Trinity architecture light storage and firewood intelligent power supply system under field environment

Abstract

This invention relates to the field of distributed generation technology and discloses a three-in-one photovoltaic-storage-diesel intelligent power supply system for outdoor environments. The system includes: a photovoltaic energy acquisition module, an energy storage scheduling and execution module containing independent energy storage units, a backup power control module, and a core scheduling module. The core scheduling module acquires the pose parameters of each module and constructs a physical distribution model. Based on this model, it analyzes geometric shading to determine the shadow feature matrix and the power allocation correction amount. Simultaneously, it identifies consistency differences among energy storage units based on state feature vectors containing temperature gradients and issues differential power commands. This invention, through dynamic perception and logical reconstruction of the random evolution of physical space topology, offsets balance deviations caused by support deformation and shading offsets, ensuring the microgrid system's adaptive disturbance rejection and power supply continuity under disordered outdoor conditions.

Description

Technical Field

[0001] This invention relates to a three-in-one intelligent power supply system for photovoltaic, energy storage, and diesel power generation in outdoor environments, belonging to the field of distributed power generation technology. Background Technology

[0002] Currently, in the field of distributed generation, off-grid microgrid systems composed of photovoltaics, energy storage, and diesel generator sets are used to provide power security for remote work sites such as mines and drilling sites. These systems rely on the fixed topology between photovoltaic arrays and energy conversion modules to maintain the stability of power dispatch. Field mobile operations are characterized by high-frequency migration, and equipment needs to be moved periodically with the entire camp. The geographical location, shading environment, and thermodynamic background after migration will undergo random changes. Existing control methods are based on the prior assumption of constant equipment topology and use preset electrical parameter thresholds to allocate power.

[0003] Changes in physical layout lead to nonlinear shifts in the irradiance shadows received by photovoltaic cells. Simultaneously, they cause inconsistencies in the intrinsic characteristics of energy storage units due to differences in thermal radiation environments. Linear improvements through increased hardware redundancy or increased power of individual components incur high deployment costs and cannot fundamentally eliminate uneven polarization of battery modules caused by spatial topology mismatch. Existing solutions lack the ability to dynamically perceive and reconstruct physical spatial parameters, causing a logical disconnect between power routing rules and the actual physical field distribution. Current technological improvements largely focus on enhancing the mechanical strength of the support structure or adding hardware redundancy; improvements solely at the hardware level are insufficient to address dynamically evolving environmental constraints. The control logic is not adaptable enough to sudden changes in physical field parameters. For example, Chinese invention patent application CN121367328A discloses a distributed photovoltaic-storage-diesel-charging microgrid collaborative control system, which distributes expected power data through cloud-based regional division. The underlying logic of this type of solution relies on stable cloud communication and a preset static distribution model of the region. In field conditions where there is a lack of external communication network and the physical topology is randomly reconstructed, the cloud planning-application execution architecture fails. This solution focuses on macro-level load scheduling and cannot perceive the local shading and thermal field distribution differences caused by changes in the placement and orientation of equipment, resulting in a disconnect between scheduling commands and the actual physical field state.

[0004] Therefore, how to achieve dynamic mapping between power dispatching logic and random physical topology, and solve the problem of control strategy inaccuracy caused by sudden layout changes in mobile scenarios, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A three-in-one intelligent power supply system for photovoltaic, energy storage, and diesel power generation in a field environment, comprising:

[0006] Photovoltaic energy harvesting module, used for solar energy conversion and mechanical posture adaptation;

[0007] The energy storage scheduling execution module includes two independently arranged energy storage skid-mounted units for energy storage and path routing;

[0008] The backup power control module is connected to the diesel generator set and is used for auxiliary power support.

[0009] The core scheduling module is communicatively connected to the photovoltaic energy acquisition module, the energy storage scheduling execution module, and the backup power control module. The core scheduling module includes a topology analysis unit, which reconstructs power allocation rules based on the random evolution of the camp's physical topology and performs the following steps: Step S101: Obtain the pose parameters of the photovoltaic energy acquisition module and the energy storage scheduling execution module, and construct a physical distribution model describing the spatial relationship between each module; Step S102: Perform geometric shading analysis based on the physical distribution model, obtain the shadow feature matrix of the shaded area of ​​the photovoltaic array, and convert the shadow feature matrix into a power allocation correction amount for energy storage scheduling; Step S103: Obtain the state feature vector containing the temperature gradient of each energy storage skid-mounted unit in the energy storage scheduling execution module, identify the electrochemical consistency differences of each energy storage skid-mounted unit based on the state feature vector, and issue asymmetric charging and discharging commands to the power conversion module corresponding to each energy storage skid-mounted unit.

[0010] Preferably, the photovoltaic energy acquisition module includes photovoltaic modules, a fixed base, and a photovoltaic panel extendable telescopic bracket; wherein, the fixed base is adapted to the camping roof by welding, the photovoltaic panel extendable telescopic bracket has a fixed tilt angle, and the telescopic end of the photovoltaic panel extendable telescopic bracket is connected to the photovoltaic module; the core scheduling module is used to collect real-time telescopic displacement and tilt angle data of the photovoltaic panel extendable telescopic bracket, and combine it with the preset solar geometric trajectory equation to complete the projection shading analysis, so as to output the power conversion path switching command.

[0011] Preferably, the energy storage scheduling execution module adopts a modular physical structure, wherein the energy storage skid unit integrates a lithium iron phosphate battery pack and a power conversion module, and the energy storage skid unit is equipped with hoisting and reinforcement nodes for overall relocation of the camp; the core scheduling module is used to monitor the operating parameters of each power conversion module and adjust the power weight components of each power conversion module in real time according to the state feature vector.

[0012] Preferably, when the core scheduling module operates step S101, it completes the following steps: Step S401, calling the built-in parameterized mechanical physical model and inputting the pose parameters as geometric constraints into the parameterized mechanical physical model; Step S402, calculating the relative spatial coordinates between the photovoltaic energy acquisition module and the energy storage scheduling execution module through coordinate transformation operators, and generating a logical entity that reflects the real-time physical deployment status of the camp.

[0013] Preferably, when the core scheduling module operates step S102, it completes the following steps: Step S501, based on logical entities, simulates the projection path of solar radiation at different times; Step S502, identifies the projection shadow area between the photovoltaic energy acquisition module and the existing buildings in the camp, and extracts the shadow feature matrix that reflects the shading area, shadow duration and dust deposition distribution pattern to compensate for the interference of physical topological changes on power output.

[0014] Preferably, when the core scheduling module operates step S103, it completes the following steps: Step S601, obtains the real-time state of charge, battery cell temperature, heat dissipation duct flow rate and shell temperature on the back side of each energy storage skid unit to construct a temperature gradient describing environmental thermal stress; Step S602, identifies battery clusters in high-temperature areas or limited heat dissipation environments based on the temperature gradient, and achieves intrinsic consistency compensation for energy storage units under different physical conditions by adjusting the charging and discharging power amplitude.

[0015] Preferably, the core scheduling module adjusts the node potential threshold during the charging and discharging process in real time according to the power allocation correction amount, so as to offset the power balance deviation caused by the sudden change in the physical distribution model at the control logic level and maintain the power balance of the microgrid in the field environment.

[0016] Preferably, the backup power control module is connected to the energy storage scheduling execution module and the critical load respectively; the core scheduling module is used to start the diesel generator set according to the prediction result of the shadow feature matrix when the predicted photovoltaic output drop value exceeds the support threshold of the energy storage scheduling execution module, so as to complete the energy supply path switching.

[0017] Preferably, step S602 specifically involves: calculating the power allocation coefficient μ of each power conversion module based on the temperature gradient, the calculation logic of which follows these rules: , where α is a preset thermal conductivity weighting factor, and ΔT is the temperature gradient difference between monitoring points inside the energy storage skid unit.

[0018] Preferably, the system also includes a spatial perception module, which has a built-in Beidou positioning unit and a gyroscope for acquiring pose parameters; the topology parsing unit is configured to complete the power supply scheduling modeling of the field three-dimensional space through local computing power in an environment without external communication network support.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. In the three-in-one architecture of photovoltaic-storage-diesel intelligent power supply, by introducing a parameterized topology rendering mechanism based on edge computing into the control logic, the power allocation strategy of distributed power sources can be deeply coupled with the random spatial layout of the field operation site. This effectively avoids the negative interference to energy output caused by photovoltaic array damage, local shading, and sudden changes in equipment placement due to frequent camp relocation. When the physical spatial topology changes, the system automatically reconstructs the three-dimensional logical model through the pose parameters obtained by local sensors, and maps the extracted spatial topology shading tensor into the correction factor of energy storage scheduling in real time. This offsets the energy field distortion caused by physical deformation at the logical level, ensuring the operational stability of the microgrid system in the disordered field environment.

[0021] 2. The asymmetric linkage between dual skid-mounted energy storage units and a regularized control strategy enables intrinsic state compensation for complex thermodynamic environments in the field, extending the service life of the battery pack and improving the accuracy of energy routing. The system utilizes intrinsic state vectors containing thermodynamic gradients to identify electrochemical consistency differences among the skid-mounted energy storage units under different geographical orientations and solar radiation intensities. Based on this, it issues asymmetric charge and discharge commands to each independent power conversion module, enabling battery clusters in high-radiation and high-heat-load areas to avoid local polarization overheating by dynamically adjusting discharge weights. This avoids the nonlinear degradation of individual cells caused by the one-size-fits-all static electrical thresholds in traditional microgrids.

[0022] 3. The mechanically expandable structure and the feedback regulation mechanism of the control subsystem work together to construct a distributed power supply closed loop with adaptive disturbance rejection capabilities. This ensures the continuity of power supply without relying on external remote communication support. The system feeds back the real-time extension and tilt data of the mechanical support to the edge scheduling module. Combined with the built-in solar geometric trajectory equation, ray projection calculations are performed. This ensures that the triggering of power switching commands no longer depends solely on voltage drop, but can seamlessly connect the power path in advance based on the predicted output changes. This predictive scheduling method based on physical state perception keeps the coordinated switching response time of the diesel generator and energy storage system within 20ms, meeting the stringent requirements of critical field operation loads for voltage fluctuations of around 3%. Attached Figure Description

[0023] Figure 1 This is a schematic diagram illustrating the integrated architecture and core scheduling principle of the photovoltaic-storage-diesel intelligent power supply system of the present invention.

[0024] Figure 2 This is a flowchart illustrating the power supply scheduling logic and dynamic power allocation reconfiguration process under field conditions according to the present invention.

[0025] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0027] A three-in-one intelligent power supply system integrating photovoltaic, energy storage, and diesel power generation for outdoor environments includes:

[0028] Photovoltaic energy harvesting module, used for solar energy conversion and mechanical posture adaptation;

[0029] The energy storage scheduling execution module includes two independently arranged energy storage skid-mounted units for energy storage and path routing;

[0030] The backup power control module is connected to the diesel generator set and is used for auxiliary power support.

[0031] The core scheduling module is communicatively connected to the photovoltaic energy acquisition module, the energy storage scheduling execution module, and the backup power control module. The core scheduling module includes a topology analysis unit, which reconstructs power allocation rules based on the random evolution of the camp's physical topology and performs the following steps: Step S101: Obtain the pose parameters of the photovoltaic energy acquisition module and the energy storage scheduling execution module, and construct a physical distribution model describing the spatial relationship between each module; Step S102: Perform geometric shading analysis based on the physical distribution model, obtain the shadow feature matrix of the shaded area of ​​the photovoltaic array, and convert the shadow feature matrix into a power allocation correction amount for energy storage scheduling; Step S103: Obtain the state feature vector containing the temperature gradient of each energy storage skid-mounted unit in the energy storage scheduling execution module, identify the electrochemical consistency differences of each energy storage skid-mounted unit based on the state feature vector, and issue asymmetric charging and discharging commands to the power conversion module corresponding to each energy storage skid-mounted unit.

[0032] Preferably, the photovoltaic energy acquisition module includes photovoltaic modules, a fixed base, and a photovoltaic panel extendable telescopic bracket; wherein, the fixed base is adapted to the camping roof by welding, the photovoltaic panel extendable telescopic bracket has a fixed tilt angle, and the telescopic end of the photovoltaic panel extendable telescopic bracket is connected to the photovoltaic module; the core scheduling module is used to collect real-time telescopic displacement and tilt angle data of the photovoltaic panel extendable telescopic bracket, and combine it with the preset solar geometric trajectory equation to complete the projection shading analysis, so as to output the power conversion path switching command.

[0033] Preferably, the energy storage scheduling execution module adopts a modular physical structure, wherein the energy storage skid unit integrates a lithium iron phosphate battery pack and a power conversion module, and the energy storage skid unit is equipped with hoisting and reinforcement nodes for overall relocation of the camp; the core scheduling module is used to monitor the operating parameters of each power conversion module and adjust the power weight components of each power conversion module in real time according to the state feature vector.

[0034] Preferably, when the core scheduling module operates step S101, it completes the following steps: Step S401, calling the built-in parameterized mechanical physical model and inputting the pose parameters as geometric constraints into the parameterized mechanical physical model; Step S402, calculating the relative spatial coordinates between the photovoltaic energy acquisition module and the energy storage scheduling execution module through coordinate transformation operators, and generating a logical entity that reflects the real-time physical deployment status of the camp.

[0035] Preferably, when the core scheduling module operates step S102, it completes the following steps: Step S501, based on logical entities, simulates the projection path of solar radiation at different times; Step S502, identifies the projection shadow area between the photovoltaic energy acquisition module and the existing buildings in the camp, and extracts the shadow feature matrix that reflects the shading area, shadow duration and dust deposition distribution pattern to compensate for the interference of physical topological changes on power output.

[0036] Preferably, when the core scheduling module operates step S103, it completes the following steps: Step S601, obtains the real-time state of charge, battery cell temperature, heat dissipation duct flow rate and shell temperature on the back side of each energy storage skid unit to construct a temperature gradient describing environmental thermal stress; Step S602, identifies battery clusters in high-temperature areas or limited heat dissipation environments based on the temperature gradient, and achieves intrinsic consistency compensation for energy storage units under different physical conditions by adjusting the charging and discharging power amplitude.

[0037] Preferably, the core scheduling module adjusts the node potential threshold during the charging and discharging process in real time according to the power allocation correction amount, so as to offset the power balance deviation caused by the sudden change in the physical distribution model at the control logic level and maintain the power balance of the microgrid in the field environment.

[0038] Preferably, the backup power control module is connected to the energy storage scheduling execution module and the critical load respectively; the core scheduling module is used to start the diesel generator set according to the prediction result of the shadow feature matrix when the predicted photovoltaic output drop value exceeds the support threshold of the energy storage scheduling execution module, so as to complete the energy supply path switching.

[0039] Preferably, step S602 specifically involves: calculating the power allocation coefficient μ of each power conversion module based on the temperature gradient, the calculation logic of which follows these rules: , where α is a preset thermal conductivity weighting factor, and ΔT is the temperature gradient difference between monitoring points inside the energy storage skid unit.

[0040] Preferably, the system also includes a spatial perception module, which has a built-in Beidou positioning unit and a gyroscope for acquiring pose parameters; the topology parsing unit is configured to complete the power supply scheduling modeling of the field three-dimensional space through local computing power in an environment without external communication network support.

[0041] Example 1: In a drilling camp located deep in the mountains, due to the undulating terrain and limited space of the working face, the campsites integrating the power supply system of this invention and the heavy drilling equipment are irregularly and scatteredly distributed. When the camp faces frequent relocation every three months, the relative posture between the mechanically extended photovoltaic support subsystem and the drilling rig derrick changes randomly after each redeployment. This random evolution of the physical spatial topology causes the photovoltaic energy harvesting module to suffer irregular dynamic projection shading from the derrick during daytime operation. At the same time, the two energy storage skid-mounted units, located in low-lying sheltered areas and on windward ridges, experience electrochemical inconsistencies due to differences in thermal radiation in the local microenvironment. A fixed electrical threshold is used. The energy distribution method at this time cannot perceive the geometric distortion of the physical space, which leads to the risk of hot spot damage to the photovoltaic modules in the shaded area. In addition, the energy storage battery clusters under different thermal stress environments will experience nonlinear life decay, threatening the power supply continuity of critical loads such as drilling mud pumps. When the system starts under the above conditions, the topology analysis unit in the core scheduling module obtains the real-time extension displacement and tilt angle data of the mechanically extended photovoltaic support subsystem and the absolute spatial azimuth angle parameters of the two energy storage skid units. It calls the built-in parameterized mechanical physical model and uses it as the geometric constraint input. The relative spatial coordinates between each subsystem are calculated through the coordinate transformation operator to generate a logical entity that reflects the real-time physical deployment status of the camp.

[0042] Based on this, the core scheduling module combines the preset solar geometric trajectory equation to perform periodic light projection analysis and extract the shadow feature matrix projected onto the photovoltaic module surface at the current time node in situ. Simultaneously, read the state feature vectors of each energy storage unit, including the temperature gradient. And reconstruct the formula based on the lower limit of scheduling. To ensure the reproducibility of the conversion between spatial geometric shading and internal thermal stress into electrical correction quantities, the core scheduling module incorporates a parameter quantization procedure based on the equivalent output model of the photovoltaic array and Newton's law of cooling, respectively. The topology analysis unit divides the physical plane of the photovoltaic array into 10 rows and 10 columns, totaling 100 logical sampling grids. For each grid point in the shadow feature matrix, a binary judgment of 0 and 1 is performed. A grid is judged as 1 if its shading area exceeds 50%, otherwise as 0. The photovoltaic output attenuation function is obtained by accumulating the total number of grid points with a value of 1 and dividing by the total number of 100. This function value is then multiplied by a photovoltaic shading compensation weight of 0.15 as the increment for the preset threshold of the correction. Simultaneously, a separate thermal conductivity weight gain is defined when calculating the power distribution coefficient. This gain is obtained by measuring the temperature rise slope of the energy storage unit maintained at 100% rated discharge power for 10 minutes, with an initial calibration value of 0.12. The topology analysis unit then performs a binary judgment of the shadow feature matrix... Perform matrix trace operation to extract the absolute scalar of the shaded area, calculate the ratio of the scalar to the total physical area of ​​the photovoltaic array, and directly assign it as the photovoltaic power output attenuation function. The core scheduling module extracts state feature vectors. It includes the highest single-cell temperature of the current battery cluster, calculates the absolute value of the difference between this and the environmental benchmark temperature, and assigns it as a normalized value to characterize the thermal offset correction function of the energy storage unit. The function values ​​obtained through objective parameter calculations are all used as dimensionless constraint factors in the control logic. The lower limit threshold for the reconstructed scheduling. Based on preset threshold, The shadow feature matrix, For state feature vectors, The photovoltaic power output attenuation function is extracted from the shadow feature matrix. The correction function characterizes the thermal offset of the energy storage unit, where α and β are preset mapping coefficients. When the real-time monitored energy storage parameters touch upon the dynamically evolving... At the boundary, the core scheduling module issues an asymmetric charging / discharging command to the power conversion module corresponding to the energy storage scheduling execution module. This command is encapsulated in a 16-bit controller area network bus message data segment. The high 8 bits represent the power allocation coefficient, mapping the calculated power weight components to integer values ​​between 0 and 255, corresponding to the duty cycle adjustment range of the power conversion module's pulse width modulation signal from 0% to 100%. The low 8 bits serve as a checksum and current direction control bit. If the current direction bit is 0, charging mapping is executed; if it is 1, discharging mapping is executed. The power weight components are calculated... To adjust the energy flow direction, where μ is the power weight component, λ is the preset adjustment coefficient, and ΔT is the temperature gradient difference between monitoring points inside the energy storage skid unit.

[0043] This utilizes local edge computing power to offset physical field distortions. When facing power balance deviations caused by the aforementioned physical topology abrupt changes, the power supply system coordinates with the dynamic mapping logic of the core scheduling module through mechanical pose feedback to transform output fluctuations caused by shadow occlusion into predictive switching of the energy supply path. The diesel generator set's mechanical rotation and pressure-building process inherently has a physical time constant of several seconds. To smooth out the spatiotemporal scale differences between electromagnetic transient response and mechanical inertial action, a supercapacitor transient buffer medium is physically integrated into the DC bus circuit of the energy storage scheduling execution module. When the predictive intervention conditions are met, the power conversion module controls the release of electric field energy within the buffer medium, providing millisecond-level transient reactive and active power output. This fills the gap in the time window before the backup mechanical power supply is simultaneously connected to the grid in the objective energy transfer chain. In actual operation, when the shadow feature matrix... Before the predicted drop in photovoltaic output reaches the support threshold, the backup power control module triggers the power support logic of the diesel generator set 20ms in advance, limiting the voltage fluctuation of critical operating loads to within 3%. Due to the completion of consistent compensation for energy storage units under different physical conditions, the equipment loss rate of the system after four high-frequency relocations remains below 1%, and the relocation and deployment period is reduced from 5 days to 1 day. Ultimately, the operational stability of the distributed generation system in disordered outdoor spaces is no longer subject to the randomness of equipment placement.

[0044] Example 2: On a verification platform equipped with a distributed generation physical simulation system, an experimental environment simulating high-frequency migration conditions in the field was constructed to verify the dynamic response performance and power supply continuity indicators of the core scheduling module under physical space topological abrupt changes. This verification platform includes a 50kW photovoltaic energy acquisition module, two 100kWh energy storage skid-mounted units, and an inductive load unit simulating the characteristics of a 100kW drilling mud pump. Data acquisition was completed using a physical board with a 16-bit analog-to-digital conversion depth and a sampling frequency set to 10kHz, ensuring the original telecommunications... The sampling accuracy is better than 0.1%. The sampling period of key parameters in the experiment is set according to the Nyquist sampling theorem and the constraints of edge computing power. It is weighed between 50ms and 200ms. In order to capture the spatial gradient evolution when the projection edge sweeps across the photovoltaic module at a speed of 0.5m / s, the sampling period is determined to be 100ms. The mapping coefficients α and β are derived from the internal resistance consistency calibration experiment of the energy storage unit under different thermal field distributions. By measuring the voltage drop deviation of the single cell under a temperature gradient of 10K / cm, the value of α is determined to be 0.12 and the value of β is 0.08.

[0045] To simulate non-ideal factors in the field environment, additive white Gaussian noise with a signal-to-noise ratio of 25dB was actively injected into the data input, and the dynamic shading projection of the drilling rig onto the photovoltaic array was simulated. During the baseline setting phase, the irradiance was maintained at 800W / m². 2 The initial state of charge (SOC) of both energy storage skid-mounted units was adjusted to 60%. The experiment was divided into an experimental group based on the present invention and a control group. The control group used a fixed SOC limit of 20% as the scheduling trigger threshold, while the experimental group based on the present invention used a dynamic threshold scheme reconstructed by a topology rendering engine. In the core derivation phase, the physical pose of the simulated derrick was adjusted to make the shadow feature matrix... The shading ratio was increased from 0% to 45% in 5% increments. When the shading ratio was at a low intensity gradient of 15%, the bus voltage fluctuation rate of the comparison sample was 1.45%, while that of the present invention sample was 1.12%. When the shading ratio increased to a medium intensity gradient of 30%, the bus voltage fluctuation rate of the comparison sample rose to 4.82% because it could not perceive the physical field distortion, accompanied by an abnormal temperature rise of 3.6K in some photovoltaic modules due to current mismatch. At this time, the present invention sample reconstructed the system by calling the scheduling lower limit formula. The current scheduling lower limit threshold is calculated. The voltage fluctuation rate was 24.8%, and an asymmetric charging and discharging command was issued to the power conversion module to maintain the bus voltage fluctuation rate at 1.65% and control the local temperature rise below 1.1K. The lower limit threshold for the reconstructed scheduling. Based on preset threshold, The shadow feature matrix, For state feature vectors, The photovoltaic power output attenuation function is extracted from the shadow feature matrix. The correction function characterizes the thermal offset of the energy storage unit, with α and β being preset mapping coefficients.

[0046] Entering the performance inflection point verification stage under high-intensity shading gradients, when the shading ratio exceeds 35%, the photovoltaic output exhibits a non-linear drop. The comparative sample, lacking prediction of physical topology evolution, experienced a 165.4ms response lag in energy routing, causing the bus voltage to drop to 472.8V and triggering system undervoltage protection. The present invention's sample, however, detected the shading feature matrix... Within 18.5ms of the indicator reaching the critical value, the core scheduling module calculates the power weight component. The bias correction is supported by the backup power control module guiding the diesel generator set to the grid, and the bus voltage is maintained at 515.2V. Here, μ is the power weight component, λ is the preset adjustment coefficient, and ΔT is the temperature gradient difference between monitoring points inside the energy storage skid unit. The system frequency data of the entire process from shading ratio to 45% is statistically analyzed. The standard deviation of frequency fluctuation of the sample group of the present invention is maintained at 0.08Hz, while the standard deviation of frequency fluctuation of the comparison sample group rises to 0.42Hz after the shading ratio exceeds 30%. The above measured data show that the present invention, through dynamic perception and logical reconstruction of physical space topology, limits the voltage fluctuation of key loads to within 3% under the condition of environmental noise and physical shading interference, and solves the problem of power distribution imbalance caused by geometric distortion in the disordered field application of distributed generation system.

[0047] Example 3: In the operation site of a distributed generation system requiring high-precision control of energy flow paths, the topology analysis unit determines the spatial relationship between the photovoltaic energy acquisition module and the energy storage scheduling execution module based on the following physical mapping path; the parameterized mechanical physical model called by the topology analysis unit uses the geometric dimensions of the photovoltaic panel extendable telescopic support as boundary constraints, and sets the telescopic length... The value range is limited to 0m to 3.5m, and a fixed inclination angle is set. The design value is fixed at 25.5 degrees. After system startup, the topology analysis unit acquires real-time pose parameters via a rope-type displacement sensor installed at the end of the telescopic mechanism, transmits real-time attitude data using an inclination sensor installed on the support beam, and calls the coordinate transformation operator to determine the three-dimensional offset vector of the photovoltaic module's center point relative to the fixed base. The calculation logic of the coordinate transformation operator includes: based on the real-time telescopic displacement... The system calculates the displacement components of the photovoltaic module's center point in the vertical plane using a fixed tilt angle θ. Combined with the absolute spatial azimuth angle ϕ of the energy storage skid-mounted unit collected by the spatial sensing module, a three-dimensional rectangular coordinate system is established. The coordinates (x, y, z) of the photovoltaic module's center point are defined as a spatial vector relative to the camp's physical origin, thereby generating a logical entity reflecting the camp's real-time physical deployment status. To determine key parameters in scheduling decisions, the system employs a calibration procedure based on dynamic load characteristic analysis to obtain basic preset thresholds. The calibration procedure includes: when inductive loads such as drilling mud pumps are connected to the busbar, the busbar voltage is continuously sampled at a frequency of 10kHz using a data acquisition unit to obtain an original voltage sequence containing 2000 sampling points; secondly, the voltage drop depth ΔU under load surge conditions is calculated, and the root mean square algorithm is used to evaluate the transient support capability of the system under different charge states; when the measured busbar voltage fluctuation rate reaches the 3% stability threshold, the corresponding energy storage unit charge state value is recorded and set as the basic preset threshold. In the specific application of this embodiment, the calibration points are... The value was determined to be 22.4%.

[0048] When determining the adaptive logic for the adjustment coefficient λ, the core scheduling module sets λ as a function operator that is monotonically correlated with the temperature gradient difference ΔT, based on the internal thermal field distribution of the energy storage skid-mounted unit. This operator is calculated to balance the electrochemical response rate and thermal stress risk. When the temperature gradient difference ΔT between internal monitoring points is in the range of 0.5 K / cm to 1.5 K / cm, the adjustment coefficient λ is selected as 0.05 to maintain linear adjustment of the power weight, ensuring that the power allocation matches the thermal load offset. If the real-time monitored temperature gradient difference ΔT exceeds the performance inflection point of 2.5 K / cm, the core scheduling module identifies that the energy storage unit has entered a thermal resistance state. In the nonlinear growth region, the calculation operator of the power weight component μ automatically increases the amplitude of the adjustment coefficient λ to 0.15, which produces a suppression effect on the discharge power of the energy storage skid unit and limits the polarization damage of the battery cluster. The above-mentioned mapping method for pose parameters and calibration process for scheduling thresholds eliminate the logical uncertainty of the system under the condition of disordered deployment in the field, enabling personnel in the relevant technical field to reproduce the adaptive compensation process of distributed generation system for physical field distortion based on the published geometric constraints and calibration steps. Under complex load fluctuations, the response delay of the system bus voltage is kept within 20ms, realizing the closed-loop alignment of the power supply system physical topology and energy logic.

[0049] Example 4: When the system faces the initial deployment after the overall relocation of the camp, the parameterized physical model invoked by the core scheduling module determines the alignment benchmark between the logical space and the physical space through a pre-calibration process. This calibration process collects the camp's geographical coordinates and magnetic north azimuth data through the spatial perception module, aligns and maps the rotation axis of the photovoltaic energy acquisition module with the geographical latitude and longitude grid, thereby determining the initial offset vector of the coordinate transformation operator. ,in, A vector operator characterizing the deviation of the physical origin is used; the topology analysis unit controls the mechanically extended photovoltaic support subsystem to generate full-stroke telescopic displacement under no-load conditions, collects sensor feedback electrical signals at calibration points of 0.5m, 1.5m and 3.0m, and uses linear interpolation logic to construct a mapping function between displacement components and electrical signals, thereby eliminating geometric system errors introduced by field installation posture deviations.

[0050] After completing the initial pose alignment, the core scheduling module compares the real-time acquired open-circuit voltage of the photovoltaic module with the value obtained from the shadow feature matrix. The calculated predicted power values ​​are used to verify the matching degree of the physical distribution model. When the deviation rate between the predicted and measured values ​​exceeds the preset tolerance boundary of 5%, the topology analysis unit initiates the parameter fine-tuning procedure, adjusting the fixed tilt angle θ parameter in the model in 0.1-degree increments until the deviation rate converges to within 2%, correcting the power deviation caused by the local terrain shading effect. In addition, the backup power control module, based on the output characteristics of the diesel generator set under different altitude environments, collects intake pressure sensor data and combines it with the power correction formula. The trigger point for reconstructing the backup support logic, among which, The corrected output power, The rated power is γ, the altitude attenuation coefficient is γ, and the altitude deviation is ΔH. This pre-calibration process eliminates the interference of the field geographical environment on the control closed loop, enabling the system to maintain the correspondence between physical parameters and control commands in environments without communication coverage.

[0051] Example 5: When the system is in a high-albedo operating environment deep in a valley, interference from secondary reflected light from the surrounding rock walls and signal drift from the pose sensor in the spatial perception module due to long-term mechanical vibration cause the real-time pose parameters obtained by the topology analysis unit to deviate from the physical reference. This perception error causes the logical entities generated by the parametric topology rendering engine to be misaligned with the actual geometric topology, thereby triggering the scheduling lower limit reconstruction formula. The calculation deviation caused the system to experience power regulation oscillations when the cloud layer swept across rapidly, affecting the voltage quality of the distributed generation system and the operational life of the energy storage subsystem.

[0052] The core scheduling module utilizes an online closed-loop self-calibration program to calibrate the pose sensor. This program configures a circular buffer with a capacity of 1000 sampling points at the control chip's underlying layer, acquiring the raw voltage signals from the gyroscope and displacement sensor at a frequency of 10kHz. A sliding window calculation is performed every 50ms with a 50% overlap rate. Within the window, a 20-point median filtering algorithm is executed to eliminate high-frequency mechanical vibration interference above 50Hz generated by drilling rig operation, thereby identifying the cumulative drift component of the pose sensor. The program extracts the short-circuit current envelope of the photovoltaic energy acquisition module at the moment of highest daytime irradiance. By comparing the solar spatial azimuth angle corresponding to the measured current peak moment with the predicted angle calculated by the coordinate transformation operator, the cumulative drift component Δσ of the pose sensor is identified, where Δσ is the cumulative angle error offset. The topology parsing unit corrects the geometric transformation matrix of the rendering engine based on the drift component and recalibrates the shadow feature matrix. The mapping weights are determined, and to address high albedo interference, the backup power control module collects the radiation intensity of the auxiliary photosensitive component in a specific spectral frequency band and constructs a reflected light interference suppression model. The power component generated by reflected light is then decoupled from the photovoltaic output attenuation function. When the real-time bus voltage deviation converges to within 0.5% within three consecutive sampling periods, the self-calibration program corrects the physical space topology logic entity, and the frequency fluctuation range of the system bus voltage is maintained within 0.05Hz.

[0053] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A three-in-one architecture light storage and firewood intelligent power supply system in a field environment, characterized in that, include: Photovoltaic energy harvesting module, used for solar energy conversion and mechanical posture adaptation; The energy storage scheduling execution module includes two independently arranged energy storage skid-mounted units for energy storage and path routing. The backup power control module is connected to the diesel generator set and is used for auxiliary power support. The core scheduling module is communicatively connected to the photovoltaic energy acquisition module, the energy storage scheduling execution module, and the backup power control module. The core scheduling module includes a topology analysis unit, which reconstructs power allocation rules based on the random evolution of the camp's physical topology and performs the following steps: Step S101: Obtain the pose parameters of the photovoltaic energy acquisition module and the energy storage scheduling execution module, and construct a physical distribution model describing the spatial relationship between each module; Step S102: Perform geometric shading analysis based on the physical distribution model, obtain the shadow feature matrix of the shaded area of ​​the photovoltaic array, and convert the shadow feature matrix into a power allocation correction amount for energy storage scheduling; Step S103: Obtain the state feature vector containing the temperature gradient of each energy storage skid-mounted unit in the energy storage scheduling execution module, identify the electrochemical consistency differences of each energy storage skid-mounted unit based on the state feature vector, and issue asymmetric charging and discharging commands to the power conversion module corresponding to each energy storage skid-mounted unit.

2. The three-in-one architecture light storage and firewood intelligent power supply system in a field environment according to claim 1, characterized in that, The photovoltaic energy acquisition module includes photovoltaic modules, a fixed base, and a photovoltaic panel extendable telescopic support. The fixed base is adapted to the camping roof by welding, and the photovoltaic panel extendable telescopic support has a fixed tilt angle, with the telescopic end of the support connected to the photovoltaic module. The core scheduling module is used to collect real-time telescopic displacement and tilt angle data of the photovoltaic panel extendable telescopic support, and combine it with the preset solar geometric trajectory equation to complete the projection shading analysis, so as to output the power conversion path switching command.

3. The three-in-one architecture light storage firewood intelligent power supply system in a field environment according to claim 1, characterized in that, The energy storage scheduling execution module adopts a modular physical structure. The energy storage skid unit integrates a lithium iron phosphate battery pack and a power conversion module. The exterior of the energy storage skid unit is equipped with hoisting and reinforcement nodes for the overall relocation of the camp. The core scheduling module is used to monitor the operating parameters of each power conversion module and adjust the power weight components of each power conversion module in real time according to the state feature vector.

4. The three-in-one architecture light storage firewood intelligent power supply system in a field environment of claim 1, wherein, When the core scheduling module operates in step S101, it completes the following steps: Step S401, it calls the built-in parameterized mechanical physical model and inputs the pose parameters as geometric constraints into the parameterized mechanical physical model; Step S402, it calculates the relative spatial coordinates between the photovoltaic energy acquisition module and the energy storage scheduling execution module through the coordinate transformation operator, and generates a logical entity that reflects the real-time physical deployment status of the camp.

5. The three-in-one architecture light storage-chip intelligent power supply system in a field environment according to claim 1, characterized in that, When the core scheduling module operates in step S102, it completes the following steps: Step S501, based on logical entities, simulates the projection path of solar radiation at different times; Step S502, identifies the projection shadow area between the photovoltaic energy acquisition module and the existing buildings in the camp, and extracts the shadow feature matrix that reflects the shading area, shadow duration and dust deposition distribution pattern to compensate for the interference of physical topological changes on power output.

6. The three-in-one architecture light storage firewood intelligent power supply system in a field environment of claim 1, wherein, When the core scheduling module operates in step S103, it completes the following steps: Step S601, it acquires the real-time state of charge, cell temperature, heat dissipation duct flow rate, and shell temperature on the back side of each energy storage skid unit to construct a temperature gradient describing environmental thermal stress; Step S602, it identifies battery clusters in high-temperature areas or limited heat dissipation environments based on the temperature gradient, and achieves intrinsic consistency compensation for energy storage units under different physical conditions by adjusting the charging and discharging power amplitude.

7. The integrated photovoltaic, energy storage and firewood intelligent power supply system of claim 1, wherein, The core scheduling module adjusts the node potential threshold during the charging and discharging process in real time according to the power allocation correction amount, so as to offset the power balance deviation caused by the sudden change in the physical distribution model at the control logic level and maintain the power balance of the microgrid in the field environment.

8. The three-in-one architecture light storage firewood intelligent power supply system in a field environment of claim 1, characterized in that, The backup power control module is connected to the energy storage scheduling execution module and the critical load respectively; the core scheduling module is used to start the diesel generator set according to the prediction results of the shadow feature matrix when the predicted photovoltaic output drop value exceeds the support threshold of the energy storage scheduling execution module, so as to complete the energy supply path switching.

9. The three-in-one architecture light storage firewood intelligent power supply system in a field environment of claim 6, wherein, The step S602 specifically comprises: calculating the power distribution coefficient μ of each power conversion module according to the temperature gradient, and the calculation logic follows the following rules: wherein, α is a preset heat conduction weight factor, and ΔT is the difference of the temperature gradient between the internal monitoring points of the energy storage skid-mounted unit.

10. The integrated photovoltaic, energy storage and firewood intelligent power supply system of claim 1, wherein, The system also includes a spatial perception module, which has a built-in Beidou positioning unit and a gyroscope to acquire pose parameters; the topology resolution unit is configured to complete the power supply scheduling modeling of the field three-dimensional space through local computing power in an environment without external communication network support.

Images