection pump, thereby forming the overall predicted load curve of the well group.

[0228] According to step S3, when the predicted peak power reaches or exceeds the well site demand warning value, the well site is identified as a demand-limited scenario; when the periodic load peaks of multiple pumping units overlap, or the predicted power change of the water injection pump exceeds the preset sudden change threshold, the well site is identified as a high-frequency pulse scenario; when the above scenario judgment conditions are met simultaneously, the well site is identified as a composite operation scenario. The edge controller further calculates the urgency parameters corresponding to the four objectives of economic operation, demand control, load shock mitigation, and power supply reliability assurance.

[0229] According to step S4, the edge controller schedules energy storage charging during off-peak electricity price periods and energy storage discharging during peak electricity price periods, and increases the target state of charge before the predicted peak period. For high-frequency pulse scenarios, the bidirectional adjustment power margin and adjustment power margin required in step S6 are reserved; for operating periods with power supply guarantee requirements, the energy storage backup power requirements and energy storage backup power requirements are determined based on the critical load power, preset guarantee duration, and available output of other backup power sources.

[0230] According to step S5, the edge controller performs minute-level rolling corrections on the hourly base charge and discharge power based on the real-time well group load, the latest forecast results, and the target urgency parameters output in step S3. When the grid-side power of the well group approaches the demand limit, the urgency of demand control and its corresponding dynamic weights are increased; when the load peak-to-average power ratio and load fluctuation variance increase, the urgency of load shock mitigation and its corresponding dynamic weights are increased. The minute-level energy storage power command is determined within the charge and discharge power boundaries and the state-of-charge operating boundaries after deducting the adjustment margin in step S6.

[0231] According to step S6, the edge controller generates stroke phase synchronous compensation power for each pumping unit based on the real-time stroke phase, peak stroke phase, trough stroke phase, and real-time power deviation. For non-periodic or impact loads such as water injection pumps, rapid compensation power is generated based on predicted start-stop status, predicted power change times, and real-time power deviation. The superimposed compensation power is limited to the bidirectional adjustment power margin reserved in step S4 and superimposed with the minute-level energy storage power command. After verification of state of charge, charge / discharge power boundaries, and equipment availability, the final total energy storage power command is formed.

[0232] According to step S7, the edge controller performs a basic allocation of the final total energy storage power command based on the available charging and discharging power of each normally mobile energy storage device, and performs a state of charge balancing adjustment based on the deviation between the state of charge of each device and the average state of charge. When the total available power is sufficient and each device has not reached its operating boundary, the sum of the individual power commands of each device remains equal to the final total energy storage power command.

[0233] When a mobile energy storage device reaches its state of charge boundary, charge / discharge power boundary, or temperature limit, the edge controller limits its individual power command and reallocates the remaining power demand to other devices with adjustment capabilities. When a device experiences over-temperature, over-current, insulation abnormality, communication failure, or other unavailability, it is removed from the collaborative power allocation, the set of devices participating in operation is re-determined, and the base allocation power and average state of charge are recalculated. When the total available power of normally operating devices is insufficient, each device outputs according to its current available power boundary, and an energy storage output insufficiency alarm is generated.

[0234] The above methods enable hourly economic dispatching by combining time-of-use pricing and well group load forecasting, minute-level rolling optimization control of the maximum power on the well site grid side, second-level compensation to reduce power fluctuations caused by the superposition of multiple pumping unit loads and the start-up and shutdown of water injection pumps, and the coordination and continuous regulation capability of mobile energy storage equipment in parallel operation to improve the power distribution of multiple units and the balance of charge status.

[0235] Example 3: Temporary power supply for well workover / well rotation operations (Scenario C): This embodiment is applied to well workover, well rotation, or other temporary work sites. The on-site power load includes workover rigs, mud pumps, hoisting equipment, lighting equipment, and communication equipment, and is characterized by frequent changes in deployment locations, short work cycles, insufficient historical load data, and significant equipment start-ups, shutdowns, and power fluctuations.

[0236] Based on the maximum on-site load, operation duration, and backup power requirements, one or more integrated three-compartment mobile energy storage units are configured, with a diesel generator reserved as a backup power source. After the mobile energy storage unit arrives at the work site, it connects to the workover rig, mud pump, on-site power distribution bus, backup diesel generator, and safety protection system via the well site bus interface, diesel generator interface, communication interface, grounding interface, and fire alarm linkage interface. After completing grounding confirmation, communication confirmation, insulation testing, and voltage, frequency, and phase sequence verification, the control and off-grid switching device completes the connection.

[0237] According to step S1, the edge controller collects the current, power, start / stop status and operating status of the workover rig, mud pump and other operating equipment, and collects the energy storage charge status, charging and discharging power, battery temperature, equipment availability status and the current available output of the diesel generator.

[0238] For new work sites with insufficient historical data, the edge controller, following step S2, generates initial prediction results for non-periodic or impact loads based on the rated parameters, short-term trial operation data, current changes, power change trends, and start-up / shutdown information of equipment such as workover rigs and mud pumps. This prediction yields the predicted start-up / shutdown status, predicted power surge times, and load change trends. As on-site operational data accumulates, when communication conditions permit, the edge controller uploads new operational data and load prediction deviations to the cloud-based model training and parameter management module. This module then trains or corrects the load prediction model and related parameters, and sends the verified model parameters back to the edge controller.

[0239] According to step S3, when the load power change between adjacent predicted times exceeds the preset mutation threshold due to the lifting or lowering of the workover rig or the start of the mud pump, the operating scenario is directly identified as a high-frequency pulse scenario, regardless of the predicted load peak-to-average ratio determination condition, and the urgency of load impact mitigation is set to 1. When the power grid is unavailable, the available output of the diesel generator is insufficient, the energy storage backup capacity is insufficient, or there is a risk of power outage for critical operating loads, the operating scenario is identified as a power supply guarantee scenario; when multiple scenario conditions are met simultaneously, it is identified as a composite operating scenario.

[0240] According to step S4, the edge controller generates an hourly basic charge-discharge plan based on the expected operation phase and increases the target state of charge before the workover rig is lifted, the mud pump is started, or other high-power operations begin. For high-frequency pulse scenarios, bidirectional power and energy margins are reserved; for power supply guarantee scenarios, the energy storage backup power and energy storage backup requirements to be undertaken by the energy storage system are determined based on the critical operation load power, preset guarantee duration, and available diesel generator output.

[0241] According to step S5, the edge controller performs minute-level rolling adjustments to the hourly baseline charge and discharge power based on the actual work progress, real-time load, latest forecast results, and energy storage system status. The minute-level optimization is performed within the power boundary and state-of-charge operating boundary after deducting the adjustment margin in step S6, so as to avoid prematurely consuming the energy storage capacity used for sudden load compensation and power supply for critical work loads.

[0242] Since workover rigs, mud pumps, and hoisting equipment do not possess stable pumping unit stroke phase characteristics, step S6 does not perform stroke phase window determination for these devices. When step S2 predicts that the corresponding device will start, stop, or experience a power surge, or detects that the absolute value of the deviation between its real-time load power and short-term average load power exceeds a preset power surge threshold, the edge controller generates a non-periodic impact load rapid compensation power based on this power deviation. When the real-time load power is higher than the short-term average load power, the mobile energy storage equipment is controlled to perform discharge compensation; when the real-time load power is lower than the short-term average load power and the energy storage system allows charging, the mobile energy storage equipment is controlled to absorb power.

[0243] The rapid compensation power limit is set within the bidirectional adjustment power margin reserved in step S4, and is superimposed with the minute-level energy storage power command to form the second-level energy storage target power. The edge controller performs safety verification based on the dynamic state of charge lower limit, fixed safety upper limit, energy storage converter charge and discharge power boundary, and equipment availability status. When the site is in a power supply guarantee scenario and the risk of power loss for critical work loads is 1, the fixed basic safety lower limit preset by the battery management system is used as the discharge state lower limit to prioritize power supply to critical work loads.

[0244] When the available power or electricity of the energy storage is insufficient, the edge controller generates an alarm indicating insufficient energy storage output and coordinates the diesel generator to start or increase its output. When one mobile energy storage device is configured on-site, the final energy storage power command is directly sent to the local bidirectional energy storage converter for execution; when multiple mobile energy storage devices are configured, the final total energy storage power command is distributed to individual units and the state of charge is balanced according to step S7.

[0245] When the site is in a public network communication blind spot or cloud communication is interrupted, the edge controller continues to perform load forecasting, power regulation and safety protection based on the locally stored prediction model, control parameters and the most recent valid configuration. The backup communication link is used to transmit necessary operating status and fault alarms.

[0246] After the operation is completed, stop the power output of the mobile energy storage equipment according to the preset shutdown sequence, disconnect and disconnect the off-grid switching device, disconnect the electrical, communication, grounding and fire protection linkage connections, and transfer the mobile energy storage equipment to the next operation location.

[0247] The above methods can improve the deployment and relocation efficiency of temporary work sites by utilizing the integrated three-compartment structure and standardized quick-connect interface of mobile energy storage equipment. It can also reduce the power impact on the site bus by providing second-level rapid compensation for non-periodic or impact loads such as workover rigs and mud pumps. At the same time, it can improve the continuity of power supply for temporary operations through energy storage backup management, diesel generator coordination, and edge local autonomy.

[0248] Implementation Method 2 This embodiment provides a digital management and control system for mobile energy storage at well sites based on edge-cloud collaboration. The system uses the mobile energy storage method for well sites based on dynamometer card coupling as described in Embodiment 1. The system includes end-side equipment, an edge control layer, a cloud management layer, and a communication network connecting the end-side equipment, the edge control layer, and the cloud management layer. The end-side equipment includes a mobile energy storage device, a pumping unit dynamometer diagram acquisition device, and well site power supply and distribution equipment. It is used to collect pumping unit displacement, load, stroke phase, well site load, energy storage charge status, energy storage charging and discharging power, battery temperature, and equipment operating status, and to perform energy storage charging, energy storage discharging, load impact compensation, and power regulation according to the control commands issued by the edge control layer. The edge control layer is used to perform time synchronization, abnormal data processing, feature extraction and local storage of the data collected by the end-side equipment; to perform load prediction based on the physical characteristics of the pumping unit's dynamometer diagram and historical load data; to identify well site operation scenarios and calculate the urgency parameters corresponding to the four objectives of economic operation, demand control, load impact mitigation and power supply reliability assurance; to generate hourly basic charge and discharge plans; to perform minute-level dynamic weighted multi-objective rolling optimization, pumping unit second-level stroke phase synchronization compensation, non-periodic impact load rapid compensation and state of charge safety verification; and to perform single-unit power allocation and state of charge balancing control when multiple mobile energy storage devices are running in parallel. The cloud management layer includes a data management module, an energy storage-operating condition digital twin module, a model training and parameter management module, a remote monitoring module, and a remote control module, which are used to store, analyze, perform model feedback correction, remotely monitor, and manage remote control commands for well site operation data. The communication network is used to enable data interaction between the end-side equipment, the edge control layer, and the cloud management layer.

[0249] Furthermore, the energy storage-operating condition digital twin module is used to establish the correlation between the operating conditions of the pumping unit, the well site load and the energy storage operation response, and to compare the predicted load with the actual load, the energy storage control command with the actual power response and the target state of charge with the actual state of charge to obtain the load prediction deviation and the energy storage response deviation. The model training and parameter management module is used to train or correct the load prediction model and control parameters based on historical operating data, the load prediction deviation and the energy storage response deviation, and to send the verified model parameters and control parameters to the edge control layer. The edge control layer is also used to verify the permissions, integrity, device status and security constraints of the operating mode, scheduling parameters, model parameters, power limit values ​​and remote control commands issued by the cloud management layer, and to execute the corresponding control after the verification is passed, and to refuse to execute and feed back abnormal information when the verification fails. When communication between the edge control layer and the cloud management layer is abnormal or interrupted, the edge control layer continues to execute well site energy storage control based on the locally stored prediction model, control parameters and the most recent valid configuration. When the primary communication link fails, the communication network switches to an available backup communication link; The system is equipped with identity authentication, data encryption, access control, operation auditing, fault signal linkage, and fault isolation mechanisms.

[0250] Implementation Method 3 This embodiment provides a mobile energy storage equipment for well sites, including a container body that can be transported and hoisted as a whole. The mobile energy storage equipment is connected as an end-side device to a well site mobile energy storage digital management and control system based on edge-cloud collaboration, and performs charging, discharging and power regulation according to the energy storage power command generated by the multi-timescale collaborative optimization scheduling method based on dynamometer diagram coupling. The container is equipped with a battery compartment, a power conversion compartment, and a control compartment that are isolated from each other. The battery compartment is equipped with battery clusters, a battery management system, a thermal management device, and a fire-fighting device. The power conversion compartment is equipped with a bidirectional energy storage converter, AC / DC power distribution equipment, and grid-connected / off-grid switching equipment. The control cabin is equipped with an energy management controller, an edge controller, a communication gateway, and a human-machine interaction device; The battery compartment, power conversion compartment and control compartment are connected by power lines, control lines and communication bus, and are equipped with fireproof isolation devices, pressure relief devices, fire-fighting linkage devices and fault isolation devices; The outer side of the container body is provided with one or more of the following interfaces: AC interface, DC interface, diesel generator interface, photovoltaic interface, well site bus interface, communication interface, grounding interface, and fire linkage quick-connect interface. The interface is connected to the power conversion compartment through a circuit breaker, disconnect switch, or contactor, and is equipped with anti-misinsertion device and electrical interlock device. The edge controller is used to perform load prediction, well site operation scenario identification, hourly basic charge and discharge plan generation, minute-level dynamic weighted rolling optimization, second-level stroke phase synchronization compensation and state of charge safety verification based on pumping unit dynamometer data, well site load power data and energy storage system status data, generate energy storage power commands, and control the operation of the bidirectional energy storage converter according to the energy storage power commands. The edge controller and / or energy management controller are also used to control the grid-connected switching device to complete the connection of the well site power supply and distribution system after completing grounding confirmation, communication confirmation, insulation detection, and voltage, frequency and phase sequence verification. When over-temperature, over-current, insulation abnormality, smoke or communication failure is detected, alarm, power limitation, fault branch disconnection or fault compartment isolation are performed according to the fault level.

[0251] The mobile energy storage equipment can operate in grid-connected or islanded mode, and performs peak shaving and valley filling, impact load compensation, bus voltage and frequency support, and backup power coordination control according to the well site load, energy storage charge status, and backup power status. The equipment is equipped with dustproof, waterproof, corrosion-proof, explosion-proof, and wide-temperature-range thermal management measures according to the well site environment, and is suitable for single wells, well groups, and temporary operation scenarios such as well workover and well transfer.

Claims

1. A multi-timescale collaborative optimization scheduling method for a mobile energy storage system at a well site, characterized in that, The method includes the following steps: Step S1: Collect pumping unit dynamometer data, well site load power data, energy storage system status data, time-of-use electricity price information, well site electrical equipment operation status data, and well site power supply and backup power status data. Perform time synchronization, abnormal data identification and processing, and feature extraction on the collected data to form a well site operation dataset. Step S2: Establish a load prediction model combining the physical model of the pumping unit's dynamometer diagram and a Long Short-Term Memory (LSTM) network. Input the physical load characteristics obtained from the physical model of the pumping unit's dynamometer diagram, as well as the pumping unit's displacement, load, stroke phase, and historical load time series data, into the LSTM network to obtain the pumping unit load prediction results. Based on the operating status, current variation characteristics, power variation trend, and start-up / shutdown information of aperiodic or impulsive loads, predict the start-up / shutdown status, power mutation time, and load variation trend of the aperiodic or impulsive loads. Superimpose the predicted power of each well site's electrical equipment at the same prediction time to form the overall predicted load curve for the well site. Step S3: Calculate the predicted average load power, predicted peak power, predicted load fluctuation variance and predicted load peak-to-average ratio based on the total predicted load curve of the well site. Combined with the well site power grid status, backup power status, energy storage charging status, demand control limits and critical load power supply status, identify the well site operation scenario and calculate the urgency parameters corresponding to the four objectives of economic operation, demand control, load impact mitigation and power supply reliability assurance. Step S4: Generate an hourly basic charge and discharge plan based on the load forecast results, well site operation scenario and target urgency parameters, and determine the bidirectional adjustment power margin and adjustment energy margin for second-level rapid compensation, as well as the energy storage backup power requirement and energy storage backup energy requirement undertaken by the energy storage system. Step S5: Based on the hourly basic charge and discharge plan, real-time well site load, latest load prediction results, energy storage system status, and the target urgency parameter most recently updated by step S3, calculate the dynamic weights corresponding to the four objectives, and perform dynamic weighted multi-objective rolling optimization within the minute-level feasible domain after reserving adjustment power and adjustment amount for second-level rapid compensation, to obtain minute-level energy storage power instructions; Step S6: Determine the stroke phase corresponding to the load peak and the stroke phase corresponding to the load trough based on the pumping unit indicator diagram data and real-time load power data within a series of strokes. Generate pumping unit stroke phase synchronous compensation power based on the real-time stroke phase of the pumping unit and the deviation of the real-time load power relative to the average load power of the corresponding pumping unit stroke. Generate aperiodic impact load rapid compensation power based on the predicted start-stop status of the non-periodic or impact load, the predicted power change time, the load change trend, and the deviation of the real-time load power relative to the short-term average load power. The pumping unit stroke phase synchronization compensation power and non-periodic impact load rapid compensation power are superimposed, and the superposition result is limited within the bidirectional adjustment power margin. Then, it is superimposed with the minute-level energy storage power command to form the second-level energy storage target power. The lower limit of dynamic state of charge operation is determined based on the predicted peak-to-average load ratio. Based on whether the well site is in a power supply guarantee scenario and the risk of power failure of critical loads, the lower limit of the current discharge state is determined from the lower limit of dynamic state of charge operation and the fixed basic safety lower limit preset by the battery management system. The second-level energy storage target power is safely verified based on the lower limit of discharge state, the fixed safety upper limit preset by the battery management system, the charging and discharging power boundary of the energy storage converter, and the availability status of the equipment. The final energy storage power command is generated, and the energy storage converter of the mobile energy storage equipment is controlled to charge or discharge according to the final energy storage power command. Step S7: When a single mobile energy storage device is running, the final energy storage power command is sent to the corresponding energy storage converter for execution; when multiple mobile energy storage devices are running in parallel, the final energy storage power command is used as the final total energy storage power command. Based on the state of charge, available charge and discharge power, battery temperature and equipment availability of each mobile energy storage device, basic power allocation and state of charge balancing adjustment are performed to form the individual charge and discharge power command of each mobile energy storage device, and the individual charge and discharge power command is sent to the corresponding energy storage converter for execution.

2. The method according to claim 1, characterized in that, Step S2 specifically involves: establishing a physical model of the pumping unit's dynamometer diagram, which consists of a static load term, a linear load term corresponding to the pumping unit's displacement, and a periodic dynamic load term corresponding to the pumping unit's stroke period; identifying and fitting the model parameters of the physical model of the pumping unit's dynamometer diagram to obtain physical load characteristics that reflect the mechanical operating state of the pumping unit. The physical load characteristics, along with the measured displacement, load, stroke phase, and historical load time series data of the pumping unit, are input into the trained Long Short-Term Memory (LSTM) network to predict the load power of each pumping unit in the future within a preset time domain. For water injection pumps, well workover rigs, mud pumps and other non-periodic or impact loads, the load status is identified based on their operating status, current variation characteristics, power variation trend and start-stop information, and their start-stop status, power change time and load variation trend are predicted. The predicted power of each pumping unit and the non-periodic or shock load at the same predicted time is superimposed to form the total predicted load curve of the well site, and the predicted peak power, the predicted average load power of the well site, and the predicted time of the peak are obtained.

3. The method according to claim 1, characterized in that, Step S3 specifically involves calculating the predicted average load power, predicted peak power, predicted load fluctuation variance, and predicted load peak-to-average ratio based on the total predicted load curve of the well site. When the predicted load peak-to-average ratio exceeds the preset peak-to-average ratio judgment threshold, and the predicted load fluctuation variance exceeds the preset variance reference upper limit, or when the well site total predicted load curve has repeated peak-to-valley changes corresponding to the pumping unit stroke cycle, the well site operation scenario is identified as a high-frequency pulse scenario. When the change in load power between adjacent predicted times of non-periodic or impact loads exceeds a preset abrupt change threshold, the well site operation scenario is identified as a high-frequency pulse scenario, regardless of the predicted load peak-to-average ratio determination condition. When the predicted load peak-to-average ratio does not exceed the preset peak-to-average ratio judgment threshold, the predicted load fluctuation variance does not exceed the preset variance reference upper limit, and there are no periodic shocks, instantaneous power surges, power grid anomalies, or critical load power supply risks, the well site operation scenario will be identified as a continuous production scenario. When the predicted peak power reaches or exceeds the well site demand warning value, the well site operation scenario will be identified as a demand-limited scenario. When at least one of the following situations occurs: grid voltage or frequency reaches the warning limit, grid power failure or risk of power failure, well site enters islanded operation or there is a need for islanding switchover, backup power supply capacity is insufficient, energy storage backup capacity is lower than the preset safety value, or critical loads are at risk of power outage, the well site operation scenario is identified as a power supply guarantee scenario; when the judgment conditions of two or more operation scenarios are met simultaneously, the well site operation scenario is identified as a composite operation scenario; the urgency of economic operation, the urgency of demand control, the urgency of load impact mitigation, and the urgency of power supply reliability guarantee are calculated respectively.

4. The method according to claim 1, characterized in that, The specific steps of step S4 are as follows: Based on the load forecast results, well site operation scenario and target urgency parameters, and combined with time-of-use electricity price, current state of charge of energy storage, available capacity of energy storage and power supply reliability requirements of well site, determine the charging, discharging or standby status of each hourly scheduling period within the future preset scheduling cycle, and calculate the corresponding hourly basic charging and discharging power and target state of charge. In high-frequency pulse scenarios, the bidirectional adjustment power margin for the second-level rapid compensation in step S6 is determined based on the difference between the predicted peak power and the predicted average load power, the expected number of loads participating in compensation simultaneously, and the current available charging and discharging power of the energy storage. The adjustment power margin is determined based on the bidirectional adjustment power margin, the expected compensation duration, and the current available energy storage power. In demand-constrained scenarios, the target state of charge before the predicted peak period is increased, and the hourly base discharge power is set during the predicted peak period to reserve the discharge capacity required during the predicted peak period. In the scenario of power supply guarantee, based on the critical load power, the preset guarantee duration and the available output of diesel generators and other backup power sources, the backup power supply capacity that non-energy storage backup power sources can undertake is deducted, and the energy storage backup power requirements and energy storage backup electricity requirements undertaken by the energy storage system are determined. In a combined operation scenario, when the current available power or available electricity of the energy storage is insufficient to simultaneously meet the regulation margin of second-level rapid compensation and the backup requirements for critical load power supply, priority is given to reserving the energy storage backup power and energy storage backup electricity required for critical load power supply, and the bidirectional regulation power margin and regulation electricity margin are adjusted accordingly based on the remaining available power and remaining available electricity.

5. The method according to claim 4, characterized in that, Specifically, step S5 involves reading the economic operation urgency obtained from the most recent update in step S3, and calculating the dynamic weights corresponding to each target based on the urgency parameters, basic weights, and weight adjustment coefficients of each target, so that each dynamic weight is not less than 0 and the sum of each dynamic weight is 1. Based on the deviation between the minute-level energy storage charging and discharging power to be determined and the hour-level basic charging and discharging power for the current time period determined in step S4, a minute-level economic operation target is constructed. The power of the grid side of the well site after energy storage regulation is determined based on the current total load power of the well site and the minute-level energy storage charging and discharging power, and the load demand control target is constructed based on the grid side power of the well site, the well site demand warning value and the maximum allowable power; Based on the change between the power on the well site grid side corresponding to the current minute-level control cycle and the power on the well site grid side corresponding to the previous minute-level control cycle, a load shock mitigation target is constructed. Calculate the power margin that can be further used for critical load power backup in the discharge direction at the current operating point after deducting the bidirectional adjustment power margin determined in step S4. Based on the current energy storage state of charge, minute-level energy storage charge and discharge power, minute-level control cycle, rated available energy storage capacity and charge and discharge efficiency, determine the expected state of charge after executing the minute-level energy storage charge and discharge power. Then determine the energy storage capacity that can be used for critical load backup power after deducting the adjustment power margin determined in step S4. Based on the power margin, energy storage capacity, and the energy storage backup power and energy storage backup capacity requirements determined in step S4, a power supply reliability assurance target is constructed. The minute-level economic operation target, load demand control target, load impact mitigation target, and power supply reliability guarantee target are weighted according to the dynamic weights to form a minute-level comprehensive objective function. Based on the principle of minimizing the minute-level comprehensive objective function, the minute-level energy storage power command is determined within the minute-level feasible domain jointly formed by the energy storage charge and discharge power boundary after deducting the bidirectional adjustment power margin determined in step S4 and the state of charge operation boundary after deducting the adjustment power margin determined in step S4. At the beginning of each minute-level control cycle, the latest well site load, the latest load prediction result, the current energy storage charge status and the current equipment status are reacquired, and the target urgency parameter obtained in the most recent update in step S3 is read. The dynamic weights and various targets are recalculated, and the minute-level energy storage power command is updated. In step S5, the most recent valid results of the operation scenario identification and target urgency parameter calculation in step S3 are read, without repeating the well site operation scenario identification and target urgency parameter calculation in step S5.

6. The method according to claim 1, characterized in that, Step S6 specifically involves: determining the first stroke based on the indicator diagram data and real-time load power data within a series of consecutive strokes. i The peak load of the pumping unit corresponds to the stroke phase and the trough load corresponds to the stroke phase. The annular phase deviation between the real-time stroke phase and the stroke phase corresponding to the peak load and the trough load is calculated. Calculate the first based on the real-time load power of the current stroke or the most recent complete strokes. i The average load power of the pumping unit during its stroke; When the real-time stroke phase enters the preset phase window corresponding to the load peak, and the first i When the real-time load power of the pumping unit is higher than the average load power of the stroke, a positive value of the pumping unit stroke phase synchronization compensation power is generated to control the energy storage converter to perform discharge compensation. When the real-time stroke phase enters the preset phase window corresponding to the load trough, and the real-time load power is lower than the average stroke load power, a negative value of the pumping unit stroke phase synchronous compensation power is generated to control the power absorption of the energy storage converter. For non-periodic or impulsive loads that do not have stable stroke phase characteristics, the short-time average load power is calculated based on the real-time load power within the most recent preset second-level time window. When step S2 predicts that the load will start, stop, or experience a power surge, or when the absolute value of the deviation between the real-time load power and the short-time average load power exceeds a preset power surge threshold, the non-periodic impulsive load rapid compensation power is generated based on the deviation. The stroke phase synchronization compensation power of each pumping unit and the rapid compensation power of each non-periodic or impact load are superimposed, and the total second-level compensation power is limited within the bidirectional adjustment power margin determined in step S4. The total second-level compensation power is superimposed with the minute-level energy storage power command output in step S5 to form the second-level energy storage target power; The lower limit of the state of charge is dynamically adjusted according to the predicted peak-to-average load ratio. When the predicted peak-to-average load ratio does not exceed the preset peak-to-average load ratio reference value, the lower limit of the dynamic state of charge is set to the fixed basic safety lower limit preset by the battery management system. When the predicted peak-to-average load ratio exceeds the preset peak-to-average load ratio reference value, the lower limit of the dynamic state of charge operation is increased as the predicted peak-to-average load ratio increases, but does not exceed the preset maximum lower limit of operation. When the well site is in a power supply guarantee scenario and the risk of power loss of critical loads is 1, the currently used discharge state lower limit is switched to the fixed foundation safety lower limit; otherwise, the dynamic charge state operating lower limit is used.

7. The method according to claim 1, characterized in that, Specifically, step S7 involves sending the final energy storage power command generated in step S6 to the corresponding energy storage converter for execution when a single mobile energy storage device is in operation. When multiple mobile energy storage devices are running in parallel, the final energy storage power command generated in step S6 is used as the final total energy storage power command, and the state of charge, available charge and discharge power, battery temperature and device availability of each mobile energy storage device are obtained. Let M be the number of normal mobile energy storage devices participating in the collaborative operation. When M=0, the basic power allocation and state of charge balancing control are stopped, the current achievable total energy storage power is set to 0, and an alarm for all mobile energy storage devices being unavailable is generated. When M>0, the basic power allocation is performed on the final total energy storage power command based on the current available charging or discharging power of each normal mobile energy storage device. When the total available power of each normal mobile energy storage device in the corresponding charging or discharging direction is sufficient to meet the final total energy storage power command, the basic power allocation of each mobile energy storage device is determined according to the ratio of the current available charging power or the current available discharging power of each normal mobile energy storage device; when the total available power in the corresponding direction is insufficient, each normal mobile energy storage device outputs according to its current available power boundary. The average state of charge (SOC) of the normally operating mobile energy storage equipment participating in the coordinated operation is calculated. Based on the deviation between the current SOC of each mobile energy storage equipment and the average SOC, the SOC equalization power adjustment amount is determined. Based on the SOC equalization power adjustment amount, the basic power allocation of each mobile energy storage equipment is adjusted so that the mobile energy storage equipment with an SOC lower than the average SOC reduces its discharge power or increases its charging power, and the mobile energy storage equipment with an SOC higher than the average SOC increases its discharge power or decreases its charging power. When the mobile energy storage equipment reaches the charging and discharging power boundary or the state of charge safety boundary, the charging and discharging power command of the corresponding single unit is limited, and the remaining power demand generated after the limit is redistributed to other normal mobile energy storage equipment with adjustment capabilities. When the single-machine charging and discharging power command is continuously limited due to the charging and discharging power boundary or the state of charge safety boundary, anti-saturation processing such as pausing integration or limiting integration is performed on the corresponding state of charge balance integral term. When a mobile energy storage device experiences over-temperature, over-current, insulation abnormality, communication failure, or other unavailable conditions, the abnormal mobile energy storage device will be removed from the collaborative power allocation. The set and number M of normal mobile energy storage devices participating in collaborative operation will be re-determined, and the average state of charge and basic allocated power will be recalculated. At the same time, the state of charge balance integral term will be re-initialized. When the total available power of the normally mobile energy storage equipment participating in the coordinated operation is insufficient to meet the final total energy storage power command in the corresponding charging or discharging direction, each normally mobile energy storage equipment outputs according to the current available power boundary to determine the current achievable total energy storage power, generates an energy storage power shortage alarm, and coordinates the diesel generator or other backup power sources to be put into operation.

8. A digital management and control system for mobile energy storage at well sites based on edge-cloud collaboration, characterized in that, The system uses the multi-timescale collaborative optimization scheduling method for mobile energy storage systems at well sites as described in any one of claims 1-7. The system includes end-side equipment, an edge control layer, a cloud management layer, and a communication network connecting the end-side equipment, the edge control layer, and the cloud management layer. The end-side equipment includes mobile energy storage equipment, a pumping unit dynamometer card acquisition device, and well site power supply and distribution equipment. The pumping unit indicator diagram acquisition device is used to acquire displacement, load, and stroke phase data of the pumping unit. The well site power supply and distribution equipment is used to provide data on well site load, power grid operation status, backup power status, and critical load power supply status. The mobile energy storage equipment is used to collect energy storage state of charge, energy storage charging and discharging power, battery temperature and equipment operating status, and to perform energy storage charging, energy storage discharging, load impact compensation and power regulation according to the control commands issued by the edge control layer. The edge control layer is used to perform time synchronization, abnormal data processing, feature extraction and local storage of the collected data, and to perform load prediction, well site operation scenario identification, hourly basic charge and discharge plan generation, minute-level dynamic weighted multi-objective rolling optimization, second-level stroke phase synchronization compensation of pumping unit, rapid compensation of non-periodic impact load and state of charge safety verification, and to perform single-unit power allocation and state of charge balancing control when multiple mobile energy storage devices are running in parallel. The cloud management layer includes a data management module, an energy storage operating condition digital twin module, a model training and parameter management module, a remote monitoring module, and a remote control module; The communication network is used to enable data interaction between the end-side equipment, the edge control layer, and the cloud management layer.

9. The system according to claim 8, characterized in that, The energy storage operating condition digital twin module is used to establish the correlation between the operating conditions of the pumping unit, the well site load and the energy storage operation response. It compares the predicted load with the actual load, the energy storage control command with the actual power response and the target state of charge with the actual state of charge to obtain the load prediction deviation and the energy storage response deviation. The model training and parameter management module is used to train or correct the load prediction model and control parameters based on historical operating data, the load prediction deviation and the energy storage response deviation, and to send the verified model parameters and control parameters to the edge control layer. The edge control layer is used to verify the permissions, integrity, device status, and security constraints of the operating modes, scheduling parameters, model parameters, power limits, and remote control commands issued by the cloud management layer. When communication between the edge control layer and the cloud management layer is abnormal or interrupted, the edge control layer continues to execute well site energy storage control based on the locally stored prediction model, control parameters and the most recent valid configuration. When the primary communication link fails, the communication network switches to an available backup communication link; The system is equipped with identity authentication, data encryption, access control and operation auditing mechanisms, and a fault signal linkage and fault isolation mechanism between the battery compartment, power conversion compartment and control compartment of the mobile energy storage equipment.

10. A mobile energy storage device for well sites, comprising a container capable of being transported and hoisted as a whole, characterized in that, The mobile energy storage equipment is connected as an end-side device to the well site mobile energy storage digital management and control system based on edge-cloud collaboration, and performs charging, discharging and power regulation according to the energy storage power command generated by the multi-time-scale collaborative optimization scheduling method of the well site mobile energy storage system. The container is equipped with a battery compartment, a power conversion compartment, and a control compartment that are isolated from each other. The battery compartment is equipped with battery clusters, a battery management system, a thermal management device, and a fire-fighting device. The power conversion compartment is equipped with a bidirectional energy storage converter, AC / DC power distribution equipment, and grid-connected / off-grid switching equipment. The control cabin is equipped with an energy management controller, an edge controller, a communication gateway, and a human-machine interaction device; The battery compartment, power conversion compartment and control compartment are connected by power lines, control lines and communication bus, and are equipped with fireproof isolation devices, pressure relief devices, fire-fighting linkage devices and fault isolation devices; The outer side of the container body is provided with one or more of the following interfaces: AC interface, DC interface, diesel generator interface, photovoltaic interface, well site bus interface, communication interface, grounding interface, and fire linkage quick-connect interface. The interface is connected to the power conversion compartment through a circuit breaker, disconnect switch, or contactor, and is equipped with anti-misinsertion device and electrical interlock device. The edge controller is used to perform load prediction, well site operation scenario identification, hourly basic charge and discharge plan generation, minute-level dynamic weighted rolling optimization, second-level stroke phase synchronization compensation and state of charge safety verification based on pumping unit dynamometer data, well site load power data and energy storage system status data, generate energy storage power commands, and control the operation of the bidirectional energy storage converter according to the energy storage power commands. The edge controller and / or energy management controller are also used to control the grid-connected switching device to complete the connection of the well site power supply and distribution system after completing grounding confirmation, communication confirmation, insulation detection, and voltage, frequency and phase sequence verification. When over-temperature, over-current, insulation abnormality, smoke or communication failure is detected, alarm, power limitation, fault branch disconnection or fault compartment isolation are performed according to the fault level.

Images