A multi-point high-precision coordinated ignition system for fireworks display

Abstract

The present application belongs to the technical field of distributed cooperative control system, and discloses a multi-point high-precision cooperative ignition system for firework display, comprising: a central controller which non-timely distributes delay time stamps to a plurality of distributed ignition modules and broadcasts a content-neutral synchronization event at a starting moment; the distributed ignition module starts a local timer upon detecting the synchronization event and autonomously counts down according to the pre-stored delay time stamp to execute ignition, the present application decentralizes intelligent time decision to the terminal, avoids the interference of signal physical transmission delay on the overall ignition precision of the system, and replaces the complex real-time clock synchronization and delay compensation mechanism in the traditional control system, so that the architecture simplifies the physical design and deployment complexity of the system, and improves the synchronization robustness and scalability when a large number of nodes work cooperatively.

Description

TECHNICAL FIELD

[0001] The present application relates to a multi-point high-precision coordinated ignition system for fireworks display, belonging to the technical field of distributed coordinated control system. BACKGROUND

[0002] In the current control system engineering practice, it is a basic and key technical task to realize the coordinated action of a large number of remotely distributed execution units at a preset time. Generally, it relies on a central control mode of synchronous triggering, that is, a central controller broadcasts data packets containing address and instruction information to all execution units to realize the synchronous control of each unit. This mode is clear in logic and provides a feasible solution for many conventional applications.

[0003] However, when the scale of the control system expands to tens of thousands of nodes and the physical span reaches several kilometers, while requiring all actions to be completed in microsecond-level time accuracy, the inherent physical limitations of the aforementioned control mode will be revealed. The synchronization accuracy of the system is bound to the ability to accurately measure and compensate the physical transmission delay of the signal, which leads to a principled dilemma in the design and deployment of the system. In order to pursue higher synchronization accuracy, more investment in communication links and clock synchronization hardware is required, which increases the complexity of the system, the deployment cost and the operational vulnerability.

[0004] To alleviate this problem, the industry has also explored the idea of using high-precision clock sources, such as the Global Positioning System, to provide a unified time reference for each execution unit. However, this approach not only increases the hardware cost and power consumption of each node, making it difficult to popularize in cost-sensitive applications, but also challenges the reliability of the system in poor signal reception environments. It does not fundamentally solve the core contradiction between cost and reliability faced by large-scale inclusive high-precision coordinated control systems. Specifically, the existing technology mainly has the following deficiencies: 1. The synchronization accuracy of the system is severely dependent on the instantaneous health status of the communication link and the environmental stability. Any temporary electromagnetic interference or poor interface contact can disrupt the accurate arrival of a signal, leading to synchronization failure; 2. As the number of system nodes and the physical range increase, the delay compensation data that needs to be calculated and managed by the central controller becomes increasingly large, and the control logic becomes increasingly complex, limiting the scalability of the system; 3. The high-speed communication network high-precision clock synchronization module and other hardware used to achieve high-precision synchronization result in high overall system cost, restricting its application in a wider field. Therefore, how to build a control system architecture that can enable large-scale distributed nodes to achieve high-precision coordinated action without relying on expensive hardware and complex delay compensation algorithms, and make the synchronization accuracy of the system inherently avoid the interference of physical transmission delay, has become a technical problem to be solved by the present application. SUMMARY

[0005] The application provides a multi-point high-precision coordinated ignition system for fireworks display, which mainly aims to solve the problem that the existing distributed coordinated control system excessively relies on complex real-time communication and delay compensation mechanism to achieve high-precision synchronization, thereby resulting in complex system architecture and high cost.

[0006] To achieve the above-mentioned purpose, the application provides a multi-point high-precision coordinated ignition system for fireworks display, which comprises:

[0007] a central controller and a plurality of distributed ignition modules, the central controller and the plurality of distributed ignition modules are connected via a pair of shared DC power supply lines, the central controller is configured to generate a delay timestamp for each distributed ignition module according to a fireworks display script, and modulate the delay timestamp via a power line carrier modulator and then distribute it via the DC power supply line in a non-timed manner; at the starting moment of the fireworks display, a voltage transient not carrying address and instruction information is applied on the voltage baseband of the DC power supply line by controlling a switching circuit as a synchronization event; after the synchronization event, the switching circuit continuously generates system metronome pulses at a unified frequency on the same DC power supply line;

[0008] Each distributed ignition module comprises a power line carrier demodulation module, a voltage comparator, a local timer and a microcontroller; the distributed ignition module is configured to obtain power via the DC power supply line, decode and store the delay timestamp via the power line carrier demodulation module; monitor the synchronization event via the voltage comparator, and when the synchronization event is monitored, reset the local timer to zero by the microcontroller and autonomously count down according to the stored delay timestamp to drive ignition; during the autonomous countdown, count the received system metronome pulses in parallel to obtain an external beat count value, and periodically adjust the timing step of the local timer according to the difference between the external beat count value and the internal beat count value generated based on the local timer.

[0009] Preferably, the central controller is configured to modulate the delay timestamp into a high-frequency data signal superimposed on the DC power supply line; the power line carrier demodulation module of the distributed ignition module is configured to filter out the baseband voltage of the DC power supply line and demodulate the high-frequency data signal to obtain the delay timestamp; the central controller is configured to generate a voltage sag with a determined amplitude and duration as the synchronization event; the voltage comparator of the distributed ignition module is configured to generate a hardware interrupt signal when a voltage sag matching the determined amplitude and duration is monitored, and the microcontroller responds to the hardware interrupt signal to perform the reset of the local timer.

[0010] Preferably, the microcontroller is configured to perform the adjustment of the timing step in each correction period, which specifically comprises: calculating a corrected timing step for the next timing period of the local timer, wherein, is a nominal timing step of the local timer before correction, is an external beat count value accumulated in the correction period, is an internal beat count value generated by the local timer in the same correction period, is a correction gain coefficient stored in the distributed ignition module.

[0011] Preferably, the distributed ignition module further comprises an analog-to-digital converter; the distributed ignition module is configured to, when the voltage comparator detects a voltage change, first instruct the analog-to-digital converter to sample the waveform of the voltage change to obtain the rise time and pulse amplitude of the voltage change; the microcontroller is further configured to: compare the obtained rise time and pulse amplitude with a health parameter template stored in the microcontroller, and only perform the reset operation of the local timer when the obtained rise time is less than the upper limit threshold of the rise time and the pulse amplitude falls within the pulse amplitude range.

[0012] Preferably, the distributed ignition module is further configured to, in a deployment self-checking phase, apply a detection current to the connected fireworks product with an identity coding resistor through the microcontroller, and instruct the analog-to-digital converter to measure the voltage drop generated thereby to calculate the resistance value of the identity coding resistor; and report the resistance value as physical type information to the central controller; the central controller is further configured to compare the received resistance value with a database stored in the central controller, which records the mapping relationship between the type of fireworks product and the resistance value, to confirm the correctness of the physical deployment before ignition execution.

[0013] Preferably, the central controller is configured to generate a voltage sag with an amplitude ranging from 20% to 40% of the system nominal supply voltage and a duration ranging from 5 microseconds to 20 microseconds.

[0014] Preferably, the central controller is further configured to, when it is confirmed that there is an error in the physical deployment, output an error report to the operation interface, the error report comprising the address of the distributed ignition module with the error, and the type of fireworks product that should be connected at the address and the type of fireworks product actually connected, which are parsed according to the mapping relationship.

[0015] Preferably, the central controller is further configured to broadcast a polling instruction after the distribution of the delay timestamp is completed, and enter a waiting state until receiving confirmation information from all distributed ignition modules indicating that the delay timestamp has been successfully received and stored, and then enter a working state in which it can broadcast a synchronization event.

[0016] Preferably, the distributed ignition module is further configured to send, through the power line carrier modulator, a status return message containing its own address and ignition execution status to the central controller after driving the ignition; and the central controller is further configured to monitor and record the status return message of all distributed ignition modules within a time window after broadcasting the synchronization event, and list the addresses of the distributed ignition modules that have not received successful return messages for subsequent fault diagnosis.

[0017] Preferably, the local timer is driven by a quartz crystal oscillator; and the microcontroller is configured to perform adjustment of the timing step to compensate for the frequency drift of the quartz crystal oscillator due to manufacturing tolerances and changes in the field environment temperature.

[0018] Compared with the prior art, the present application has the following beneficial effects:

[0019] 1. The central controller first distributes the delay timestamp of each distributed ignition module to the corresponding module in a non-timed manner, and then broadcasts a unified asynchronous synchronization trigger signal when execution is required; each module starts local timing at the moment of receiving the signal, and executes ignition after the local timing reaches the preset delay timestamp. In this way, the overall ignition accuracy of the system is no longer related to the time difference of the trigger signal in the physical transmission path, but depends on the stability of the internal timer of each distributed module. The complex clock synchronization and delay compensation link that must be set in the original control system to cope with the uncertainty of signal transmission time becomes unnecessary in this method.

[0020] 2. After broadcasting the asynchronous synchronization trigger signal, the central controller can also broadcast a metronome signal of a unified frequency to all distributed ignition modules through a public channel; each distributed ignition module receives and counts the metronome signal while performing autonomous countdown, and compares the internal beat count value generated based on the local timer with the count value of the metronome signal, and then dynamically corrects the timing rate of the local timer according to the deviation between the two values. This makes the timing pace of each module consistent with the high-stability clock reference of the central controller throughout the execution period, even if the module uses a low-cost timing element with limited long-term stability, avoiding the cumulative decline in synchronization accuracy caused by the inherent timing frequency drift of each unit.

[0021] 3. A deployment self-check step is added before the pre-setting stage. Each distributed ignition module first measures the electrical characteristics of the connected pyrotechnic products with preset identification codes to identify their physical type, and reports the results to the central controller for comparison to verify the correctness of the physical deployment. Furthermore, during the execution stage, each module, upon receiving an asynchronous synchronous trigger signal, also measures its waveform physical parameters in real time. Only when the measurement results match the preset health parameter template is the local timer started. This dual verification of the correctness of the physical deployment and the physical integrity of the trigger signal before execution avoids ignition errors caused by human wiring mistakes or on-site electromagnetic interference. The system utilizes a shared physical wire to connect the central controller to multiple distributed ignition modules. The central controller modulates the delayed timestamp into a high-frequency data signal for distribution, while the asynchronous synchronous trigger signal is a baseband electrical event, such as a momentary drop in power supply voltage. Each distributed ignition module receives power through this wire and can obtain the delayed timestamp and trigger signal through the high-frequency demodulation circuit and the baseband monitoring circuit, respectively. This method of carrying signals of different properties on different frequency bands of a single physical medium integrates the data network and synchronization signal network that originally needed to be deployed in parallel, simplifying the physical architecture of the system and the complexity of on-site construction. Attached Figure Description

[0022] Fig. 1 This is a block diagram of the signal flow and functional modules of the coordinated ignition system of the present invention;

[0023] Fig. 2 This is a schematic diagram of a wide-area deployment scenario for the collaborative ignition system of the present invention;

[0024] Fig. 3 This is a flowchart illustrating the closed-loop self-checking and error correction process for the physical deployment connection of this invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] The present invention provides a multi-point high-precision coordinated ignition system for fireworks displays. Its overall architecture consists of a central controller and multiple distributed ignition modules configured according to actual needs. The central controller and all distributed ignition modules are physically connected and exchange information via a pair of shared DC power lines running throughout the deployment site. The system's workflow is divided into a non-timed preset phase and a real-time execution phase. In the preset phase, the central controller's core task is to generate independent delay timestamps for each ignition module based on the display script, and distribute this timestamp data to the corresponding modules for storage using power line carrier modulation technology. In the execution phase, at a preset start time, the central controller applies a globally neutral voltage transient on the DC power line as a synchronization event to trigger all distributed ignition modules to immediately start their local timers and autonomously count down based on the pre-stored delay timestamps until the ignition action is finally completed.

[0027] After the system is first deployed or any changes occur in the physical network topology, in order to match the physical layer authentication mechanism of the synchronization event with the actual electrical characteristics of the field lines, a network-level health parameter template self-calibration procedure needs to be executed. This procedure first involves the central controller broadcasting a command to all connected distributed ignition modules to put them into a data acquisition mode. Subsequently, the central controller generates a statistically significant set of reference voltage sag event samples within a preset time period, for example, 100 consecutively. Each distributed ignition module in the network uses its analog-to-digital converter to instantaneously sample the waveform of each received sag event and calculates its actual pulse amplitude and rise time—two key characteristic parameters. After completing the acquisition of all samples, all modules transmit their measured pulse amplitude and rise time data back to the central controller. After aggregating tens of thousands of data points from the entire network, the central controller performs statistical distribution calculations on the pulse amplitude and rise time data, and takes the mean plus or minus six times the standard deviation. The result is taken as the effective range of pulse amplitude, and its mean is taken plus six times the standard deviation. The result is used as the upper limit threshold for rise time. Finally, this health parameter template, generated by actual field data, is distributed and solidified into all distributed ignition modules, serving as the basis for validating synchronization events in subsequent execution phases.

[0028] To ensure that the frequency drift correction logic of the local timers in each distributed ignition module has dynamic response characteristics, after the system completes the calibration of the aforementioned parameter templates, it also needs to perform a step on the correction gain coefficient. The iterative optimization procedure involves the central controller selecting a distributed ignition module at the physical end of the network as the calibration target. It then artificially introduces a known positive frequency offset, such as +50ppm, into the quartz crystal oscillator frequency of the module's local timer to simulate cumulative clock drift. Subsequently, the central controller initiates an iterative test program, starting from an initial safety value, for example… Initially, the calibration module performs dynamic timing step correction, and the central controller monitors its external beat count value. With internal beat count value The difference The time required for convergence from the initial frequency offset state to near zero is incremented by a fixed step size, for example, 0.01, after each test. The value is then used to repeat the above convergence test until an overshoot is detected for the first time during the convergence process, i.e., the difference changes from positive to negative or from negative to positive. At this point, in order to achieve a technical balance between convergence speed and system stability, the value that leads to the critical overshoot is taken. 90% of the value is used as the final working point in this deployment environment. This coefficient value is then distributed by the central controller and fixed to all distributed ignition modules in the network for dynamic adjustment of the timing step size during the subsequent autonomous countdown. In specific engineering implementation, a major challenge faced by large-scale distributed control systems in achieving microsecond-level coordinated actions over a wide physical range is that the transmission delay of signals in the physical medium is objectively present and uneven. This makes it impossible for traditional centralized synchronization commands to reach all nodes simultaneously. To address this challenge, the solution of this invention is configured as a spatiotemporally decoupled working mode. In the preset stage, the central controller first loads the fireworks display script, which defines that each ignition channel should ignite at the zero moment of the performance. After a certain time interval before the action is executed, the central controller compiles it into an ignition task list, whose data structure consists of distributed ignition module addresses and delay timestamps. For example, module address 101, 30.125s. Subsequently, the central controller modulates these data packets into high-frequency data signals and superimposes them onto the DC power supply line via a power line carrier modulator, broadcasting them to the entire network in a non-timing manner. Each distributed ignition module integrates a power line carrier demodulation module, which is configured to continuously monitor the high-frequency signals on the line and decode only data packets matching its own address, extracting the delay timestamp from them. The data is stored in its internal non-volatile memory. It should be noted that this data distribution process is not time-sensitive. Data packet transmission delays, jitter, and even retransmissions caused by interference do not affect the final synchronization accuracy. It is only necessary to ensure that all modules have correctly received and stored their task parameters before entering the execution phase. In this way, by sending out the timing information that determines when to ignite in advance, the task of establishing the time base, which has a decisive impact on the synchronization accuracy, is delegated from the central controller to each terminal execution unit.

[0029] To ensure that all distributed ignition modules have correctly loaded their independent delay timestamps before receiving synchronization events, the system implements a mandatory data reception integrity verification procedure after completing the data distribution in the preset phase. Specifically, after broadcasting the last data packet containing the delay timestamp via the power line carrier channel, the central controller immediately broadcasts a global polling command and pauses subsequent operations, entering a waiting state. Each distributed ignition module in the network, after successfully decoding and storing its delay timestamp in non-volatile memory, is configured to automatically send back an acknowledgment message containing its own address and data reception completion status. The central controller maintains an internal list of all registered modules and checks the received acknowledgment messages one by one. Only after confirming receipt of acknowledgment messages from all modules in the list within a preset timeout period, such as 5 seconds, does the central controller unlock itself and enter a ready working state where it can broadcast synchronization events. If any module fails to send back acknowledgment messages after the timeout, the system reports the address list of unacknowledged modules to the operation interface and remains locked until all modules have been acknowledged.

[0030] Furthermore, when the system enters the execution phase, to address the issue of establishing a unified logical starting point for signals arriving at physically different times across all distributed nodes, the system adopts the following procedure at the start of the fireworks display. The central controller applies a voltage transient without any address or instruction information to the baseband voltage of the power supply line by controlling a high-power electronic switching circuit connected in series with the DC power supply line. This transient is used as a global synchronization event. This voltage transient is deterministically defined as a voltage sag with definite physical properties. Its amplitude is set to drop by 20% to 40% from the system's nominal supply voltage, for example, 24V, i.e., a drop to between 14.4V and 19.2V. Its duration is calibrated to be between 5 microseconds and 20 microseconds. This parameter range is set because a sag that is too small or too short is easily drowned out by line noise, while a sag that is too large or too long may affect the normal power supply to the distributed ignition module. Therefore, this range achieves a technical balance between ensuring signal detectability and maintaining system power supply stability. Select the working window; accordingly, each distributed ignition module is equipped with a voltage comparator. The inverting input of the comparator is connected to a reference voltage, such as 20V, while the non-inverting input directly monitors the voltage of the DC power supply line. It is configured to generate a hardware interrupt signal when a voltage dip matching a determined amplitude and duration is detected. Once the microcontroller in the module responds to this hardware interrupt signal, it unconditionally resets its internal local timer, driven by a quartz crystal oscillator, to zero and immediately starts timing. Through this mechanism, a synchronous event that arrives at different times physically defines a logically unified execution starting point for all nodes in the entire distributed system, thereby changing the overall ignition accuracy of the system from dependence on physical transmission delay to dependence only on the stability of the local timers of each module.

[0031] However, during the several-minute-long ignition cycle, the quartz crystal oscillators used in each distributed ignition module experience slight but cumulative frequency drift due to inherent manufacturing tolerances and ambient temperature variations. This constitutes a secondary technical problem affecting the long-term synchronization accuracy of the system. To address this issue, this invention introduces a dynamic self-correcting time base maintenance mechanism. After the central controller broadcasts the aforementioned voltage transient as a synchronization event, it continues to generate system metronome pulses using the same switching circuit on the same DC power line at a uniform frequency, for example, 10.000Hz, driven by its own high-stability temperature-compensated crystal oscillator (TCXO). Inside each distributed ignition module, its microcontroller, while driving a local timer for autonomous countdown, simultaneously uses an input capture unit to count the received system metronome pulses to obtain an external metronome count value. And based on the external beat count value and an internal beat count value generated within the same time period based on its local timer. The difference is periodically adjusted, for example, every second, to change the timing step size of the local timer. The adjustment logic follows a closed-loop feedback correction algorithm, specifically calculating a corrected timing step size. The formula for calculating the next timing cycle for the local timer is determined as follows: ,in, It is the nominal timing step size of the local timer before calibration. It is the accumulated external beat count value (ideally 10) within this correction period, for example, 1 second. It is the internal tick count value generated by the local timer within the same correction cycle, while This is a pre-stored correction gain coefficient within the module. Its value (e.g., 0.1) is determined by testing modules with different drift rates during the system debugging phase to calibrate a value that can both quickly respond to drift and avoid system overshoot and oscillation. For example, if within a 1-second correction cycle... The count value is 10, and a certain module's problem is caused by the crystal oscillator being too fast. If the count value is 11, its timing step will be adjusted to... This mechanism adjusts the timing rate to align with the system's standard beat, compensating for the long-term instability of the hardware components through software.

[0032] To further improve the system's operational reliability in complex electromagnetic environments, this invention also incorporates a synchronization pulse health self-check gating mechanism. Given that strong electromagnetic interference in the field may distort the voltage sag waveform of synchronization events or generate spurious interference pulses, the distributed ignition module does not respond immediately when the voltage comparator detects a voltage change. Instead, its microcontroller first instructs an analog-to-digital converter to instantaneously sample the waveform of the voltage change at a sampling rate on the order of MHz to obtain the actual rise time and pulse amplitude of the voltage change. The microcontroller then combines these two measured parameters with a stored parameter containing the rise time and pulse amplitude. The upper limit threshold for time, such as 50 nanoseconds, is compared with a health parameter template, such as the aforementioned pulse amplitude range of 14.4V to 19.2V. A valid synchronization event is considered received only when the acquired rise time is less than the upper limit threshold and the pulse amplitude falls within the amplitude range, and a local timer reset operation is performed. If any parameter does not meet the template requirements, the microcontroller will actively suppress the start of the timer and record the waveform parameters of this abnormal event in the log for post-fault diagnosis. This pre-execution signal physical layer integrity verification step can effectively prevent the system from being misled by a damaged control signal.

[0033] Finally, to address the high risk of physical wiring errors due to human error in large systems, this invention adds a deployment self-test function to the initial stage of the system workflow. During this self-test, each distributed ignition module, through its microcontroller, applies a weak and safe probe current to each connected flame product equipped with an identification-coded resistor of a specific resistance value, and instructs its analog-to-digital converter to measure the resulting voltage drop, calculating the precise resistance value of the identification-coded resistor based on Ohm's law. Subsequently, the module reports this channel number and resistance value data as physical type information to the central controller via power line carrier communication. Upon receiving this information, the central controller... After obtaining the physical deployment information of all modules, it compares it point by point with a database stored within it that records the mapping relationship between fireworks product types and resistance values. This verifies the correctness of the physical deployment before ignition. If an error is found in the physical deployment, the central controller outputs an error report to the operation interface, which includes the address of the faulty module, the product type it should be connected to, and the actual product type connected. The system is only allowed to enter the subsequent ignition preparation state after the on-site engineer completes the correction based on the report and passes a new round of self-checks. This closed-loop verification procedure enables the system to verify the correctness of its physical deployment before execution, thereby systematically avoiding errors in the ignition that may be caused by human error.

[0034] Example 1: In a large-scale cross-river fireworks display project, the specific deployment and operation are as follows. This project requires the deployment of over 10,000 distributed ignition modules along a stretch of over 5 kilometers of riverbanks, including multiple bridges, the rooftops of several high-rise buildings, and floating platforms on the river. The coordinated action time error of all modules must be less than 100 microseconds. Given the physical span of the site environment, the physical transmission delay of signals along the power cables varies significantly. Furthermore, the multi-day deployment involves multiple construction teams, increasing the possibility of human error in wiring. The 25-minute display process also places demands on the long-term stability of the local timers of the distributed ignition modules. After system deployment, before formally entering the execution phase, the central controller first initiates a deployment self-check phase. It broadcasts instructions to all distributed ignition modules via power line carrier communication, requiring them to identify the physical type of their connected fireworks products. The distributed ignition module with address 7852, located on a floating platform on the river, receives the instructions... After the command is given, a detection current is applied to a pyrotechnic product connected to it, and the corresponding identification code resistor value is measured to be 15.0kΩ through its analog-to-digital converter. The module then reports the data (address: 7852, resistance: 15.0kΩ) to the central controller. The central controller compares it with the internally stored firing script and finds that the script requires that the address should be connected to a pyrotechnic product with a resistance of 8.2kΩ. Based on this, the central controller determines that there is an error in the physical deployment and immediately outputs an error report containing the address, the product type that should be connected, and the actual product type connected to the field engineer on the operation interface. At the same time, the system is locked, preventing it from entering the next working state until the engineer verifies and corrects the wiring error on-site. After a new round of self-test confirms that the physical deployment of all ignition points is completely consistent with the digital script, the deployment self-test phase is completed. In this procedure, the power line carrier communication channel carries the reporting of physical type information, so that the preset of time decision and the verification of physical deployment are completed under the same hardware architecture.

[0035] At the preset start time of the ignition script, the central controller, through the control switching circuit, applies a voltage dip from 24V to 18V for 10 microseconds on a pair of shared DC power lines running through all deployment areas as a synchronization event. A distributed ignition module installed at the end of a bridge 5 kilometers away and another located near the control center, despite a delay of tens of microseconds in the physical moment they receive the voltage dip due to the physical distance difference, both trigger a hardware interrupt via their internal voltage comparators the instant they detect the event, immediately resetting their local timers and beginning an autonomous countdown based on pre-stored delay timestamps. Thus, the physical transmission delay, a factor affecting synchronization accuracy in traditional control methods, is not measured or compensated for in this architecture but is avoided by the procedure itself, preventing it from affecting the final coordinated accuracy. During the 25-minute autonomous countdown, the central controller... The system continuously generates metronome pulses at a frequency of 10.000Hz on the same DC power supply line. While each distributed ignition module performs its own countdown, it continuously counts and compares the count with the internally generated beat count value. Based on the published correction formula, it dynamically adjusts the timing step size of its local timer, thereby locking the timing rate of all modules to the high-stability clock reference of the central controller. In the final stage of the display, a design script requires two ignition positions, 5 kilometers apart, to be triggered simultaneously. Data collected by high-speed optical sensors deployed on-site shows that the actual ignition time difference between the two positions is less than 15 microseconds. Throughout the 25-minute display, no synchronization accuracy degradation caused by accumulated clock drift was observed in any of the distributed ignition modules. The data collected on-site indicates that the technical approach of combining non-timed timestamp presets with asynchronous synchronous triggering can achieve high-precision collaborative operation in a distributed control system with a large physical span.

[0036] Example 2: To objectively verify the initial synchronization accuracy of the cooperative ignition system of the present invention under long-distance, multi-node conditions, and its suppression effect on the frequency drift of the local timers of each distributed ignition module during long-term operation, the following test platform was built. This platform consists of a central controller and 100 distributed ignition modules. All modules are connected in series in a daisy-chain configuration via a 1-kilometer-long dual-core DC power supply line to simulate the physical distance and signal propagation delay in field deployment. To assess the system's stability under different temperature environments, the entire test platform was placed in a programmable environmental temperature chamber, with a temperature that can be maintained at 10°C. Up to 40 The data acquisition section uses a multi-channel digital oscilloscope with a time resolution better than 1 nanosecond to monitor the ignition output signals of three distributed ignition modules located at the beginning (address 001), middle (address 050), and end (address 100) of the line, respectively, to record their precise action moments. The experiment was conducted in three sample groups to isolate different influencing factors. Control group A used a traditional centralized synchronous triggering method, where the central controller broadcasts an ignition data packet containing address and instruction information at zero time. Test group B used the technical solution of this invention, which first distributes delayed timestamps in a non-timing manner, and then the central controller broadcasts a synchronization event to start the local countdown. However, in this sample group's experiment, the system metronome pulse function was disabled. Test group C used the complete technical solution of this invention. Based on test group B, it enabled the system metronome pulse continuously generated at a frequency of 10.000Hz by the central controller after broadcasting the synchronization event, as well as the dynamic correction function of the timing step size inside the distributed ignition module, with a correction gain coefficient of [missing information]. Set to 0.1.

[0037] In the experiment, the test results of control group A showed that the actual action time of the three monitored modules had a significant delay related to their physical position on the cable. Specifically, the end module with address 100 had a 21.5 microsecond delay compared to the start module with address 001. This time difference was mainly due to the signal propagation time on the 1-kilometer cable. Test group B, on the other hand, implemented delay timestamps. For a short-duration synchronous task of 1.000000 seconds, the ignition timing of the three monitored modules showed a high degree of consistency, with a maximum time difference of only 1.7 microseconds. This indicates that the asynchronous synchronous triggering mechanism effectively avoided the impact of physical transmission delay. However, when the delay timestamp is... When the duration was extended to 600.000000 seconds and a long-term synchronization task was performed under temperature variation conditions, the action times of the three modules in test group B showed significant dispersion, with the maximum time difference increasing to 511.6 microseconds. Modules with addresses 001 and 100 exhibited execution time deviations of +255.4 microseconds and -256.2 microseconds, respectively. This revealed the cumulative timing error caused by the inherent dispersion of the local crystal oscillators of each module under no external constraints and the influence of temperature. In contrast, test group C, when performing the exact same long-term synchronization task (… =600.000000 seconds (under the same temperature change conditions), the final synchronization accuracy was improved. The action times of the three monitored modules regained a high degree of consistency, and the maximum time difference between them was controlled at 4.3 microseconds, which is on the same order of magnitude as the result of the short-time synchronization task. The comparison results of the test data show that the introduction of the system metronome pulse and the local timing step dynamic correction function constitutes an effective closed-loop feedback calibration loop. This loop can continuously correct the local clock of each distributed module, so that its timing rate is consistent with the high stability clock reference of the central controller, thereby maintaining the high-precision synchronization of the system throughout the entire execution cycle.

[0038] Example 3: This example combines Figs. 1 to 3 This describes a multi-point high-precision coordinated ignition system for fireworks displays, such as... Fig. 1 As shown, the fireworks display script serves as the top-level input of the system, defining the execution time of each ignition module for the core processing unit. Based on this, the core processing unit generates control commands and a global clock reference. The commands are sent to the shared DC power supply line through two paths. One path modulates the data into a high-frequency signal via a power line carrier modulator, while the other path applies transient events to the voltage baseband via a high-power switching circuit. On the distributed terminal side, the deployment self-test module reports physical type information by measuring the resistance of the fireworks identification code during the deployment self-test phase. This information is compared with the database by the deployment correctness verification module to confirm the correctness of the physical deployment. During the execution phase, the power line carrier demodulation module is responsible for decoding and storing the delay timestamp, while the synchronization event monitoring module is responsible for monitoring voltage dips and verifying their waveform health. The outputs of both are fed into the microcontroller and local timer to achieve reset and start the autonomous countdown. During the countdown, the timing step dynamic correction module compensates for crystal oscillator frequency drift by comparing internal and external beat count values, and finally achieves ignition execution at the preset time to drive the fireworks product.

[0039] like Fig. 2 As shown, it depicts a river surface 1 spanning 5 kilometers, with the central controller 3 located on the bank, while a large number of distributed ignition modules 2 are dispersed and deployed in a vast physical space including the left bank building complex 4, the right bank building complex 5, the left bridge 6, the right bridge 7, and the floating platform 8 on the river surface. The central controller 3 sends delay timestamps and synchronization signals to all distributed ignition modules 2 through physical links, demonstrating the phenomenon of inconsistent signal transmission delays faced by the system in a large physical span and multi-node deployment environment.

[0040] like Fig. 3As shown, the process begins with the central controller broadcasting a self-test command to all distributed ignition modules. Upon receiving the command, distributed ignition module 1 applies a probe current to its connected pyrotechnic product 1 and receives the returned voltage drop. It then calculates its identification code resistance value and reports the data containing its own address and the resistance value, for example, address: 001, resistance: 8.2kΩ, to the central controller. In parallel, distributed ignition module 2 performs the same operation on its connected pyrotechnic product 2 and reports its measurement results, for example, address: 002, resistance: 15.0kΩ. After receiving this reported data, the central controller compares it with the internally stored database. If the comparison result shows that the deployment is correct, for example, module 001, the system will be unlocked and enter the ready state. If the comparison result shows that the deployment is incorrect, for example, module 002, the system will output an error report to the operation interface and remain locked until the physical deployment is corrected.

[0041] Example 4: In a system with completed physical wiring, including a central controller and 100 distributed ignition modules connected by a 1-kilometer-long DC power supply line, to match the internal control parameters of the system with the characteristics of the physical network deployed on-site, a system-level parameter self-tuning and calibration procedure needs to be executed before loading the ignition script. This procedure aims to determine the discrimination threshold for the health self-checking gating mechanism of synchronous events and to determine the correction gain coefficient for the dynamic correction mechanism of the system metronome's time base. To determine the health parameter template for the synchronization event, the central controller is first placed in an offline calibration mode. In this mode, the central controller continuously broadcasts 100 voltage dips generated at the nominal voltage as reference samples to all distributed ignition modules. Each distributed ignition module in the network uses its internal analog-to-digital converter to sample each received voltage dip waveform and calculates its actual rise time and pulse amplitude. These 100 sets of measurement data are then transmitted back to the central controller via power line carrier communication. After collecting a total of 10,000 data points from all 100 modules, the central controller performs statistical analysis, calculating that the mean rise time of all samples is 25 nanoseconds with a standard deviation of 4 nanoseconds, and the mean pulse amplitude is 18.5V with a standard deviation of 0.2V. According to the preset calibration rules, the upper limit threshold for rise time is set to the mean plus six times the standard deviation, i.e. Nanoseconds, rounded to 50 nanoseconds, and the effective range of pulse amplitude is set to the mean plus or minus six standard deviations, i.e. The range is calculated to be 17.3V to 19.7V. Finally, the central controller solidifies and distributes this health parameter template, which is obtained by calibrating the field data, namely, the upper limit threshold of rise time = 50 nanoseconds and the pulse amplitude range = [17.3V, 19.7V], to all distributed ignition modules.

[0042] Immediately afterwards, the system entered the calibration gain coefficient. The calibration process involves the central controller instructing the 100th distributed ignition module at the end of the line to artificially introduce a +50ppm positive frequency offset into the quartz crystal oscillator frequency of its local timer via software, simulating clock drift. Subsequently, the central controller begins broadcasting the system metronome pulse at a frequency of 10.000Hz and initiates an iterative optimization program to determine... The value, the program starts from an initial value Initially, module 100 executes the timing step dynamic correction algorithm disclosed in the aforementioned specific embodiment and monitors its external beat count value. With internal beat count value The speed at which the difference converges to zero over time; after completing one test, the program will... The value was increased by 0.01 and the test was repeated until an overshoot was observed in the difference during convergence, i.e., the difference changed from positive to negative; experimental data showed that when As the value increases from 0.01 to 0.12, the convergence time continues to shorten. When the value was set to 0.13, overshoot was observed for the first time. Based on the optimization criteria set in the system design to balance response speed and stability, 90% of the critical overshoot value was taken as the operating point, and the system finally determined the correction gain coefficient for this deployment environment. for This coefficient value is distributed by the central controller and embedded into all distributed ignition modules. By executing the above-mentioned system-level self-tuning and calibration procedures, the two key control parameters inside the control system, namely the health parameter template of the synchronous event and the gain coefficient of the time base correction, have obtained set values ​​that match the characteristics of their physical network environment, thereby setting traceable engineering parameters for subsequent system operation.

[0043] Example 5: Before the collaborative ignition system of this invention is put into application, it is necessary to first complete the offline construction of a database mapping relationship between the physical type of fireworks products and their identification code resistors. This procedure is carried out in a controlled laboratory environment. Engineers take samples of each specification of fireworks products to be used in the pyrotechnic display and connect them one by one to a distributed ignition module that serves as a reference measurement device. The central controller instructs the module to perform 100 repeated measurements on the identification code resistor of each sample. Subsequently, the central controller calculates the arithmetic mean of these 100 measurements and uses this mean as the standard resistance value of this specification of fireworks product. Along with its product name and specification model, it is entered into the database. At the same time, the system will also set an allowable matching tolerance range for this standard resistance value. This range is determined according to the accuracy level of the resistor element itself and is usually set to ±5%. By repeating the above process for all types of fireworks products involved, a mapping relationship database containing all types of fireworks products and their corresponding standard resistance values ​​and tolerance ranges is finally generated. This database is loaded into the central controller during the deployment phase and serves as the basis for comparing the correctness of physical deployment when performing the deployment self-check function.

[0044] After all on-site deployment, connection, and parameter calibration are completed, the system will initiate a global physical link quality diagnostic procedure in the final stage before executing the ignition mission. The central controller first instructs all distributed ignition modules to enter a diagnostic mode. In this mode, the controller, in order from the far end to the near end, instructs each distributed ignition module to apply a transient load of 50 mA to the shared DC power supply line through a calibration load circuit within its internal circuit. During the load application by any module, all other distributed ignition modules in the network use their analog-to-digital converters to synchronously measure and record the power supply at their respective locations. The system detects the voltage drop amplitude of the power line and reports its own address and voltage drop amplitude data to the central controller. After collecting and analyzing the voltage drop data of all modules under different load points, the central controller calculates the equivalent impedance between any two adjacent modules in the network topology based on the electrical model of the line. It then compares this calculated value with a theoretical impedance value based on the cable specifications and laying length. If the difference between the two exceeds a preset fault judgment threshold, it indicates that there may be physical faults such as poor connector contact or cable damage in this section of the line. The system will report the specific location of the faulty section to the operation interface for technicians to investigate and maintain.

[0045] Example 6: In a scenario where a coordinated ignition system with all parameters calibrated is put into formal operation, in order to establish an initial health baseline for comparison during subsequent online status monitoring and fault diagnosis in the mission cycle, the system will automatically execute a zero-time health baseline establishment procedure after broadcasting a global synchronization event. The first step of this procedure aims to determine the set of effective distributed ignition modules that have entered the synchronization state. Specifically, at the initial moment, the central controller... Within one second of a voltage dip as a synchronization event broadcast via the DC power supply baseband, a status query command is immediately broadcast to all registered modules in a polling manner through the power line carrier channel on the same pair of power supply lines. Each distributed ignition module that has normally received the synchronization event and started its local timer is configured to immediately send back a response containing its own address and synchronization confirmation status upon receiving this query command. The central controller receives and verifies the response information of all modules within a preset 2-second time window. If no response is received from a registered module within this window, the module is determined to be disconnected or has failed to synchronize. The central controller then removes the module from the current ignition task sequence and uses the set of modules that have confirmed the response as the valid node baseline for this task.

[0046] After establishing the baseline of the effective nodes, the second step of the procedure aims to establish the physical link health baseline connecting these effective nodes. The central controller then initiates a dynamic network impedance measurement. This process reuses a distributed network diagnostic method, which sequentially instructs each module to apply a calibration load while the remaining modules synchronously measure voltage drops, thereby calculating the actual equivalent impedance between every two adjacent modules in the network. This entire set of line impedance values ​​measured and calculated at the start of the task is recorded and stored by the central controller as the zero-time physical link health baseline for this task. Based on a deterministic rule, the fault judgment threshold for subsequent online diagnostics is set to 150% of this baseline impedance value plus a fixed tolerance of 0.5Ω. During subsequent operation, the system can periodically repeat this network diagnostic process and compare the real-time measured line impedance with this zero-time health baseline to... The system identifies new physical link quality degradation during task execution; after the online status confirmation and baseline establishment procedures are completed and the system finishes all ignition tasks, the central controller enters a post-event log analysis phase. In this phase, the central controller organizes and analyzes the status reports sent back by all distributed ignition modules via the power line carrier channel after their preset delay timestamps arrived and ignition was initiated during the entire task execution. A typical status report log shows that most modules reported successful ignition, but module 8192 did not send back any information within the expected report time window, while module 4096 reported an open-circuit fault status code. Based on this log, the central controller automatically generates a maintenance work order containing the addresses of these two modules and their final status to guide technicians in targeted repairs.

[0047] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-point high-precision coordinated ignition system for fireworks displays, characterized in that, include: A central controller and multiple distributed ignition modules are connected via a pair of shared DC power lines. The central controller is configured to: generate a delay timestamp for each distributed ignition module according to the fireworks display script, modulate the delay timestamp using a power line carrier modulator, and distribute it non-timingly via the DC power lines; at the start of the fireworks display, apply a voltage transient without address or instruction information to the voltage baseband of the DC power lines by controlling a switching circuit as a synchronization event; after the synchronization event, continuously generate system metronome pulses at a uniform frequency on the same DC power lines via the switching circuit. Each distributed ignition module includes a power line carrier demodulation module, a voltage comparator, a local timer, and a microcontroller. The distributed ignition module is configured to: obtain power via a DC power supply line, decode and store a delay timestamp through its power line carrier demodulation module; monitor synchronization events through its voltage comparator, and when the synchronization event is detected, the microcontroller resets the local timer to zero and performs an autonomous countdown based on the stored delay timestamp to drive ignition. During the autonomous countdown, the received system metronome pulses are counted in parallel to obtain the external beat count value, and the timing step size of the local timer is periodically adjusted based on the difference between the external beat count value and the internal beat count value generated by the local timer. Furthermore, the microcontroller is configured to perform timing step size adjustment within each calibration cycle, specifically by calculating a calibrated timing step size. For use in the next timing cycle of the local timer, ,in, It is the nominal timing step size of the local timer before calibration. It is the cumulative external beat count value within this correction period. It is the internal tick count value generated by the local timer within the same correction cycle. It is a correction gain coefficient stored within the distributed ignition module.

2. The multi-point high-precision coordinated ignition system for fireworks display according to claim 1, characterized in that, The central controller is configured to modulate the delayed timestamp into a high-frequency data signal and superimpose it onto the DC power supply line; the power line carrier demodulation module of the distributed ignition module is configured to filter out the baseband voltage of the DC power supply line and demodulate the high-frequency data signal to obtain the delayed timestamp. The central controller is configured to generate a voltage dip with a defined amplitude and duration as a synchronization event; the voltage comparator of the distributed ignition module is configured to generate a hardware interrupt signal when a voltage dip matching the defined amplitude and duration is detected, and the microcontroller responds to the hardware interrupt signal to perform a reset of the local timer.

3. The multi-point high-precision coordinated ignition system for fireworks display according to claim 1, characterized in that, The distributed ignition module also includes an analog-to-digital converter (ADC). The distributed ignition module is configured such that, when the voltage comparator detects a voltage change, the microcontroller first instructs the ADC to sample the waveform of the voltage change to obtain its rise time and pulse amplitude. The microcontroller is also configured to compare the acquired rise time and pulse amplitude with a health parameter template stored internally, containing a rise time upper limit threshold and a pulse amplitude range. Furthermore, the microcontroller will only perform a local timer reset operation if the acquired rise time is less than the rise time upper limit threshold and the pulse amplitude falls within the pulse amplitude range.

4. The multi-point high-precision coordinated ignition system for fireworks display according to claim 1, characterized in that, The distributed ignition module is also configured to, during a deployment self-test phase, apply a probe current to the pyrotechnic product connected to it via its microcontroller, which has an identification-coded resistor, and instruct its analog-to-digital converter to measure the resulting voltage drop in order to calculate the resistance value of the identification-coded resistor. The resistance value is then reported to the central controller as physical type information. The central controller is also configured to compare the received resistance value with a database stored within it that records the mapping relationship between fireworks product types and resistance values, in order to confirm the correctness of the physical deployment before ignition is executed.

5. A multi-point high-precision coordinated ignition system for fireworks display according to claim 2, characterized in that, The central controller is configured to generate voltage dips ranging from 20% to 40% of the system's nominal supply voltage, with a duration ranging from 5 to 20 microseconds.

6. A multi-point high-precision coordinated ignition system for fireworks display according to claim 4, characterized in that, The central controller is also configured to output an error report to the operating interface when an error is confirmed in the physical deployment. The error report includes the address of the faulty distributed ignition module, as well as the type of pyrotechnic product that should be connected to that address and the type of pyrotechnic product that was actually connected, as resolved by the mapping relationship.

7. A multi-point high-precision coordinated ignition system for fireworks display according to claim 1, characterized in that, The central controller is also configured to: after completing the distribution of delayed timestamps, broadcast polling instructions and enter a waiting state until it receives confirmation information from all distributed ignition modules that the delayed timestamps have been successfully received and stored, before it enters the working state of broadcasting synchronization events.

8. A multi-point high-precision coordinated ignition system for fireworks display according to claim 1, characterized in that, The distributed ignition module is also configured to send a status report containing its own address and ignition execution status to the central controller via a power line carrier modulator after it drives ignition. The central controller is also configured to monitor and record the status report information of all distributed ignition modules within a time window after the broadcast synchronization event, and to list the addresses of distributed ignition modules that have not received successful report information.

Images