Mobile emergency DC power supply intelligent dispatching system

Abstract

The application relates to the technical field of power emergency guarantee and intelligent scheduling, and particularly discloses a mobile emergency DC power supply intelligent scheduling system, which acquires static ability elements and dynamic ability elements of a power supply unit in real time through a heterogeneous data acquisition module, constructs a time-space ability field, dynamically deduces the strength attenuation and spatial migration of the ability field by using a time-space convolution kernel, and realizes accurate prediction of a future power supply ability coverage range; a task matching and decision generation module generates a comprehensive gravity weight by combining power demand and time urgency of a task point, calculates cumulative power supply contribution degrees of each power supply unit to the task point within a specified time window, and generates an optimal scheduling scheme by a multi-objective optimization algorithm with the global cumulative strength maximization and mobile cost minimization as targets; and an instruction execution and state synchronization module encapsulates scheduling instructions and resource state updates as atomic transactions, so that the reliability of instruction execution and the consistency of system state are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power emergency guarantee and intelligent scheduling, and in particular to a mobile emergency DC power intelligent scheduling system. BACKGROUND

[0002] As important emergency power supply guarantee equipment, mobile emergency DC power plays a key role in scenarios such as power grid failure, natural disaster rescue, and power supply for important activities. The traditional scheduling management mode mainly relies on manual experience and semi-automated information systems. The scheduling center obtains limited power supply state information (such as location and power) through a monitoring platform. When an emergency event occurs, the scheduling personnel need to contact and assign the power supply vehicles that are closest to the fault point and appear to be usable based on personal experience through communication means such as telephone and intercom.

[0003] The prior art has the following disadvantages:

[0004] The traditional emergency power scheduling system cannot model and predict the space-time accessibility and power supply capacity decay of mobile DC power units in dynamic and uncertain emergency scenarios, resulting in serious lag or inaccuracy in scheduling decisions. Existing systems are based on rough matching of static location and power information, ignoring key dynamic factors such as power consumption, health status changes, and power supply capacity decay with distance during the movement of power units, which can easily cause resource mismatching, making it difficult to supply power in practice despite the nominal coverage. When the power unit arrives at the task point, it may run out of power, be insufficient, or respond too slowly, severely weakening the reliability and timeliness of emergency power supply. SUMMARY

[0005] The present application relates to the technical field of power emergency guarantee and intelligent scheduling, and in particular to a mobile emergency DC power intelligent scheduling system.

[0006] The object of the present application can be achieved by the following technical solutions:

[0007] The mobile emergency DC power intelligent scheduling system comprises:

[0008] A heterogeneous data acquisition module acquires static capacity factors and dynamic capacity factors of each power supply unit in real time. The static capacity factors include the rated capacity and inherent physical properties of the power supply unit. The dynamic capacity factors include the real-time spatial position, real-time remaining power, and real-time health status of the power supply unit.

[0009] A space-time capability field construction module constructs a space-time capability field for each power supply unit based on static capability elements and dynamic capability elements; the space-time capability field is used to represent the effective coverage range of the power supply capability of the power supply unit in space and the attenuation law with time and space movement in a specific time period in the future;

[0010] A task matching and decision generation module calculates the cumulative intensity of any task point covered by the capability field of one or more power supply units in a specified time window based on the space-time capability field of each power supply unit; a global scheduling decision scheme is generated with the optimization goal of maximizing the total cumulative intensity of all task points; the global scheduling decision scheme explicitly specifies the execution order and movement path of the scheduled power supply unit;

[0011] An instruction execution and state synchronization module issues scheduling instructions to the corresponding power supply unit according to the global scheduling decision scheme, and synchronously updates the space-time capability field state of the power supply unit, marking its power supply capability as an allocated state.

[0012] As a further scheme of the present application: the construction process of the space-time capability field is:

[0013] The initial intensity of the capability field is determined based on the rated capacity in the static capability elements, and the initial intensity is corrected based on the real-time health status in the dynamic capability elements;

[0014] The effective coverage range of the power supply capability of the power supply unit in space is divided into three concentric annular regions of core region, effective region and extension region with the real-time spatial position of the power supply unit as the center, and different capability intensity coefficients are assigned to each region, wherein the intensity coefficient of the core region is the highest and the intensity coefficient of the extension region is the lowest;

[0015] The corrected capability field intensity generates a space-time capability field that can be visualized on an electronic map, with the intensity gradually decaying from the center to the edge.

[0016] As a further scheme of the present application: the deduction process of the space-time capability field includes:

[0017] A space-time convolution kernel is defined, and the parameters of the space-time convolution kernel are determined by the real-time residual power consumption rate in the dynamic capability elements and the average movement speed of the power supply unit;

[0018] The space-time capability field map at the current time is convolved using the space-time convolution kernel to simulate the dynamic process of spatial position movement and overall intensity attenuation of the space-time capability field with the movement of the power supply unit and the consumption of power in a future period of time;

[0019] A set of space-time capability field snapshots arranged in a future time sequence are output by continuously performing convolution operations, which are used to predict the schedulable capability of the power supply unit at future time points.

[0020] As a further scheme of the present application: the process of obtaining the intensity coefficient is:

[0021] According to the preset mapping relationship, an initial reference intensity coefficient is allocated to the core area, the effective area and the extended area respectively, wherein the reference intensity coefficient of the core area is set as the maximum value;

[0022] Based on the ratio of the real-time residual power to the rated capacity in the dynamic capability element, an electric quantity attenuation factor is calculated, and the reference intensity coefficient is scaled by the electric quantity attenuation factor;

[0023] Based on the real-time health state value in the dynamic capability element, a health impairment factor is calculated, and the intensity coefficient after the electric quantity attenuation correction is multiplied by the factor to obtain the actual capability intensity coefficient of each area;

[0024] The actual capability intensity coefficient of each area calculated is bound to the corresponding annular area, and the quantitative construction of the internal structure of the space-time capability field is completed.

[0025] As a further scheme of the present application: the process of calculating the cumulative intensity includes:

[0026] According to the power demand and time urgency of the emergency task, the emergency degree weight and power level weight of each task point are calculated, and the two are fused into a comprehensive gravity weight;

[0027] For the capability field of a power supply unit, the capability intensity value at the geographical location of the task point within the time window required by the task is calculated, and the capability intensity value is multiplied by the comprehensive gravity weight of the task to obtain the single contribution degree of the power supply unit to the task point;

[0028] For the same task point, the single contribution degrees of all power supply units to the task point are linearly superimposed to obtain the final cumulative intensity of the corresponding task point.

[0029] As a further scheme of the present application: the process of calculating the capability intensity value is:

[0030] The path and time window of moving the power supply unit from the current position to the task point planned by the task point are combined to determine the space-time projection sequence of the capability field at the task point within the time window;

[0031] The theoretical capability intensity on the space-time projection sequence is weighted and integrated;

[0032] The result of the weighted integration is divided by the length of the time window to obtain an average effective intensity value representing the power supply unit to the task point within the task time window, and the capability intensity value is obtained.

[0033] As a further scheme of the present application, the generating the global scheduling decision scheme specifically comprises:

[0034] Randomly generating a group of initial scheduling decision links containing different power supply unit and task point matching relations and execution sequences, to form an initial population;

[0035] Iteratively optimizing the population, in each generation, performing cross and mutation operations on the decision links, and introducing a conflict detection and resolution operator, which detects and corrects path overlap, resource over-allocation and time conflict;

[0036] Setting a composite optimization function with the goal of maximizing total cumulative intensity and minimizing total scheduling path, after iteration termination, selecting the optimal decision scheme located on the Pareto frontier from the final population as the global scheduling decision scheme.

[0037] As a further scheme of the present application, the updating the space-time capability field state of the power supply unit specifically comprises:

[0038] Before issuing the scheduling instruction, creating a current state snapshot of the space-time capability field of the affected power supply unit and generating a unique transaction identifier;

[0039] Encapsulating the issuance of the scheduling instruction and the update of the space-time capability field as an atomic transaction;

[0040] In the atomic transaction, setting the intensity value of the space-time capability field of the scheduled power supply unit to zero at the current time and in the future task duration, so as to immediately exclude it from the available resource pool.

[0041] As a further scheme of the present application, the allocated state further includes an enhanced step to ensure the eventual consistency of the system state:

[0042] Waiting for the instruction confirmation signal returned by the execution end of the power supply unit, and only after receiving the confirmation signal, the temporary state allocated in the atomic transaction is solidified into a persistent state;

[0043] If the instruction confirmation signal is not received within the preset timeout time, it is determined that the instruction execution is abnormal, and a rollback operation is automatically triggered to cancel the allocated mark of the space-time capability field of the power supply unit and restore its intensity to the transaction snapshot state;

[0044] After the rollback is completed, the system verifies the consistency of the space-time capability field state and the actual state of all power supply units, and if a deviation is found, an alarm is generated and a manual intervention process is started.

[0045] The present application has the following beneficial effects:

[0046] (1) The application realizes the dynamic prediction of the future power supply capacity coverage of the power supply unit by constructing a "spatiotemporal capacity field" model integrating the real-time power, health status, moving speed and power supply capacity attenuation with distance of the power supply unit, overcomes the resource mismatch problem caused by the dependence on static information in the traditional scheduling, ensures that the scheduling decision is more scientific and reliable, and effectively avoids the situation that the power supply unit cannot complete the task due to power consumption or response timeout.

[0047] (2) The application introduces an atomic operation mechanism based on state snapshot and transaction identification in the instruction execution link, binds the scheduling instruction issuing and resource state updating as an indivisible transaction, and has an abnormal rollback function. When communication fails or the device does not respond, it can automatically recover to the state before scheduling to prevent resource duplication allocation; at the same time, through consistency verification comparison between the system and field data, it can timely find and alarm the abnormality, and ensures the robustness of the scheduling process and the accurate synchronization of the global state of the system. BRIEF DESCRIPTION OF DRAWINGS

[0048] The application will be further described below with reference to the drawings.

[0049] Figure 1 is a flow chart of the system of the application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0051] Please refer to Figure 1 The application is a mobile emergency DC power intelligent scheduling system, which comprises:

[0052] A heterogeneous data acquisition module acquires the static capacity elements and dynamic capacity elements of each power supply unit in real time. The static capacity elements include the rated capacity and inherent physical properties of the power supply unit. The dynamic capacity elements include the real-time spatial position, real-time residual power and real-time health status of the power supply unit.

[0053] A spatiotemporal capacity field construction module constructs a spatiotemporal capacity field for each power supply unit based on the static capacity elements and dynamic capacity elements. The spatiotemporal capacity field is used to represent the effective coverage range of the power supply capacity of the power supply unit in space and the attenuation law with time and space movement within a specific time period in the future.

[0054] The task matching and decision generation module calculates the cumulative intensity of any task point covered by the ability field of one or more power supply units within a specified time window based on the spatiotemporal ability field of each power supply unit; generates a global scheduling decision scheme with the optimization objective of maximizing the total cumulative intensity of all task points; and specifies the execution order and movement path of the scheduled power supply units in the global scheduling decision scheme.

[0055] The instruction execution and state synchronization module issues scheduling instructions to the corresponding power supply units according to the global scheduling decision scheme and synchronously updates the spatiotemporal ability field state of the power supply units, marking their power supply capabilities as allocated.

[0056] In the heterogeneous data acquisition module, the heterogeneous data acquisition module is responsible for acquiring the basic data constituting the power supply capability image of the mobile emergency DC power supply unit and its operating environment. The core of the design of this module lies in the parallel and independent acquisition of two kinds of data streams with different properties: static capability elements and dynamic capability elements. Static capability elements represent the inherent characteristics of power supply units that are relatively stable and not easily changed, and are the basis for evaluating their long-term capability potential. Dynamic capability elements reflect the real-time state of the power supply unit that changes constantly during task execution, and are the key basis for making immediate scheduling decisions. Through the separate acquisition of such heterogeneous data, a clear structure and complete dimension input are provided for subsequent construction of an accurate spatiotemporal ability field model.

[0057] The acquisition of static capability elements first targets the rated capacity and core inherent physical properties of the power supply unit. The rated capacity data is usually pre-installed in the control system of the power supply unit, and the acquisition process reads it from the local memory or management chip of the unit through a standard device information query protocol. The value represents the maximum energy that the unit can continuously output under ideal conditions, with the unit being kilowatt-hours. The acquisition range of inherent physical properties includes but is not limited to the overall weight, size, configured interface type and standard, and supported special function modes of the unit. Some of these attribute data are obtained from the nameplate information database pre-installed with the unit at the factory, and some are obtained by querying the central device archive after identifying the unit model. The data obtained in this step constitutes the basic capability profile of the power supply unit and is the prerequisite for determining whether it can meet the physical requirements of a specific task.

[0058] The first task of collecting dynamic capability elements is to obtain the spatial position information of the power supply unit in real time. This function is achieved through the global satellite navigation system positioning integrated in the power supply unit or its carrier. By continuously receiving satellite signals and solving, coordinate data containing latitude, longitude, elevation and time stamp are generated. Position data is not only used to determine the absolute geographical position of the unit, but more importantly, it provides the basis for calculating the relative spatial relationship between the unit and the potential task point. The collection system ensures that the position information is updated at a set period and transmitted to the dispatch center through the mobile communication network, so as to realize the real-time visual tracking of the position of the power supply unit on the electronic map.

[0059] Another key component of dynamic capability elements is the real-time remaining power of the power supply unit. This data is obtained by monitoring the state of charge parameter provided by the unit battery management system. The battery management system calculates the percentage of the current remaining power to the total capacity by monitoring the voltage, current and temperature of the battery pack in real time. This percentage value is packaged into a standard data frame and continuously reported through wired or wireless communication interface. Real-time remaining power is a core variable for predicting the sustainable power supply time of the power supply unit, which directly determines the time coverage range of its space-time capability field.

[0060] Dynamic capability elements also include the real-time health status assessment of the power supply unit. Health status is a comprehensive evaluation parameter, and its calculation depends on the comprehensive analysis of the running status of multiple subsystems. Data sources include the running efficiency and temperature data of the battery health index provided by the battery management system, as well as the state feedback of key components such as cooling system and connector. After these original state parameters are collected, they are fused and analyzed by a built-in health assessment algorithm, and finally a quantitative health status score or grade is output. The health status value reflects the degree of performance degradation of the power supply unit relative to its brand-new state, and directly affects the reliability and safety of its actual output capability.

[0061] In the spatiotemporal capacity field construction module, the construction process of the spatiotemporal capacity field starts with the quantitative assessment of the basic capacity of the power supply unit. The input of this process is the static and dynamic capacity elements obtained by the heterogeneous data acquisition link. The initial intensity of the capacity field is determined according to the rated capacity value in the static capacity element. The rated capacity reflects the total amount of energy reserves of the power supply unit in the ideal state, usually in units of kilowatt-hours. The setting of the initial intensity is proportional to the rated capacity, for example, a proportion can be set to linearly map the rated capacity value to a dimensionless initial intensity reference value. Subsequently, the real-time health status value in the dynamic capacity element is introduced to correct the initial intensity. Health status is a parameter between 0 and 1 or expressed as a percentage, which represents the degree of retention of the current performance of the power supply unit relative to its brand-new state. The correction process usually directly multiplies the initial intensity by the health status value, thereby obtaining a capacity field intensity that has been corrected by health loss. This step ensures that the intensity basis of the capacity field not only considers the theoretical potential of the unit, but also reflects its current actual reliable degree.

[0062] With the real-time spatial position coordinates of the power supply unit as the center, the effective influence range of its power supply capacity is divided into three concentric ring-shaped regions on the two-dimensional electronic map, namely the core region, the effective region, and the extended region. The core region is the smallest ring-shaped region adjacent to the position of the power supply unit, and its radius range is set according to the typical optimal power supply distance of this type of unit. The effective region is the ring-shaped zone between the core region and the extended region, and its outer boundary radius is usually set to the maximum economic radius that the unit can effectively reach and provide service in a single task. The extended region is the largest ring-shaped region, and its outer boundary represents the theoretical maximum coverage limit of the unit's power supply capacity considering extreme or emergency situations. Each ring-shaped region is assigned a specific capacity intensity coefficient, which defines the relative intensity of the power supply capacity that a task point located in that region can feel. The intensity coefficient of the core region is set to the highest value, followed by the effective region, and the extended region is the lowest. This setting reflects the physical law that the power supply capacity decays with increasing spatial distance, i.e., the closer to the power supply unit, the higher the reliability, efficiency, and power quality of the power supply.

[0063] The corrected overall capacity field intensity needs to be combined with the above spatial structure to generate a visual field model. Multiplying the corrected capacity field intensity value by the intensity coefficient of each region, the actual intensity value of each region can be obtained. In the visual presentation of the electronic map, this intensity distribution usually appears as a gradient color map or contour map with the unit position as the center and the intensity gradually decaying from inside to outside. The core region has the deepest color or the densest contour, indicating the highest intensity; the extended region has the lightest color or the sparsest contour, indicating the lowest intensity. This spatiotemporal capacity field map statically depicts the distribution of the power supply unit's power supply capacity in the spatial range at the current time.

[0064] To predict the capacity changes in future time periods, the spatiotemporal capacity field needs to be dynamically extrapolated. This extrapolation relies on a conceptual tool called the spatiotemporal convolution kernel. This convolution kernel is essentially a weight distribution function, whose shape and parameters are determined by key dynamic capacity factors, particularly the real-time residual power consumption rate and the average moving speed of the supply unit. The power consumption rate affects how quickly the capacity field strength decays over time, while the average moving speed determines the trajectory and speed at which the capacity field center moves in space. The convolution kernel jointly defines an influence range in both time and space, simulating how the capacity field evolves over a small future time period due to unit movement and power consumption.

[0065] Dynamic extrapolation is achieved through convolution operations. The spatiotemporal capacity field map at the current time is convolved with the defined spatiotemporal convolution kernel. This mathematical operation simulates the propagation and decay effects of the capacity field over time and space. Specifically, it calculates the weighted cumulative value of the supply capacity strength at each potential spatial location in the future, with the weights determined by the convolution kernel, while considering the center position shift due to movement and the overall strength decay due to power consumption. By performing multiple convolution operations in succession, each advancing a small simulation time step, the continuous changes in the capacity field over a period of time in the future can be simulated.

[0066] Finally, the extrapolation process outputs a set of discrete spatiotemporal capacity field snapshots, each corresponding to a specific future time point. This sequence of snapshots clearly shows how the supply unit's capacity field evolves over time, including its movement in coverage range, changes in shape, and overall strength decay. These prediction information is crucial for determining whether the supply unit can meet the spatiotemporal and capacity requirements of a specific task at various future time points.

[0067] The determination of the intensity coefficient is a dynamic adjustment process. First, set an initial benchmark intensity coefficient for the core area, the effective area and the extended area respectively. These benchmark values are pre-set based on historical data, physical models or expert experience, reflecting the basic law of capacity decay with distance under ideal full power and healthy state, with the core area having the highest benchmark coefficient and the extended area having the lowest. Then, correction based on real-time power is carried out. Calculate the ratio of real-time residual power to rated capacity in the dynamic capacity element to obtain the current power percentage. Take this percentage as a power decay factor, which is usually directly used to scale the benchmark intensity coefficients of each area by equal ratio. This means that the lower the power, the lower the capacity intensity coefficient of all areas will be proportionally reduced. Then, secondary correction based on the health state is carried out. Take the real-time health state value in the dynamic capacity element as a health impairment factor, which is also a value between 0 and 1. Multiply this health impairment factor by the intensity coefficient corrected by the power decay to further impair it. After these two corrections, the actual capacity intensity coefficients applied to each annular area at the current time are obtained. These coefficients more accurately reflect the spatial capacity distribution characteristics of the power supply unit under the current actual state. Finally, bind these calculated actual intensity coefficients with the annular areas they correspond to, thereby completing the quantitative construction of the internal structure of the space-time capacity field, providing a basis for subsequent spatial analysis and calculation.

[0068] In the task matching and decision generation module, the core goal of the task matching and decision generation process is to generate a globally optimized scheduling scheme to efficiently allocate limited mobile emergency DC power supply units to multiple emergency task points. The process starts with quantitatively evaluating the importance of each task point, i.e. calculating its comprehensive gravitational weight. This weight is determined by two key factors: power demand and time urgency. The power demand weight is set according to the size of the required power of the task, usually using a segmented function, for example, dividing the power demand into several levels, each level corresponding to a weight value, the larger the demand power, the higher the weight. The time urgency weight is calculated according to the difference between the latest start power supply time required by the task and the current time, the smaller the difference, the higher the urgency, and the larger the corresponding weight value. Then, the power demand weight and the time urgency weight are combined into a comprehensive gravitational weight by multiplication or weighted average fusion. The larger the weight value, the higher the priority of the task point in global scheduling, which should be given priority in resource allocation.

[0069] After the gravity weight of a task point is determined, the next step is to calculate the contribution degree of a single power supply unit to a specific task point, i.e. the single contribution degree. This calculation needs to first determine the capability intensity value of the spatiotemporal capability field of the power supply unit at the geographic location of the task point within the time window required by the task. The calculation of this intensity value is a spatiotemporal integration process. First, the optimal or feasible path of the power supply unit from the current location to the task point needs to be planned, and the movement process of the power supply unit in the future period of time is discretized into a spatiotemporal projection sequence in combination with the time window required by the task. This sequence consists of spatial positions at multiple time points, each of which corresponds to the theoretical intensity value of the capability field of the power supply unit at the task point at that time. Then, the theoretical capability intensity on the spatiotemporal projection sequence is weighted and integrated. The weighting function usually adopts a time decay function, and the intensity value closer to the start time of the time window has a higher weight, and vice versa, in order to reflect the timeliness preference of power supply. Finally, the result of the weighted integration is divided by the length of the entire time window, and the average effective intensity value is obtained. This value represents the average power supply capability level provided by the power supply unit to the task point during the entire task execution period. Multiply the calculated average effective intensity value by the comprehensive gravity weight of the task point, and the result is the single contribution degree of the power supply unit to the task point.

[0070] For a task point, it may be covered by the capability fields of multiple power supply units. Therefore, the final cumulative intensity of the task point needs to be obtained by superimposing the single contribution degrees of all power supply units to it. Specifically, the single contribution degrees of each power supply unit to the task point are simply added by arithmetic, and the result of the summation is the cumulative intensity of the task point. This cumulative intensity value comprehensively reflects the overall power supply guarantee level that the task point can obtain under the synergistic action of all available power supply units. The system optimization goal is to maximize the sum of the cumulative intensities of all task points, which means that the power supply resources are most effectively allocated in the global range, and the overall power supply guarantee capability is the highest.

[0071] Generating a global dispatching decision scheme is a complex combinatorial optimization problem, which is solved by a population-based optimization algorithm. At the initialization stage, a set of initial dispatching decision schemes are randomly generated, each of which is called an individual, and all of which form the initial population. Each individual contains a complete set of dispatching arrangement, which specifies which power supply units are assigned to which task points, the execution order, and the moving path of each unit to the task points. Subsequently, the algorithm enters the iterative optimization stage. In each iteration, the individuals in the population are first subjected to crossover operation, i.e., randomly selecting two individuals and exchanging their partial dispatching arrangements to generate new individuals. Then mutation operation is performed to randomly change the assignment target, execution order or path of a power supply unit in an individual with a small probability, in order to increase the diversity of the population. To ensure the feasibility of the newly generated individuals, a conflict detection and resolution operator is introduced. This operator detects whether there is a resource conflict in the dispatching scheme, such as the same power supply unit being assigned to multiple tasks, the paths of different units overlapping in time and space, or the expected task execution time crossing. Once a conflict is detected, it is resolved by adjusting the execution order, re-planning the path, or replacing the power supply unit, etc.

[0072] The optimization objective of the algorithm is defined by a composite optimization function, which simultaneously pursues the maximization of the total cumulative intensity of all task points and the minimization of the total moving path of all dispatched power supply units. Since these two objectives often conflict with each other, the algorithm adopts a multi-objective optimization strategy. The iteration process continues until the preset termination condition is met, such as reaching the maximum number of iterations or the quality of the solution no longer improves significantly. After the iteration is terminated, the Pareto optimal solution set is selected from the final population, i.e., those non-dominated solutions that cannot improve one objective without compromising the other. Finally, the system selects a compromise scheme from the Pareto optimal solution set according to the actual decision preference (e.g., prioritizing intensity or prioritizing cost saving), as the final global dispatching decision scheme output. This scheme explicitly lists the task sequence to be executed by each dispatched power supply unit, the transfer path between tasks, and the approximate timing arrangement.

[0073] In the instruction execution and state synchronization module, the instruction execution and state synchronization process is the key link to ensure that the scheduling decision can be accurately translated into on-site action and maintain the consistency of the system internal state. The process starts after the scheduling decision scheme is determined, and before any actual instruction is issued, the system first creates a current state snapshot of the time and space capability field for all power supply units allocated tasks in this scheduling decision. This snapshot records all parameters of the power supply unit capability field at the current time, including its spatial position, the capability intensity coefficient of each region, and the overall intensity reference value of the capability field. At the same time, the system generates a globally unique transaction identifier for this group of operations, which will be used to associate all subsequent operations and state updates, providing data basis for possible rollback operations. The purpose of creating a snapshot is to freeze the instantaneous state of the power supply unit before scheduling, like taking a photo of the current resource allocation situation of the system, to ensure that when an exception occurs in subsequent operations, a clear recovery point can be obtained.

[0074] Next, the system encapsulates the scheduling instruction issuing operation and the time and space capability field model updating operation as an atomic transaction. The meaning of atomic transaction is that these two operations are considered as an indivisible whole, either all successfully executed or not executed at all, without intermediate state. In technical implementation, this is usually guaranteed by transactional memory or distributed transaction protocol. The specific process is as follows: the system first sends the scheduling instruction to the communication gateway, which contains the task details, target location, execution path and the above unique transaction identifier. Almost at the same time, in the internal memory database of the system, the time and space capability field state of the power supply unit is marked as a temporary state of "allocating", and the capability field intensity value from the current time to the estimated task end time is set to zero. Setting the intensity to zero means that the unit is considered to be fully occupied during this period and no longer participates in any new scheduling decision calculation, so it is immediately logically excluded from the available resource pool, avoiding the risk of repeated allocation of the same resource. Atomicity ensures that the resource state is updated synchronously as soon as the instruction starts to issue, thus maintaining the consistency of the decision context at the system level.

[0075] After the instruction is issued and the resource state is temporarily updated, the system enters a waiting confirmation stage to ensure that the instruction has been reliably received by the on-site power supply unit. The system waits for the execution end of the power supply unit to return an instruction confirmation signal through the communication network. This signal indicates that the control system of the power supply unit has successfully received and parsed the scheduling instruction and has started to execute the corresponding movement or preparation action. Only after successfully receiving this confirmation signal, the system will formally commit the "allocating" temporary state set in the atomic transaction, converting it into a persistent "allocated" state. This step is the process of solidifying the temporary state, marking that the scheduling task has officially entered the execution phase, and the capability field of the power supply unit will remain unavailable during the task period.

[0076] However, the communication link or the execution end device can be at risk of failure. Therefore, the system presets a timeout time, for example, 300 seconds. If the instruction confirmation signal from the power supply unit is not received within the timeout time, the system automatically determines that the instruction execution is abnormal. At this time, the preset rollback operation is triggered. The rollback operation uses the initially created state snapshot and the unique transaction identifier to completely restore the corresponding power supply unit's spatiotemporal capability field state to the snapshot time, that is, to cancel the "allocated" mark and restore the capability field intensity value to the original value. This operation makes the power supply unit return to the available state and can be considered in subsequent scheduling.

[0077] After the rollback operation is completed, the system starts a consistency verification process. The process compares all the power supply units' spatiotemporal capability field states maintained internally by the system with the latest state data of each unit actually collected from the field. The comparison indicators include the unit's position, power, health status, and current task execution state. If any inconsistency is found, for example, the system internally shows that a unit is in the "allocated" state, but the actually collected data indicates that the unit is still in standby position and has no signs of task execution, the system will generate a detailed alarm information. The alarm information records the inconsistent power supply unit identifier, inconsistent data item, and deviation degree, and notifies the system administrator to start the manual intervention process to troubleshoot communication failure, device failure or other abnormal reasons, ensure that the system data and the actual situation are re-aligned, and maintain the reliability and credibility of the entire scheduling system.

[0078] The working principle of the present application is: through the isomerism data acquisition module, the static capacity elements (including rated capacity and inherent physical properties) and dynamic capacity elements (including real-time spatial position, residual power and health status) of each power supply unit are acquired in real time, and a space-time capacity field model representing the distribution and attenuation law of power supply capacity in space-time dimension is constructed by the space-time capacity field construction module based on the above elements, the model divides the core area, effective area and extended area with the current position of the power supply unit as the center, and dynamically corrects the capacity intensity coefficient of each area combined with the power, health status and movement characteristics, and then predicts the future capacity field evolution through convolution deduction; the task matching and decision generation module calculates the comprehensive gravity weight based on the power demand and time urgency of each task point, quantifies the average effective capacity intensity of each power supply unit to the task point within the specified time window, obtains the single contribution degree and accumulates to form the cumulative intensity of the task point, takes the maximum of the total cumulative intensity of all task points and the minimization of the total moving path as the goal, adopts a multi-objective optimization algorithm to generate a global scheduling decision scheme including power supply unit allocation sequence, path planning and execution timing; the instruction execution and state synchronization module encapsulates the scheduling instruction and state update as an atomic transaction, marks the space-time capacity field of the corresponding power supply unit as "allocating" and sets its capacity output to zero at the same time, and only after receiving the on-site confirmation signal, it is formally submitted to the "allocated" state, if it is not confirmed within the time limit, it will be rolled back and restored using the pre-created state snapshot, and through the consistency verification mechanism, the system state and field data are compared to ensure the accuracy of resource scheduling and the reliability of system operation.

[0079] The above describes one embodiment of the present application in detail, but the content described is only the preferred embodiment of the present application, and cannot be considered as limiting the scope of the present application. Any equivalent changes and improvements made within the scope of the present application shall still belong to the scope of the present application.

Claims

1. A mobile emergency DC power supply intelligent dispatching system, characterized in that, include: The heterogeneous data acquisition module collects the static and dynamic capability elements of each power supply unit in real time. The static capability elements include: the rated capacity and inherent physical properties of the power supply unit; the dynamic capability elements include: the real-time spatial location, real-time remaining power, and real-time health status of the power supply unit. The spatiotemporal capability field construction module constructs a spatiotemporal capability field for each power supply unit based on static and dynamic capability elements. The spatiotemporal capability field is used to characterize the effective spatial coverage of the power supply capability of the power supply unit in a specific time period in the future and the attenuation law with time and spatial movement. The construction process of the spatiotemporal capability field is as follows: The initial strength of the capacity field is determined based on the rated capacity in the static capacity elements, and the initial strength is degraded based on the real-time health status in the dynamic capacity elements. Centered on the real-time spatial location of the power supply unit, the effective coverage of the power supply capacity of the power supply unit in space is divided into three concentric ring areas: the core area, the effective area, and the extension area. Each area is assigned a different capacity strength coefficient, with the core area having the highest strength coefficient and the extension area having the lowest strength coefficient. The corrected energy field strength is used to generate a spatiotemporal energy field that is visualized on an electronic map, with its strength gradually decreasing from the center to the edge; The deduction process of the spatiotemporal capability field includes: Define a spatiotemporal convolution kernel, the parameters of which are jointly determined by the real-time remaining power consumption rate and the average moving speed of the power supply unit in the dynamic capability elements; Using the spatiotemporal convolution kernel, a convolution operation is performed on the spatiotemporal capability field map at the current moment to simulate the dynamic process of its spatial position movement and overall intensity decay in the future due to the movement of power supply units and power consumption. By continuously performing convolution operations, a set of spatiotemporal capability field snapshots arranged in future time series are output to predict the dispatchable capability of the power supply unit at various points in the future. The task matching and decision generation module calculates the cumulative intensity of the coverage of any task point by the capacity fields of one or more power supply units within a specified time window, based on the spatiotemporal capability field of each power supply unit; and generates a global scheduling decision scheme with the optimization objective of maximizing the total cumulative intensity of all task points; the global scheduling decision scheme explicitly specifies the execution order and movement path of the scheduled power supply units. The instruction execution and status synchronization module issues scheduling instructions to the corresponding power supply units according to the global scheduling decision scheme, and synchronously updates the spatiotemporal capability field status of the power supply units, marking their power supply capability as allocated.

2. The mobile emergency DC power supply intelligent dispatching system according to claim 1, characterized in that, The process of obtaining the strength coefficient is as follows: Based on the preset mapping relationship, an initial reference strength coefficient is assigned to the core area, effective area and extended area respectively, with the reference strength coefficient of the core area set to the maximum value; Based on the ratio of real-time remaining power to rated capacity in the dynamic capability elements, a power attenuation factor is calculated, and the reference strength coefficient is scaled proportionally using the power attenuation factor. Based on the real-time health status value in the dynamic capability elements, a health depreciation factor is calculated, and this factor is multiplied by the intensity coefficient after power attenuation correction to obtain the final actual capability intensity coefficient for each region. The calculated actual capacity intensity coefficients of each region are bound to the corresponding annular regions to complete the quantitative construction of the internal structure of the spatiotemporal capacity field.

3. The mobile emergency DC power supply intelligent dispatching system according to claim 1, characterized in that, The calculation process for the cumulative intensity includes: Based on the power requirements and time urgency of the emergency mission, the urgency weight and power level weight of each mission point are calculated, and the two are integrated into a comprehensive gravitational weight. For the capability field of a power supply unit, calculate its capability strength value at the geographical location of the mission point within the time window required by the mission, and multiply the capability strength value by the comprehensive gravitational weight of the mission to obtain the single contribution of the power supply unit to the mission point. For the same task point, the individual contributions of all power supply units are linearly superimposed to obtain the final cumulative intensity of the corresponding task point.

4. The mobile emergency DC power supply intelligent dispatching system according to claim 3, characterized in that, The calculation process for the capability strength value is as follows: By combining the planned path for moving the power supply unit from its current location to the mission point with the time window, the spatiotemporal projection sequence of the capability field at the mission point within the time window is determined. Weighted integral of the theoretical capability strength over the spatiotemporal projection sequence; Dividing the weighted integral by the length of the time window yields an average effective strength value of the power supply unit at the task point within the task time window, thus obtaining the capability strength value.

5. The mobile emergency DC power supply intelligent dispatching system according to claim 1, characterized in that, The generation of the global scheduling decision scheme specifically includes: A set of initial scheduling decision links containing different power supply units and task point matching relationships and execution order is randomly generated to form the initial population; The population is iteratively optimized. In each generation, the decision-making links are cross-linked and mutated, and a conflict detection and resolution operator is introduced. The resolution operator will detect and correct path overlap, resource over-allocation and time conflict. A composite optimization function is set up with the objectives of maximizing the total cumulative strength and minimizing the total scheduling path. After the iteration terminates, the optimal decision scheme located at the Pareto front is selected from the final population as the global scheduling decision scheme.

6. The mobile emergency DC power supply intelligent dispatching system according to claim 1, characterized in that, The spatiotemporal capability field state of the updated power supply unit specifically includes: Before issuing dispatch instructions, create a snapshot of the current state of the spatiotemporal capability field of the affected power supply units and generate a unique transaction identifier. The issuance of scheduling instructions and the updating of the spatiotemporal capability field are encapsulated into atomic transactions; In an atomic transaction, the spatiotemporal capability field of the scheduled power supply unit is instantaneously set to zero at the current moment and during the duration of the future task, so that it is immediately removed from the available resource pool.

7. The mobile emergency DC power supply intelligent dispatching system according to claim 1, characterized in that, The allocated state also includes enhancement steps to ensure eventual consistency of the system state: Wait for the power supply unit to return a command confirmation signal. Only after receiving the confirmation signal will the temporary state allocated in the atomic transaction be solidified into a persistent state. If no instruction confirmation signal is received within the preset timeout period, the instruction execution is determined to be abnormal, and a rollback operation is automatically triggered to cancel the allocated markers of the power supply unit's spatiotemporal capability field and restore its strength to the transaction snapshot state. After the rollback is completed, the system verifies the consistency between the spatiotemporal capability field state and the actual state of all power supply units. If a deviation is found, an alarm is generated and a manual intervention process is initiated.

Images