A method, medium, and equipment for registering fire emergency lighting fixtures with low RAM centralized power supply
By using time intervals to issue instructions and identify completed circuit data during the registration process of fire emergency lighting fixtures, the scheduling of multiple power supplies in parallel and single power supply chain connection is realized, solving the problem of long-term idle time and delay of equipment under low-cost hardware architecture and improving registration efficiency.
Patent Information
- Application Number
- CN202611558970.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-09
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies, centralized power supplies with low-cost hardware architectures cause long-term idle time of equipment and serious cumulative delay in the registration of fire emergency lighting fixtures. Especially in large-scale networking scenarios, serial registration of each fixture takes too long, and optimization solutions cannot be effective without prior data.
By issuing registration instructions at preset time intervals, identifying and processing the lighting data of completed circuits, the time overlap of the first circuit registration operations of multiple centralized power supplies is achieved. After the current circuit data is processed, the next circuit instruction is issued immediately. Combining the bus interaction results, high-priority power supplies are processed first, and a scheduling architecture for parallel connection between power supplies and chain connection within a single power supply is constructed.
It significantly reduced the overall registration latency under large-scale networking, improved bus resource utilization and single power supply registration efficiency, shortened the registration time of all lighting fixtures in the network, and improved the efficiency of on-site commissioning and rapid online deployment.
Smart Images

Figure CN123124185A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of communication dispatching for fire emergency lighting control systems, and in particular to a method, medium, and device for registering fire emergency lighting fixtures with low RAM centralized power supply. Background Technology
[0002] In fire emergency lighting systems, controllers typically communicate with multiple centralized power supplies via a bus network to perform one-click registration of lighting fixtures. However, due to the low-cost hardware architecture of centralized power supplies, their internal storage resources are extremely limited, allowing them to cache only one circuit's lighting data and execute single-circuit registration at a time. Currently, the industry-standard lighting registration solution generally adopts a serial registration mechanism, where the controller must wait for all circuits of lighting fixtures from the current centralized power supply to be registered before issuing the registration command for the next centralized power supply. This serial registration mode results in all other centralized power supplies in the network being in a passive, idle waiting state during the entire registration process of the current centralized power supply, preventing the parallel utilization of bus resources and device operation resources. Especially in large-scale network scenarios with hundreds or more centralized power supplies, the serial registration process accumulates significant waiting latency, causing the overall registration time for all lighting fixtures in the network to increase exponentially. Furthermore, existing optimization solutions often rely on prior information such as historical load or timing prediction, which is ineffective in scenarios involving initial on-site commissioning without any pre-configured settings. Therefore, there is an urgent need for a registration scheduling scheme that can adapt to hardware constraints with low storage resources and significantly reduce the registration time of all lighting fixtures in the network without prior data. Summary of the Invention
[0003] This application provides a method, medium, and device for registering fire emergency lighting fixtures with low RAM centralized power supply, which solves the technical problems of long-term idle equipment and serious cumulative delay in the registration of lighting fixtures across the entire network caused by serial registration of the whole machine in the prior art.
[0004] To achieve the above objectives, this application adopts the following technical solution: Firstly, a method for registering fire emergency lighting fixtures with low RAM centralized power supplies is provided, including: Obtain the first loop registration instruction and send it sequentially to multiple centralized power supplies based on a preset time interval to obtain the bus interaction result; Based on the bus interaction results, identify the target centralized power supply among the multiple centralized power supplies that has completed registration in the current loop and whose lamp data is pending processing. Receive and process the current circuit lighting data of the target centralized power supply; In response to the completion of the current circuit lighting data processing and the existence of an unregistered circuit in the target centralized power supply, a next circuit registration instruction is issued to the target centralized power supply; wherein, the next circuit registration instruction is used to instruct the target centralized power supply to perform lighting registration for the unregistered circuit.
[0005] Based on the above technical solutions, in the fire emergency lighting registration method for low RAM centralized power supplies provided in this application, the registration operations of the first circuit of multiple centralized power supplies are overlapped in time, eliminating the problem of long-term idle waiting of other centralized power supplies in the traditional serial method, and significantly reducing the overall registration latency under large-scale networking. At the same time, based on the bus interaction results, the centralized power supply with the current circuit registration completed and the lighting data pending processing is identified and prioritized, avoiding the data of the completed circuit from being stuck in the low RAM centralized power supply and blocking the start of the next circuit. When the lighting data of the current circuit is processed and there is an unregistered circuit, the registration instruction for the next circuit is immediately issued, realizing the windowless chain connection registration of multiple circuits within the same centralized power supply, which can adapt to the single circuit exclusive constraint without increasing the RAM of the centralized power supply or modifying its hardware. Therefore, the combination of multi-power supply parallelism and single-power supply chain connection reduces the bus invalid idle and waiting latency, improves the bus resource utilization and single power supply registration efficiency, and can effectively shorten the overall registration time of the entire network of fire emergency lighting fixtures in the scenario of hundreds of centralized power supplies networking, and improve the efficiency of on-site debugging and rapid online.
[0006] In conjunction with the first aspect above, in one possible implementation, receiving and processing the current circuit lighting data of the target centralized power supply includes: Send a data upload request to the target centralized power supply that has completed registration in the current circuit and whose lamp data is pending processing. The target centralized power supply is received frame by frame in response to the data upload request, and the multiple data frames carry the current circuit lamp data; Perform verification processing on each data frame, including power address, circuit number, frame sequence number, and check value. In response to the failure of verification processing for any data frame, a retransmission request for the data frame for which verification processing failed is initiated to the target power supply, and the retransmitted data frame is received. In response to the fact that all of the multiple data frames have passed the verification process, the current circuit lighting data is written to local storage to determine that the current circuit lighting data processing is complete.
[0007] The above scheme avoids the bus overhead of retransmitting the entire circuit due to a single frame error by performing fine-grained verification on each data frame and initiating a retransmission request only for the failed data frame when the verification fails. At the same time, by writing the verified data to local storage, the storage space of the centralized power supply is freed up, allowing it to enter the registration preparation stage of the next circuit as soon as possible.
[0008] In conjunction with the first aspect above, in one possible implementation, identifying the target centralized power supply among the plurality of centralized power supplies that has completed registration in the current circuit and whose lighting data is pending processing includes: Receive the current loop completion message returned by the multiple centralized power sources; For the current loop, perform consistency verification, address matching, and loop number verification on the message. In response to the current loop completion message passing the consistency check, the address matching and the loop number verification, it is determined that the corresponding centralized power supply is registered in the current loop and the lamp data is pending processing. In response to the bus interaction result meeting the preset abnormal conditions, the corresponding centralized power supply is marked as a communication abnormal stop state; In response to receiving a valid message from the centralized power supply that is in the communication abnormal stop state, the registration status of the corresponding centralized power supply is restored.
[0009] The above solution constructs discrete state transition logic that can be objectively identified by the controller, and determines the operating condition of the centralized power supply entirely based on the results of active query and real-time message interaction. It does not require firmware modification of the centralized power supply or active reporting, thus ensuring the accuracy and fault tolerance of state identification in complex bus environments.
[0010] In conjunction with the first aspect mentioned above, one possible implementation also includes: Obtain the number of remaining unregistered circuits, current registration status, and communication response stability of each of the multiple centralized power sources; Based on the number of remaining unregistered circuits, the current registration status, and the stability of the communication response, the inspection sequence of the multiple centralized power supplies is dynamically adjusted. In this inspection sequence, centralized power supplies that have completed the current circuit registration and whose lamp data is pending have a higher priority than centralized power supplies in other states. Central power supplies with fewer remaining unregistered circuits have a higher priority than centralized power supplies with more remaining unregistered circuits. Central power supplies in the communication abnormal stop state are postponed or have their inspection frequency reduced in the inspection sequence.
[0011] The above scheme constructs a parameter-free priority evaluation system by collecting real-time available parameters, prioritizing centralized power supplies with fewer pending states and remaining circuits, and relegating centralized power supplies that are abnormally stalled to the back. This allows bus query resources to dynamically focus on high-value nodes, reducing the redundant traversal time caused by invalid bus queries.
[0012] In conjunction with the first aspect mentioned above, one possible implementation also includes: Construct a global inspection queue that includes the multiple centralized power sources; In response to any centralized power supply entering the full-loop registration completion state, or the centralized power supply being in the communication abnormal stagnation state and not receiving the valid message for a preset duration, the centralized power supply is removed from the global inspection queue; Based on the removed global inspection queue, the inspection order is updated, and the inspection interval for the remaining centralized power supplies in the global inspection queue is shortened. The removed centralized power supply no longer participates in subsequent polling and generates a corresponding fault device record, which is used to support reactivation.
[0013] The above solution employs a dual-dimensional adaptive queue slimming mechanism to simultaneously remove centralized power supplies that have completed registration normally and those that have been abnormally stalled for a long time. This results in a continuous reduction in the effective size of the global inspection queue, a continuous reduction in the time required for a single round of inspection traversal, and a gradual improvement in the data reporting response speed of the remaining normal but incomplete centralized power supplies.
[0014] In conjunction with the first aspect above, in one possible implementation, the step of sequentially sending data to multiple centralized power supplies based on a preset time interval to obtain bus interaction results includes: The first circuit registration instruction is sent to each of the multiple centralized power supplies at the preset time interval; After issuing the first circuit registration instruction to the current centralized power source, the first circuit registration instruction is issued to the next centralized power source without waiting for the current centralized power source to complete the first circuit registration. The bus interaction result is obtained based on the response messages returned by each centralized power supply during the process of distributing the data to each unit.
[0015] The above scheme sets a fixed bus frame interval to stagger the batch issuance of instructions, and continues to issue instructions without waiting for the current centralized power supply to complete registration. This ensures that there is a bus arbitration and response interval between multiple frames of instructions to avoid bus congestion. At the same time, it forms an overlapping interval for parallel registration of multiple power supplies on the time axis, eliminating the idle waiting delay in the traditional serial mode.
[0016] In conjunction with the first aspect above, in one possible implementation, the step of issuing a next circuit registration instruction to the target centralized power supply in response to the completion of the current circuit lighting data processing and the existence of an unregistered circuit in the target centralized power supply includes: In the callback or interruption node where the current circuit lighting data processing is completed, it is determined whether the unregistered circuit exists in the target centralized power supply. In response to the presence of the unregistered circuit in the target centralized power supply, the next circuit registration instruction is immediately issued to the target centralized power supply without waiting for the next round of fixed-cycle inspection; In response to the absence of the unregistered circuit in the target centralized power supply, the target centralized power supply is marked as having completed full circuit registration. The registration and activation of subsequent circuits of the target centralized power supply are passively triggered by the completion of the data processing of the previous circuit's lighting fixtures.
[0017] The above solution achieves seamless serial connection of multiple circuits within a single power supply by immediately triggering the issuance of the next circuit registration instruction in the callback or interrupt node after data processing is completed. It strictly adapts to the hardware constraints of the single circuit exclusive storage of the centralized power supply and avoids problems such as power supply idling caused by the retention of data of completed circuits.
[0018] In conjunction with the first aspect mentioned above, one possible implementation also includes: A registration context is maintained for each of the centralized power supplies. The registration context includes the power supply address, current registration status, current loop number, number of processed loops, number of remaining unregistered loops, consecutive no-response count, consecutive verification failure count, and timeout timer. The registration context is maintained only based on real-time interaction with the controller. The registration context is updated based on the results of proactive inspection and real-time bus message interaction. Based on the updated registration context, at least one of the following is performed: identifying the target centralized power supply, receiving and processing the current circuit lighting data, and issuing the next circuit registration instruction.
[0019] The above solution establishes an independent registration context for each centralized power supply on the controller side and updates it based solely on real-time interaction results. This eliminates the need for prior information such as the number of lights or historical registration data during the entire scheduling process, making it perfectly suited for the blank scenario of one-click registration of lights at the project site for the first time.
[0020] In a second aspect, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements a method for registering fire emergency lighting fixtures with a low RAM centralized power supply as described above.
[0021] Thirdly, an electronic device is provided, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements a low RAM centralized power supply method for registering fire emergency lighting fixtures as described above.
[0022] Beneficial effects: This application provides a method, medium, and device for registering fire emergency lighting fixtures using a low-RAM centralized power supply. By overlapping the registration operations of the first circuit of multiple centralized power supplies in time, it eliminates the problem of long-term idle waiting of other centralized power supplies in the traditional serial method, significantly reducing the overall registration latency in large-scale networking. At the same time, based on the bus interaction results, it promptly identifies and prioritizes the centralized power supply whose current circuit registration is completed and whose lighting data is pending, avoiding the data of completed circuits from being stuck in the low-RAM centralized power supply and blocking the start of the next circuit. When the lighting data of the current circuit is completed and there are unregistered circuits, the registration command for the next circuit is immediately issued, realizing the seamless chained registration of multiple circuits within the same centralized power supply. It can adapt to the single-circuit exclusive constraint without increasing the RAM of the centralized power supply or modifying its hardware. Therefore, the combination of multi-power supply parallelism and single-power supply chained connection reduces the bus's invalid idle time and waiting latency, improves the bus resource utilization and single-power supply registration efficiency, and can effectively shorten the overall registration time of the entire network of fire emergency lighting fixtures in the scenario of hundreds of centralized power supplies, improving the efficiency of on-site commissioning and rapid online deployment.
[0023] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description
[0024] Figure 1 A flowchart illustrating a method for registering fire emergency lighting fixtures with low RAM centralized power supply, provided in an embodiment of this application; Figure 2 A schematic diagram of the process for rapid registration of state machine-based adaptive chain lighting fixtures provided in an embodiment of this application; Figure 3This is a schematic diagram of the state transitions of a controller-side identifiable discrete state machine provided in an embodiment of this application; Figure 4 A schematic diagram illustrating the principle of dynamic inspection queue priority reordering and two-dimensional adaptive queue slimming provided in the embodiments of this application; Figure 5 This is a flowchart illustrating the two-layer timing-driven mechanism provided in an embodiment of this application. Detailed Implementation
[0025] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.
[0026] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.
[0027] Example 1 like Figure 1 As shown, this embodiment provides a method for registering fire emergency lighting fixtures using a low-RAM centralized power supply. The executing entity in this embodiment is a controller in the fire emergency lighting system. This controller is connected to multiple centralized power supplies via a communication bus and is used to coordinate the registration process of all lighting fixtures in the network. Specifically, the method includes the following steps.
[0028] Step S100: Obtain the first loop registration instruction and send it sequentially to multiple centralized power supplies based on a preset time interval to obtain the bus interaction result.
[0029] It's important to note that low-RAM centralized power supplies refer to centralized power supply devices limited by low-cost hardware architecture, whose internal random access memory (RAM) capacity is extremely small, making it impossible to cache lighting data for two or more circuits simultaneously. This hardware constraint directly leads to the limitation of dedicated storage for a single circuit; that is, the same centralized power supply can only perform lighting registration work for one circuit at a time, and can only temporarily store the lighting information for the circuit currently being registered. For example, in an engineering scenario with a network of hundreds of centralized power supplies, a single centralized power supply may support one to eight lighting circuits, with each circuit supporting a maximum of over two hundred lighting fixtures, but its internal storage space is only sufficient to accommodate the data volume of a single circuit.
[0030] Specifically, the overall intent of obtaining the first loop registration command and issuing it sequentially based on a preset time interval is to achieve parallel startup of multiple power supplies with overlapping time intervals, rather than the traditional serial registration of each device one by one. In the traditional mode, the controller must wait for all loops of the current centralized power supply to complete registration before sending a command to the next centralized power supply, which will result in a large number of devices being in a passive idle state. In this embodiment, the controller issues the first loop registration command to each device one by one according to the preset time interval, so that the first loop registration operations of multiple centralized power supplies form an overlapping interval on the time axis. The preset time interval can be configured according to the bus type used. For example, for CAN bus or RS-485 bus, the time interval can be set to about ten milliseconds to ensure sufficient bus arbitration and response intervals between multiple frames of commands and avoid bus congestion. In this way, the controller collects the response messages or timeout feedback returned by each centralized power supply during the issuance process, thereby summarizing the bus interaction results reflecting the initial response status of all devices in the network.
[0031] Step S200: Based on the bus interaction results, identify the target centralized power supply among multiple centralized power supplies that has completed registration in the current loop and whose lighting data is pending processing.
[0032] Specifically, due to the extremely limited storage space of low-RAM centralized power supplies, once the lights in a certain loop have completed registration, the relevant data will fill its limited buffer space. If the controller cannot retrieve this data in time, the centralized power supply will not be able to free up space to start the registration process for the next loop. Therefore, the core purpose of identifying target centralized power supplies is to promptly discover those devices that have completed the registration of the current loop but whose data has not yet been transferred, so as to retrieve the data as soon as possible and thus release the limited buffer space of the centralized power supply. By analyzing the bus interaction results, the controller can objectively determine which centralized power supplies have completed the registration of the current loop and are in a state of waiting for the controller to read the data, and then lock them as target centralized power supplies. This identification mechanism is entirely driven by the software scheduling logic on the controller side and does not require the centralized power supply to have the ability to actively report or manage complex states.
[0033] Step S300: Receive and process the current circuit lighting data of the target centralized power supply.
[0034] Specifically, once the controller identifies the target centralized power supply, it interacts with it to transfer the current circuit lighting data temporarily stored by the centralized power supply to the controller for processing. This process is essentially a data transfer operation, resolving the congestion caused by the centralized power supply's full buffer by moving data from the resource-constrained centralized power supply to the relatively resource-rich controller. The processing may include data parsing, verification, and data storage to ensure the integrity and accuracy of the lighting information. Only after this data processing is complete are the centralized power supply's storage resources released, making it ready to execute subsequent tasks.
[0035] In step S400, in response to the completion of the current circuit lighting data processing and the existence of an unregistered circuit in the target centralized power supply, a registration instruction for the next circuit is issued to the target centralized power supply.
[0036] The next circuit registration instruction is used to instruct the target centralized power supply to perform luminaire registration on unregistered circuits.
[0037] Specifically, issuing the next loop registration command is a crucial step in achieving seamless continuity among multiple loops within a single power supply. Due to the hardware constraint of dedicated storage for each loop within a centralized power supply, the registration of each loop can only be performed serially. This embodiment addresses this by immediately issuing a command for the unregistered loop to the centralized power supply as soon as the previous loop's data processing is complete. This allows the centralized power supply to receive new tasks instantly upon clearing its cache, thus achieving seamless continuity within a single power supply. For example, if a centralized power supply has four loops requiring registration, once the data for the first loop is retrieved and processed by the controller, the controller immediately issues the registration command for the second loop, without waiting for the fixed polling cycle to return. If all loops of the centralized power supply have been registered, no further commands are issued. This chain-triggered mechanism perfectly adapts to the hardware characteristics of centralized power supplies, ensuring the device is always operating at high efficiency.
[0038] like Figure 2As shown in the flowchart, this application's state machine-based adaptive chain-type rapid registration process for lighting fixtures is illustrated. First, the controller establishes a global inspection queue containing multiple centralized power supplies and sequentially issues first-loop registration commands at preset time intervals, allowing the first-loop registration operations of multiple centralized power supplies to overlap in time, forming parallel startup between power supplies. Second, the controller maintains discrete state machines for each centralized power supply based on bus interaction messages, identifies target centralized power supplies that have completed current loop registration but whose lighting fixture data is pending processing, and prioritizes receiving and processing the current loop lighting fixture data of the target centralized power supply to release the limited buffer of low-RAM centralized power supplies. Third, when the current loop lighting fixture data processing is complete but the centralized power supply still has unregistered loops, the controller immediately issues the next loop registration command without waiting for the next fixed-cycle inspection, realizing multi-loop chain-type serial connection within a single power supply. Finally, the controller removes centralized power supplies that have completed full loop registration and those with long-term communication anomalies from the global inspection queue, dynamically reducing the size of the inspection queue and improving the response speed of the remaining uncompleted centralized power supplies. Figure 2 Overall, it embodies a closed-loop scheduling logic that features parallel operation between power supplies, chained serial operation within power supplies, real-time state machine-driven operation, and dynamic queue slimming.
[0039] Based on the above technical solution, this embodiment constructs a two-layer scheduling architecture of parallel startup between power supplies and serial connection within power supplies. In the first-layer scheduling, the controller uses a preset time interval to issue commands in batches, enabling multiple centralized power supplies to enter the working state synchronously, fully utilizing the parallel processing capabilities of the network devices. In the second-layer scheduling, the controller strictly executes the serial connection logic of immediately triggering the next loop after data is taken from a single centralized power supply, strictly adhering to its single-loop exclusive hardware constraint. The entire scheduling process relies entirely on the software algorithm on the controller side, without increasing the hardware cost of the centralized power supply or modifying its underlying firmware. Through this two-layer scheduling architecture, this application solves the problem of long-term device idleness in the traditional whole-machine serial registration mode, eliminates idle waiting latency in large-scale networking scenarios, and significantly shortens the overall time for one-click registration of all network lighting fixtures.
[0040] Example 2 Building upon Example 1, this example further details the underlying communication interaction, status determination logic, and data frame-level fault tolerance mechanism of a single centralized power supply. To support the stable operation of the aforementioned two-layer scheduling architecture, the controller needs to establish accurate operational profiles for each device connected to the bus. Specifically, the method in this example further includes: maintaining a registration context for each centralized power supply, the registration context including the power supply address, current registration status, current loop number, number of processed loops, number of remaining unregistered loops, consecutive no-response count, consecutive verification failure count, and timeout timer; wherein, the registration context is maintained only based on real-time interaction with the controller; updating the registration context based on the results of proactive inspection queries and real-time bus message interaction; and, based on the updated registration context, performing at least one of the following: identifying the target centralized power supply, receiving and processing the current loop lighting data, and issuing the next loop registration instruction.
[0041] Specifically, the registration context is a data structure area independently allocated by the controller in memory for each centralized power supply, used to record the device's full lifecycle status within the current registration cycle. The power address uniquely identifies a physical device in the bus network; for example, it could be a slave address on an RS-485 bus or a node identifier on a CAN bus. The current registration status reflects the current operational stage of the device. The current loop number and the number of processed loops together characterize the device's registration progress. The remaining number of unregistered loops directly determines whether further instructions need to be issued. In particular, the consecutive no-response count and consecutive verification failure count are core indicators of the health of the communication link. When the controller issues a query but does not receive a response within a specified time, the consecutive no-response count is incremented by one; when a message is received but data integrity verification fails, the consecutive verification failure count is incremented by one. The timeout timer is used to define the maximum allowable time for a single registration operation, preventing the entire scheduling process from being suspended due to device crashes.
[0042] In some implementations, the registration context is maintained solely through real-time interaction with the controller. This means the controller doesn't need prior information such as the number of lights in each loop, historical registration times, or bus load statistics. In a blank scenario during initial on-site debugging, with no pre-configured settings, traditional intelligent scheduling algorithms often fail due to a lack of historical sample data. This embodiment, however, does not rely on prior data. The controller dynamically updates the fields in the registration context solely based on feedback received from its own proactive inspection queries and the real-time captured message interaction results on the bus. For example, each time the controller sends a status query message, it updates the consecutive no-response count based on whether a response is received; each time it receives a completion message, it updates the current loop number based on the parsed content. This purely real-time driven approach gives the scheduling method strong engineering feasibility and environmental adaptability.
[0043] After establishing the registration context as the underlying data structure, the controller can then construct an objectively identifiable discrete state machine. Combined with... Figure 3 The state transition diagram shown illustrates the process of identifying a target centralized power supply among multiple centralized power supplies that has completed registration in the current loop and whose lighting data is pending processing. Specifically, this includes: receiving current loop completion messages from multiple centralized power supplies; performing consistency checks, address matching, and loop number verification on the current loop completion messages; determining that the corresponding centralized power supply has completed registration in the current loop and whose lighting data is pending processing in response to the current loop completion messages passing consistency checks, address matching, and loop number verification; marking the corresponding centralized power supply as a communication abnormality stagnation state in response to the bus interaction result meeting preset abnormality conditions; and restoring the registration state of the corresponding centralized power supply in response to receiving a valid message from a centralized power supply in the communication abnormality stagnation state.
[0044] like Figure 3 As shown, the controller defines four standardized discrete states for each centralized power supply: loop registration running state, loop registration completed and pending state, full loop registration completed state, and communication abnormal standby state. These four states constitute a complete state space that the controller can recognize. Specifically, when the controller issues a registration command for a loop to a centralized power supply, the device enters the loop registration running state. In this state, the controller continuously interacts with it through the bus. When the controller receives a current loop completion message returned by the device, it does not immediately accept it blindly, but must perform a strict triple verification. First, consistency verification, such as performing cyclic redundancy check or checksum comparison on the message, ensures that no bit flips or data corruption have occurred during the message transmission on the bus; second, address matching, that is, checking whether the source address carried in the message is consistent with the power supply address recorded in the registration context, to prevent bus crosstalk from causing misidentification of messages from other devices; finally, loop number verification, confirming that the completed loop number indicated in the message is indeed equal to the currently executing registration loop number, avoiding timing errors. Only when all three verifications pass will the controller flip the device's state from loop registration running state to loop registration completed and pending state in the registration context, thereby locking it as the target centralized power supply.
[0045] In some implementations, if the bus interaction result meets preset abnormal conditions, such as the consecutive no-response count in the registration context reaching a set threshold (e.g., no response after three consecutive queries), or the consecutive verification failure count exceeding the tolerance limit (e.g., five consecutive message verification failures), or the timeout timer overflowing, the controller will mark the centralized power supply as a communication abnormal standstill state. Figure 3As indicated by the arrows, regardless of whether the device is in the loop registration running state or the loop registration completed and pending processing state, once the above abnormal conditions are triggered, it will jump to the communication abnormal stop state.
[0046] In some implementations, this embodiment also incorporates an anomaly recovery mechanism. When a device is in a communication anomaly state, the controller does not completely discard it. If, during subsequent frequency reduction inspections, the controller receives a valid message returned by the device (i.e., a message that has passed the aforementioned consistency check, address matching, and loop number verification), it will immediately restore the device's registration status, returning it to the loop registration running state or the loop registration completed and pending processing state. This anti-false rejection mechanism effectively addresses temporary communication interruptions caused by electromagnetic interference, loose wiring, or other temporary factors at the construction site, preventing occasional faults from being misjudged as permanent failures and improving the system's robustness.
[0047] Once the controller identifies the target centralized power supply in the pending state after loop registration is completed using the aforementioned state machine, it enters the data transfer phase. Specifically, receiving and processing the current loop lighting data of the target centralized power supply includes: sending a data upload request to the target centralized power supply in the current loop registration completed and lighting data pending processing; receiving multiple data frames returned by the target centralized power supply in response to the data upload request, each data frame carrying the current loop lighting data; performing verification processing on each data frame for the power supply address, loop number, frame sequence number, and checksum; in response to the failure of verification processing for any data frame, initiating a retransmission request for the data frame that failed verification processing to the target centralized power supply, and receiving the retransmitted data frame; and in response to multiple data frames passing verification processing, writing the current loop lighting data into local storage to determine that the current loop lighting data processing is complete.
[0048] It should be noted that due to the extremely limited storage space of low-RAM centralized power supplies, a single circuit may carry information for over two hundred lamps. This data cannot be transmitted in a single message, therefore a framed transmission method must be used. The controller first sends a data upload request to the target centralized power supply, which then splits the lamp data for the current circuit into multiple data frames and sends them sequentially. During reception, the controller employs a strategy of receiving and verifying simultaneously, performing fine-grained quadruple verification on each arriving data frame: checking the power supply address to prevent data mismatch, checking the circuit number to ensure correct data attribution, checking the frame sequence number to ensure the sequential order of data splicing, and calculating a checksum to verify the integrity of the payload.
[0049] In some implementations, this embodiment incorporates a frame-level fault tolerance mechanism. In traditional solutions, if a bit in a data loop becomes incorrect during transmission, it often requires requesting the central power supply to retransmit all data for the entire loop, resulting in significant resource waste when bus bandwidth is limited. However, in this embodiment, if the verification process for a data frame fails, the controller only initiates a retransmission request for that specific failed frame, requiring the central power supply to retransmit it individually. The remaining frames that have passed verification do not need to be retransmitted. For example, suppose the data for a loop is split into twenty data frames, and the fifth frame has a mismatched checksum due to bus noise. The controller only needs to request a retransmission of the fifth frame, instead of retransmitting all twenty frames. This fine-grained retransmission strategy significantly reduces bus overhead and shortens the overall data transfer time.
[0050] After all data frames have successfully passed the verification process, the controller reassembles and parses these multiple data frames carrying the current circuit lighting data, and writes them to the controller's local storage. This local storage can be non-volatile memory on the controller's motherboard, a flash memory chip, or an external database file. Writing data to local storage has a dual significance: firstly, it signifies the completion of the controller's processing of the circuit lighting data, fulfilling the prerequisite for triggering the next circuit registration command; secondly, and more importantly, the successful transfer of data to the controller means that the minimal buffer space previously occupied within the centralized power supply is released. For centralized power supplies with low RAM, the hardware only has the physical conditions to receive the next circuit registration command and begin new circuit operations after its internal buffer is cleared. Therefore, timely data write-to-disk storage is a crucial step in ensuring seamless operation of the single-power supply's internal chain serial connection.
[0051] Example 3 Building upon Example 1, this example further details the scheduling mechanism for multi-power supply global resource allocation and queue management. Example 1 constructed a two-layer scheduling architecture, while this example focuses on addressing the question of which device the controller should prioritize querying and when to stop querying a particular device during the operation of this architecture.
[0052] Combination Figure 4The diagram illustrates the principle of dynamic inspection queue priority reordering and two-dimensional adaptive queue slimming. The method in this embodiment further includes: obtaining the number of remaining unregistered circuits, current registration status, and communication response stability of each of the multiple centralized power supplies; dynamically adjusting the inspection order of the multiple centralized power supplies based on the number of remaining unregistered circuits, current registration status, and communication response stability; wherein, centralized power supplies with completed current circuit registration and pending lamp data have a higher priority in the inspection order than centralized power supplies in other states, centralized power supplies with fewer remaining unregistered circuits have a higher priority in the inspection order than centralized power supplies with more remaining unregistered circuits, and centralized power supplies in a state of abnormal communication stagnation are postponed or have their inspection frequency reduced in the inspection order.
[0053] Specifically, the collection of the above three evaluation dimensions relies entirely on the real-time recording of the controller, without any prior data support. For the number of remaining unregistered loops, the controller can directly obtain it by reading the remaining unregistered loop number field recorded in the registration context mentioned in Example 2. This value decreases as each loop registration is completed. The current registration status is one of the four standardized states maintained by the discrete state machine on the controller side, such as loop registration running status, loop registration completed and pending processing status, etc. For communication response stability, the controller can calculate it by statistically analyzing the response status of the most recent active inspection queries. For example, a sliding window mechanism can be used to record the proportion of times a valid message was successfully received in the last ten queries, or a comprehensive evaluation can be made based on the cumulative values of consecutive no-response counts and consecutive verification failure counts.
[0054] After obtaining the real-time parameters of the above three dimensions, the controller dynamically adjusts the inspection order using a hierarchical sorting rule. The design intent of this parameter-free fuzzy priority evaluation system is that during the initial commissioning at the project site, the system has no historical load or timing prediction data, and traditional load balancing algorithms are ineffective due to a lack of input samples. This embodiment establishes the following hierarchical rules through rigorous physical logic derivation. First, centralized power supplies that have completed current loop registration and whose lighting data is pending are given the highest priority. This is because low-RAM centralized power supplies have extremely limited storage space; once a loop is registered, its limited cache is filled. If the controller does not prioritize retrieving its data, the device will be blocked due to storage overflow and unable to start any subsequent loop operations. Therefore, it must be processed with the highest priority to prevent low-RAM overflow blocking. Second, centralized power supplies with a small number of remaining unregistered loops have the next lower priority. From a systems engineering perspective, prioritizing devices that are about to complete all registrations can accelerate the closing loop of a single device, allowing it to exit the entire registration process as quickly as possible, thereby releasing bus resources faster. Finally, centralized power supplies that are in a state of abnormal communication interruption are postponed or have a reduced inspection frequency in the inspection sequence. Since such devices are likely to have loose wiring or hardware failures, sending frequent query messages to them will not only fail to get an effective response, but will also waste valuable bus bandwidth. Reducing invalid bus usage is an important means to ensure the overall registration efficiency of the entire network.
[0055] As a preferred implementation, when multiple centralized power supplies are at the same priority level, the controller also needs to perform secondary sorting to determine their order. For example, assuming three centralized power supplies are all in the loop registration operation state, and each has two remaining unregistered loops, the controller will compare their communication response stability and prioritize the device with the higher sliding window response rate, because the device with the higher response rate has a higher data transmission success rate and can complete subsequent data transfer more smoothly. If the communication response stability is also the same, they can be arranged in order from most recent valid response time to oldest, or degenerate into arranging them in natural ascending order of power supply addresses. This multi-level progressive sorting strategy ensures that the inspection queue can generate a unique and definite order under any circumstances, avoiding deadlock or random oscillations in the scheduling logic.
[0056] After resolving the question of which power source to poll first, this embodiment further addresses the question of when not to poll. Specifically, the method in this embodiment further includes: constructing a global inspection queue containing multiple centralized power sources; in response to any centralized power source entering a full-loop registration completion state, or the centralized power source being in a communication abnormal stagnation state and failing to receive a valid message for a preset duration, removing the centralized power source from the global inspection queue; updating the inspection order based on the removed global inspection queue, and shortening the inspection interval for the remaining centralized power sources in the global inspection queue; wherein, the removed centralized power source no longer participates in subsequent polling, and a corresponding fault device record is generated, which is used to support reactivation.
[0057] The global inspection queue is a dynamic data structure maintained in memory by the controller, initially containing all online centralized power supply nodes on the bus. The dual-dimensional adaptive queue slimming described in this application refers to simultaneously reducing the queue size in two directions: normal completion and abnormal stagnation. In the normal completion dimension, when all loops of a centralized power supply have been registered and its state machine transitions to the full loop registration completion state, the controller immediately removes it from the global inspection queue. In the abnormal stagnation dimension, the removal operation is not performed immediately when a communication failure occurs, but rather under strict trigger conditions. When a centralized power supply is in a communication abnormal stagnation state, the controller first performs a frequency-reduced inspection, for example, reducing the frequency of queries from every round to once every five or ten rounds. If, within a preset duration, such as three minutes, or after several rounds of frequency-reduced inspections, the controller still does not receive a valid message from the device, it is determined to be a long-term failure and permanently removed from the global inspection queue. This strategy of first lowering the frequency to test and then finally eliminating the faulty equipment can effectively prevent normal equipment from being mistakenly eliminated due to brief electromagnetic interference at the construction site.
[0058] When a centralized power supply is removed from the global inspection queue, the controller updates the inspection order based on the latest queue after the removal and correspondingly shortens the inspection interval for the remaining centralized power supplies in the global inspection queue. Assuming the initial global inspection queue contains one hundred centralized power supplies, the controller needs to send query messages to one hundred nodes for one round of inspection, resulting in a long inspection time per round. As the registration process progresses, devices are continuously removed from the queue due to completed full-loop registration or prolonged abnormal stagnation, gradually reducing the effective queue size from one hundred to eighty, fifty, or even a dozen. Because the total number of nodes requiring querying decreases, the absolute time required for the controller to complete a single round of inspection also shortens linearly. This means that for centralized power supplies still in the normal registration process but not yet completed, the waiting interval for them to be queried again by the controller becomes shorter and shorter, and the response speed of data reporting is improved step by step. This acceleration effect brought about by queue shrinkage is dynamically cumulative; the later the registration process, the faster the registration progress of the remaining devices.
[0059] Furthermore, for centralized power supplies removed due to abnormal stagnation, the controller generates a corresponding fault device record. This record can be stored in the controller's non-volatile memory or saved as a log file in media such as flash memory chips or local databases. The fault device record details the device's power address, its status information at the time of removal, and historical communication failure statistics to support subsequent reactivation. For example, after on-site construction personnel have investigated and repaired the device's wiring problem, they can issue a reactivation command through the controller's human-machine interface or host computer software. The controller reads the information from the fault device record and reinstates it to the global inspection queue to continue the registration process without requiring a power outage, restart, or reinitialization of the entire system.
[0060] Example 4 like Figure 5 As shown, based on Embodiment 1, this embodiment further details the two-layer timing drive mechanism of the system's core timing control. Embodiment 1 established the overall architecture of parallel startup between power supplies and chained serial connection within power supplies, while this embodiment focuses on defining the non-blocking boundary of batch deployment of multiple power supplies and the zero-window connection logic of passive triggering within a single power supply from the perspective of underlying timing engagement.
[0061] This embodiment details the process of obtaining the first loop registration command and sequentially sending it to multiple centralized power supplies based on a preset time interval to obtain the bus interaction result. Specifically, the process of sequentially sending the command to multiple centralized power supplies based on the preset time interval to obtain the bus interaction result includes: sending the first loop registration command to each of the multiple centralized power supplies at preset time intervals; after sending the first loop registration command to the current centralized power supply, without waiting for the current centralized power supply to complete the first loop registration, immediately sending the first loop registration command to the next centralized power supply; and obtaining the bus interaction result based on the response messages returned by each centralized power supply during the sequential sending process.
[0062] The above steps constitute the underlying timing specifications for the first layer of the two-layer scheduling architecture, namely the parallel startup of power supplies. The engineering significance of the preset time interval lies in reserving a necessary signal stability window for the physical bus. For example, for the CAN bus, this time interval can be set to approximately ten milliseconds to ensure sufficient bus arbitration time and node response interval between two adjacent instruction frames, avoiding data collisions or congestion caused by multiple devices simultaneously competing for the bus. For the RS-485 bus, due to the use of half-duplex communication mode, this time interval also needs to guarantee the transmit / receive conversion time during master-slave switching to prevent level conflicts caused by multiple centralized power supplies simultaneously uploading data.
[0063] In some implementations, this embodiment employs a non-blocking batch delivery method. After the controller sends the first circuit registration command to the current centralized power supply, it does not suspend the current task to wait for the centralized power supply to complete the entire first circuit's lamp scanning and registration process, as in the traditional serial mode. Instead, it immediately continues to send the first circuit registration command to the next centralized power supply at preset time intervals. This non-waiting boundary condition means that the controller's delivery action is only limited by the frame interval requirements of the bus physical layer, not by the execution time of the centralized power supply's service layer. For example, assuming that it takes three seconds for a single centralized power supply to complete the registration of one circuit, if the traditional serial waiting mode is used, one hundred centralized power supplies would need to spend three hundred seconds of idle waiting time just in the first circuit startup phase; however, under the non-blocking mechanism of this embodiment, the controller only needs about one second to complete the delivery of all the first circuit registration commands, and the first circuit registration operations of one hundred centralized power supplies form completely overlapping parallel intervals on the time axis. During the process of distributing commands to each device, the controller continuously listens for and collects response messages from each centralized power supply, such as response frames confirming receipt of instructions or status frames reporting progress. These real-time feedbacks are then aggregated to obtain the bus interaction results reflecting the initial response status of the entire network. This design eliminates the problems of long-term device idle time and latency accumulation caused by serial operation of the entire device in the traditional mode.
[0064] After completing the parallel startup between power supplies, this embodiment further details the underlying timing specifications of the second layer in the two-layer scheduling architecture, namely the internal chained serial connection of power supplies. Specifically, in response to the completion of the current circuit's lighting data processing and the existence of unregistered circuits in the target power supply, a registration instruction for the next circuit is issued to the target power supply. This includes: determining whether there are unregistered circuits in the target power supply in the callback or interrupt node when the current circuit's lighting data processing is completed; in response to the existence of unregistered circuits in the target power supply, issuing a registration instruction for the next circuit immediately without waiting for the next fixed-cycle inspection; and in response to the absence of unregistered circuits in the target power supply, marking the target power supply as having completed full circuit registration. The subsequent registration startup of each circuit in the target power supply is passively triggered by the completion event of the previous circuit's lighting data processing.
[0065] The above steps constitute a timing engagement mechanism specifically designed for low RAM centralized power supply hardware constraints in this application. At the software implementation level, a callback or interrupt node refers to a specific code location triggered when the function handling the data reception task in the controller-side software system completes execution, or a hardware event notification thrown by the underlying operating system when the data transfer task ends. For example, when the controller transfers the lighting data of the centralized power supply to local memory via DMA and completes verification and database entry, the system automatically calls a pre-registered callback function; or, when the serial port receive interrupt service routine confirms that the last frame of data has been completely received, it triggers the corresponding interrupt handling process. This embodiment embeds the logic for issuing the next-loop registration instruction into these specific software execution nodes.
[0066] This passive triggering mechanism differs fundamentally from conventional asynchronous polling or simple interrupt responses. In conventional polling schemes, the controller typically queries each device sequentially at fixed time intervals. Even if a device has finished processing its data, a new instruction will not be sent to it until the next polling cycle arrives, creating an unavoidable waiting period between polls. In this embodiment, however, once a callback or interrupt node is entered, the controller first reads the registration context mentioned in Embodiment 2 to determine if there are any unregistered loops in the target power supply. If an unregistered loop exists, the controller does not wait for the next fixed-cycle inspection and immediately issues the next loop registration instruction to the target power supply. This immediate triggering means that the instruction issuance action and the data processing completion action are tightly synchronized in timing. Because a low-RAM centralized power supply can only cache the lighting data of one circuit at a time, the internal cache is instantly cleared as soon as the data of the previous circuit is retrieved by the controller and written to local storage. At this moment, the registration instruction for the next circuit arrives immediately, allowing the centralized power supply to start scanning the new circuit without any idle waiting time. This achieves seamless continuity between multiple circuits within a single power supply with zero downtime. This mechanism perfectly adapts to the physical limitation of a centralized power supply's single-circuit dedicated storage, without breaking through hardware bottlenecks, and greatly improves equipment utilization.
[0067] In other implementations, if, during a callback or interruption, the controller queries the registration context and finds that all loops of the target centralized power supply have been registered (i.e., there are no unregistered loops), the controller will mark the target centralized power supply's status as "all loops registered successfully." This final state marking logic signifies that all work for this centralized power supply in this round of network-wide one-click registration task has been completed, providing a clear status basis for its subsequent removal from the global inspection queue.
[0068] Based on the above technical solution, this embodiment constructs a rigorous two-layer timing-driven mechanism by clearly defining the boundaries of batch distribution without waiting between power supplies and the boundaries of passive connection triggered immediately within a power supply. This mechanism, without changing the underlying firmware of the centralized power supply or increasing hardware costs, relies purely on software timing orchestration on the controller side to deeply integrate the parallel processing capabilities of multiple power supplies with the serial execution constraints of a single power supply, ensuring the scheduling efficiency and operational stability of rapid lamp registration in large-scale networking scenarios.
[0069] Example 5 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the low RAM centralized power supply fire emergency lighting fixture registration method as described in any one of Embodiments 1 to 4.
[0070] In practical engineering applications, the core control logic described in Examples 1 to 4, such as the two-layer scheduling architecture, discrete state machine identification mechanism, dynamic priority inspection queue scheduling, and two-dimensional adaptive queue slimming, all ultimately need to be converted into computer program code that can be executed by hardware. This program code needs to be attached to a specific physical storage medium so that it can be pre-installed and burned into the controller host of the fire emergency lighting system before leaving the factory, or distributed and deployed in the form of upgrade packages during the on-site debugging phase of the project.
[0071] Specifically, the computer-readable storage medium referred to in this application means any tangible entity capable of carrying and providing instructions or data to a processor. To fully support this overarching concept, several specific media forms are listed below as examples. For instance, the computer-readable storage medium can be read-only memory (ROM), a non-volatile medium often used to directly mask or burn the underlying scheduling firmware into the controller motherboard at the factory, ensuring that the registration program can run as soon as the device is powered on; it can also be random access memory (RAM), which serves as a temporary program storage medium when the controller loads the scheduler from an external source into memory for execution during operation; or it can be flash memory, such as NOR Flash or NAND Flash chips soldered on the controller motherboard, or portable flash memory devices such as pluggable SD cards, TF cards, and USB flash drives. This type of medium supports repeated erasing and rewriting, making it very suitable for updating the registration scheduling algorithm on-site via a bus interface by distributing firmware upgrade packages. In addition, the computer-readable storage medium can also be a magnetic disk, such as a traditional mechanical hard disk or floppy disk, which is usually used by the host computer server to store and distribute the engineering installation package containing the scheduler; or an optical disk, such as a CD-ROM, DVD, or other optical storage media, which is provided to the construction personnel for offline installation along with the equipment's instruction manual.
[0072] When the processor reads and executes the computer program from any of the aforementioned computer-readable storage media, it gains the ability to coordinate and schedule the entire network of lighting registration processes. Following the program instructions, the processor sequentially executes a series of actions, including obtaining the first circuit registration instruction and issuing it at preset time intervals, identifying the target centralized power supply, receiving and processing the current circuit's lighting data, and immediately issuing the next circuit registration instruction after data processing is complete. This effectively replicates the technical effects of parallel startup between power supplies and chained serial connection within power supplies at the physical hardware level.
[0073] It should be noted that the above enumeration of specific forms of computer-readable storage media is illustrative only and not restrictive. Those skilled in the art can choose any applicable type of storage media to carry the computer program based on the selection requirements of the actual controller hardware platform, without departing from the scope of protection of this application.
[0074] Example 6 This embodiment provides an electronic device, including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the low RAM centralized power supply fire emergency lighting fixture registration method as described in any one of Embodiments 1 to 4.
[0075] Specifically, Examples 1 to 4 detail the pure software scheduling logic for rapid registration of luminaires based on state machine adaptive chaining, executed by the controller. This example solidifies and extends the aforementioned software method at the physical entity level. In actual engineering applications, the electronic device referred to in this application is the controller host in the fire emergency lighting system. As the core hub of the entire network communication system, the controller host is responsible for coordinating the registration process of all luminaires with low RAM centralized power supplies connected to the bus. By transforming the aforementioned two-layer scheduling architecture, discrete state machine identification mechanism, dynamic priority inspection queue scheduling, and two-dimensional adaptive queue slimming into an executable computer program and deploying it on the hardware platform of the electronic device, the pure software-level algorithm innovation can run stably in a real physical environment.
[0076] The processor is the core computing unit of this electronic device, executing computational and control instructions. To fully support this higher-level concept, several specific hardware forms are listed below as examples. For instance, the processor could be a microcontroller unit (MCU), which typically integrates a central processing core and rich peripheral interfaces, making it ideal for real-time task scheduling in industrial control scenarios; it could also be a digital signal processor (DSP), which has higher computational efficiency when handling large amounts of data verification and message parsing tasks; or it could be a field-programmable gate array (FPGA) or an application-specific integrated circuit (ASIC), which can directly implement state machine transition judgments and timing engagement through underlying logic gate circuits, thereby further reducing the response latency of instruction issuance and callback triggering.
[0077] The memory is used to persistently or temporarily store the aforementioned computer program code and intermediate data generated during the registration process, such as the registration context and global inspection queue node information mentioned in Embodiment 2. In one implementation, the memory may include non-volatile storage media, such as read-only memory (ROM), flash memory, electrically erasable programmable read-only memory (EEPROM), or solid-state drive (SSD), to retain the scheduler's firmware code and fault device records even after the device is powered off. Simultaneously, the memory may also include volatile storage media, such as random access memory (RAM), static random access memory (SRAM), or dynamic random access memory (DRAM), to provide high-speed data read / write cache space when the processor executes the program, supporting the massive state update requirements when hundreds of centralized power supplies interact concurrently. Of course, in other embodiments, the memory may also be a removable storage medium such as an SD card or TF card plugged into the controller motherboard, allowing for rapid upgrades of the scheduling algorithm version on-site by replacing the storage card.
[0078] In terms of hardware topology, the electronic device establishes communication connections with multiple centralized power supplies through a bus interface integrated on its motherboard. Specifically, this bus interface can be an RS-485 serial communication interface, which achieves long-distance, interference-resistant master-slave half-duplex communication through differential signal transmission; or it can be a Controller Area Network (CAN) interface, which uses a multi-master arbitration mechanism to ensure the real-time delivery of high-priority messages. When the processor executes the computer program in memory, it drives these bus interfaces to send the first loop registration command to each device at preset time intervals, and performs fine-grained verification and local disk writing after receiving data frames, thereby completely replicating the method steps described in any one of Embodiments 1 to 4 at the physical level.
[0079] It should be noted that the above specific listing of processor types, memory formats, and bus interface types is illustrative only and not restrictive. Those skilled in the art can choose any suitable hardware combination to construct the electronic device based on the actual project's cost budget, performance requirements, and bus topology, without departing from the scope of protection of this application.
[0080] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. The specific parameter values, hardware selection, bus type, and state transition logic details described in the above embodiments are merely illustrative and not restrictive. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for registering a fire emergency luminaire with low RAM centralized power supply, characterized in that, include: Obtain the first loop registration instruction and send it sequentially to multiple centralized power supplies based on a preset time interval to obtain the bus interaction result; Based on the bus interaction results, identify the target centralized power supply among the multiple centralized power supplies that has completed registration in the current loop and whose lamp data is pending processing. Receive and process the current circuit lighting data of the target centralized power supply; In response to the completion of the current circuit lighting data processing and the existence of an unregistered circuit in the target centralized power supply, a next circuit registration instruction is issued to the target centralized power supply; wherein, the next circuit registration instruction is used to instruct the target centralized power supply to perform lighting registration for the unregistered circuit.
2. The method of claim 1, wherein the low-RAM centralized power fire emergency luminaire registration method is characterized by, The process of receiving and processing the current circuit lighting data of the target centralized power supply includes: Send a data upload request to the target centralized power supply that has completed registration in the current circuit and whose lamp data is pending processing. The target centralized power supply is received frame by frame in response to the data upload request, and the multiple data frames carry the current circuit lamp data; Perform verification processing on each data frame, including power address, circuit number, frame sequence number, and check value. In response to the failure of verification processing for any data frame, a retransmission request for the data frame for which verification processing failed is initiated to the target power supply, and the retransmitted data frame is received. In response to the fact that all of the multiple data frames have passed the verification process, the current circuit lighting data is written to local storage to determine that the current circuit lighting data processing is complete.
3. The method of claim 1, wherein the low-RAM centralized power fire emergency luminaire registration method is characterized by, The identification of the target centralized power supply among the plurality of centralized power supplies that has completed registration in the current circuit and whose lighting data is pending processing includes: Receive the current loop completion message returned by the multiple centralized power sources; For the current loop, perform consistency verification, address matching, and loop number verification on the message. In response to the current loop completion message passing the consistency check, the address matching and the loop number verification, it is determined that the corresponding centralized power supply is registered in the current loop and the lamp data is pending processing. In response to the bus interaction result meeting the preset abnormal conditions, the corresponding centralized power supply is marked as a communication abnormal stop state; In response to receiving a valid message from the centralized power supply that is in the communication abnormal stop state, the registration status of the corresponding centralized power supply is restored.
4. The method of claim 3, wherein the low-RAM centralized power fire emergency luminaire registration method is characterized by, Also includes: Obtain the number of remaining unregistered circuits, current registration status, and communication response stability of each of the multiple centralized power sources; Based on the number of remaining unregistered circuits, the current registration status, and the stability of the communication response, the inspection sequence of the multiple centralized power supplies is dynamically adjusted. In this inspection sequence, centralized power supplies that have completed the current circuit registration and whose lamp data is pending have a higher priority than centralized power supplies in other states. Central power supplies with fewer remaining unregistered circuits have a higher priority than centralized power supplies with more remaining unregistered circuits. Central power supplies in the communication abnormal stop state are postponed or have their inspection frequency reduced in the inspection sequence.
5. The method for registering fire emergency lighting fixtures with low RAM centralized power supply according to claim 4, characterized in that, Also includes: Construct a global inspection queue that includes the multiple centralized power sources; In response to any centralized power supply entering the full-loop registration completion state, or the centralized power supply being in the communication abnormal stagnation state and not receiving the valid message for a preset duration, the centralized power supply is removed from the global inspection queue; Based on the removed global inspection queue, the inspection order is updated, and the inspection interval for the remaining centralized power supplies in the global inspection queue is shortened. The removed centralized power supply no longer participates in subsequent polling and generates a corresponding fault device record, which is used to support reactivation.
6. The method for registering fire emergency lighting fixtures with low RAM centralized power supply according to claim 1, characterized in that, The process of sequentially sending data to multiple centralized power supplies based on a preset time interval to obtain bus interaction results includes: The first circuit registration instruction is sent to each of the multiple centralized power supplies at the preset time interval; After issuing the first circuit registration instruction to the current centralized power source, the first circuit registration instruction is issued to the next centralized power source without waiting for the current centralized power source to complete the first circuit registration. The bus interaction result is obtained based on the response messages returned by each centralized power supply during the process of distributing the data to each unit.
7. The method for registering fire emergency lighting fixtures with low RAM centralized power supply according to claim 1, characterized in that, The step of issuing a next circuit registration instruction to the target centralized power supply in response to the completion of the current circuit lighting data processing and the existence of an unregistered circuit in the target centralized power supply includes: In the callback or interruption node where the current circuit lighting data processing is completed, it is determined whether the unregistered circuit exists in the target centralized power supply. In response to the presence of the unregistered circuit in the target centralized power supply, the next circuit registration instruction is immediately issued to the target centralized power supply without waiting for the next round of fixed-cycle inspection; In response to the absence of the unregistered circuit in the target centralized power supply, the target centralized power supply is marked as having completed full circuit registration. The registration and activation of subsequent circuits of the target centralized power supply are passively triggered by the completion of the data processing of the previous circuit's lighting fixtures.
8. The method for registering fire emergency lighting fixtures with low RAM centralized power supply according to claim 1, characterized in that, Also includes: A registration context is maintained for each of the centralized power supplies. The registration context includes the power supply address, current registration status, current loop number, number of processed loops, number of remaining unregistered loops, consecutive no-response count, consecutive verification failure count, and timeout timer. The registration context is maintained only based on real-time interaction with the controller. The registration context is updated based on the results of proactive inspection and real-time bus message interaction. Based on the updated registration context, at least one of the following is performed: identifying the target centralized power supply, receiving and processing the current circuit lighting data, and issuing the next circuit registration instruction.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a method for registering fire emergency lighting fixtures with a low RAM centralized power supply as described in any one of claims 1 to 8.
10. An electronic device comprising a processor and a memory, characterized in that, The memory stores a computer program, which, when executed by the processor, implements a method for registering fire emergency lighting fixtures with a low RAM centralized power supply as described in any one of claims 1 to 8.
