Remote control torch ignition system based on wireless command and fault self-diagnosis function

Abstract

The application discloses a remote control torch ignition system based on wireless instruction and fault self-diagnosis function and relates to the technical field of space propulsion. The application comprises a remote control center and a field control unit. The remote control center is used for generating an encrypted ignition instruction package containing an ignition instruction sequence and adjustable parameters and sending the package through a wireless data link, simultaneously receiving and displaying control feedback and diagnosis data from the field. The field control unit is arranged on an engine test stand and is electrically connected with a torch igniter and its matched fuel valve, oxidant valve and ignition power supply, used for receiving, decrypting and executing the encrypted ignition instruction package to control the corresponding valves and ignition power supply to act in sequence. The system further comprises a multi-dimensional data acquisition module used for synchronously acquiring multi-channel physical quantity time sequence signals reflecting the working state of the igniter during the ignition process. Through embedded multi-dimensional time sequence fusion diagnosis, the application can quantitatively evaluate the performance gradual state of the torch igniter in real time.

Description

Technical Field

[0001] This invention belongs to the field of aerospace propulsion technology, and in particular relates to a remote control torch ignition system based on wireless commands and fault self-diagnosis function. Background Technology

[0002] In ground testing of liquid rocket engines, especially in scenarios involving multiple repeated ignition tests on the same engine, the reliability of the flare igniter is fundamental to ensuring the success and safety of the test. With the accumulation of ignition cycles, gradual physical changes inevitably occur inside the igniter, such as carbon buildup in the injection orifice, electrode erosion, and slight drift in the fuel-oxidizer mixture ratio. These changes are not sudden failures, but rather a slow, "sub-healthy" process of performance degradation.

[0003] Existing remote ignition control and diagnostic technologies mostly rely on setting fixed thresholds for key parameters (such as ignition voltage and final pressure) to trigger over-limit alarms, or checking whether the action logic sequence at preset time points has been completed. These methods are effective for determining a binary conclusion of "success" or "failure" in ignition, but they struggle to quantify and detect the slow-growing performance degradation. As a result, the system cannot distinguish between a "barely successful" but degraded ignition and a "perfectly healthy" ignition. Maintenance decisions often rely on a fixed number of test cycles or post-event physical inspection, lacking a refined management basis based on real-time performance degradation data. Therefore, to address these issues, the following solution is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a remote-controlled torch ignition system based on wireless commands and fault self-diagnosis. Through embedded multi-dimensional time-series fusion diagnosis, it can quantitatively evaluate the performance gradient of the torch igniter in real time, solving the problem that existing methods lack online perception of potential performance degradation.

[0005] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: This invention is a remote control torch ignition system based on wireless command and fault self-diagnosis functions, including a remote control center and a field control unit; The remote control center is used to generate an encrypted ignition command package containing an ignition command sequence and adjustable parameters and send it through the wireless data link, while receiving and displaying control feedback and diagnostic data from the field. The field control unit is deployed on the engine test bench and is electrically connected to the torch igniter and its associated fuel valve, oxidizer valve and ignition power supply. It is used to receive, decrypt and execute the encrypted ignition command packet and control the corresponding valves and ignition power supply to operate in sequence. The system also includes a multi-dimensional data acquisition module, which is used to synchronously acquire multi-channel physical quantity timing signals reflecting the working status of the igniter during the ignition process; The field control unit includes a real-time fault self-diagnosis module, which is configured to generate a comprehensive index characterizing the current health of the torch igniter based on the time-series signals synchronously acquired by the multi-dimensional data acquisition module through time alignment, feature extraction and adaptive weighted fusion calculation, and complete status diagnosis and decision feedback based on the index.

[0006] Furthermore, the remote control center includes: Command generation module: Generates encrypted ignition commands containing adjustable parameters at the remote control center; User interface display module: Provides the operator with an ignition control interface and displays the returned diagnostic data and system status; The first wireless communication module is responsible for sending encrypted commands from the remote control center to the field and receiving data from the field. The field control unit also includes: Data storage module: Locally stores diagnostic results to provide a basis for adaptive parameter adjustment for the next ignition command; The second wireless communication module: The field side is responsible for receiving remote commands and uploading locally collected diagnostic data; Command receiving and decryption module: Receives wireless commands on-site, performs decryption and verification to ensure command validity; Timing execution control module: Strictly controls valve opening and closing and ignition of the igniter according to the time nodes of the instruction sequence.

[0007] Furthermore, the real-time fault self-diagnosis module includes: Timing alignment submodule: Uses dynamic time warping algorithm to eliminate time axis jitter between the acquired signals of each channel and the reference template; Feature extraction submodule: Extracts feature vectors that characterize the degree of deviation from the benchmark from each aligned time-series signal; Adaptive weighted fusion submodule: Adaptively allocates weights based on the information entropy of each feature vector and calculates the comprehensive health index by fusion; Diagnostic Decision Submodule: Determines the status level based on the health index and generates control feedback or early warning decisions.

[0008] Furthermore, the multidimensional data acquisition module synchronously acquires timing signals from at least the following four channels: the torch igniter operating voltage signal, the ion flow signal reflecting the degree of flame ionization, the spectral intensity signal reflecting specific combustion chemical components, and the combustion chamber pressure signal.

[0009] Furthermore, the specific method for the real-time fault self-diagnosis module to perform timing alignment is as follows: a dynamic time warping algorithm is used to nonlinearly align the timing data collected in this session of each channel with its pre-stored reference timing template to eliminate the influence of minute timing jitter during the ignition process.

[0010] Furthermore, the specific method for the real-time fault self-diagnosis module to perform feature extraction is as follows: for the time-series data of each channel after alignment, calculate the deviation of its deviation from the corresponding reference data in multiple consecutive sub-time periods within the entire ignition observation time window, and form the feature vector of that channel.

[0011] Furthermore, the specific method by which the real-time fault self-diagnosis module performs adaptive weighted fusion calculation is as follows: First, calculate the information entropy of each channel's feature vector based on the uniformity of the distribution of deviation in each channel's feature vector, and then assign adaptive fusion weights to each channel according to the reciprocal relationship of the information entropy, with channels that have a more concentrated distribution of deviation being given higher weights. By combining the overall deviation of each channel feature vector from the ideal benchmark, the adaptive fusion weights, and the preset deviation normalization benchmark, a comprehensive health index with a value between zero and one is calculated. The closer the index is to one, the higher the health level.

[0012] Furthermore, the decision feedback executed by the real-time fault self-diagnosis module includes: uploading the comprehensive health index and diagnostic judgment results to the remote control center via a wireless data link; and storing the calculated comprehensive health index value locally as the basis for adjusting the ignition energy parameters when the next ignition command is generated.

[0013] A remote-controlled torch ignition method based on wireless command and fault self-diagnosis functions, the method comprising the following steps: Step S1: The remote control center generates and sends an encrypted ignition command packet; Step S2: The field control unit receives, decrypts, and executes the instruction packet, and controls the valves to open and the igniter to be powered on in sequence; Step S3: During the ignition process, synchronously acquire multi-dimensional physical quantity timing signals; Step S4: Based on the collected time-series signals, perform real-time fault self-diagnosis, calculate the comprehensive health index, and complete the status diagnosis; Step S5: Based on the diagnostic results, perform decision feedback operations, including data feedback, parameter adjustment, or safety interlock.

[0014] Furthermore, the specific steps for performing real-time fault self-diagnosis in step S4 include: Step S41: Align the timing signals acquired from each channel with their reference timing template; Step S42: Extract feature vectors from the aligned data of each channel; Step S43: Calculate the adaptive fusion weights based on the distribution characteristics of the deviation of the feature vectors of each channel; Step S44: Combine the normalized bias of the feature vectors of each channel with the adaptive weights to calculate the comprehensive health index; Step S45: Compare the comprehensive health index with a preset threshold to determine whether the current status of the igniter is healthy, under warning, or faulty.

[0015] The present invention has the following beneficial effects: 1. This invention improves the finesse and continuity of the perception of the performance status of the flare igniter by real-time fusion of multi-dimensional time-series data for online health assessment. Traditional diagnostic methods often rely on isolated parameter thresholds or fixed time-series point judgments, making it difficult to capture the gradual performance degradation that occurs with multiple ignitions. This method, by simultaneously analyzing the changes of multiple physical quantities such as voltage, ion current, spectrum, and pressure curves over time and calculating a continuous health index, can effectively identify sub-healthy states between complete normality and complete failure. This capability enables the system not only to report whether a failure has occurred but also to quantify the performance degradation trend, providing a richer dimension for status assessment.

[0016] 2. This invention enables autonomous closed-loop fine-tuning of control parameters based on evaluation results, enhancing the system's adaptability and the integrity of the control loop. The system uses the calculated health index as feedback and dynamically adjusts key parameters in subsequent ignition commands, such as the duration of ignition energy supply, through a preset strategy mapping relationship. This design allows the control process to respond to changes in the torch igniter's own state and compensate for slow performance degradation, thereby helping to maintain the reliability of the ignition process and potentially supporting the extension of the effective service life of components within permissible limits.

[0017] 3. This invention employs an adaptive weighting strategy based on information entropy for feature fusion, enhancing the relevance and robustness of diagnostic conclusions. The value and certainty of information reflected by each monitoring channel will differ depending on the fault mode or performance degradation type. This method adaptively assigns appropriate weights to different channels by analyzing the distribution concentration of deviations within each feature vector, enabling the comprehensive health index to more sensitively focus on key features exhibiting abnormalities during the current ignition process. This data-driven weight allocation method reduces the limitations of fixed-weight strategies, making the diagnostic logic more closely aligned with actual data performance in specific scenarios.

[0018] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the structure of the remote control torch ignition system based on wireless command and fault self-diagnosis function of the present invention. Figure 2 This is a flowchart illustrating the remote control torch ignition method based on wireless commands and fault self-diagnosis function according to the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described 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.

[0022] Please see Figure 1 As shown, the present invention is a remote control torch ignition system based on wireless command and fault self-diagnosis function, including a remote control center and a field control unit; The remote control center is used to generate encrypted ignition command packets containing ignition command sequences and adjustable parameters and send them via a wireless data link, while simultaneously receiving and displaying control feedback and diagnostic data from the field. The field control unit is deployed on the engine test bench and is electrically connected to the torch igniter and its associated fuel valve, oxidizer valve and ignition power supply. It is used to receive, decrypt and execute encrypted ignition command packets and control the corresponding valves and ignition power supply to operate in sequence. The system also includes a multi-dimensional data acquisition module, which is used to synchronously acquire multi-channel physical quantity timing signals that reflect the working status of the igniter during the ignition process; The field control unit includes a real-time fault self-diagnosis module, which is configured to generate a comprehensive index characterizing the current health of the flare igniter based on the time-series signals synchronously acquired by the multi-dimensional data acquisition module through time alignment, feature extraction and adaptive weighted fusion calculation, and complete status diagnosis and decision feedback based on the index.

[0023] The remote control center includes: Command generation module: Generates encrypted ignition commands containing adjustable parameters at the remote control center; User interface display module: Provides the operator with an ignition control interface and displays the returned diagnostic data and system status; The first wireless communication module is responsible for sending encrypted commands from the remote control center to the field and receiving data from the field. The field control unit also includes: Data storage module: Locally stores diagnostic results to provide a basis for adaptive parameter adjustment for the next ignition command; The second wireless communication module: The field side is responsible for receiving remote commands and uploading locally collected diagnostic data; Command receiving and decryption module: Receives wireless commands on-site, performs decryption and verification to ensure command validity; Timing execution control module: Strictly controls valve opening and closing and ignition of the igniter according to the time nodes of the instruction sequence.

[0024] The real-time fault self-diagnosis module includes: Timing alignment submodule: Uses dynamic time warping algorithm to eliminate time axis jitter between the acquired signals of each channel and the reference template; Feature extraction submodule: Extracts feature vectors that characterize the degree of deviation from the benchmark from each aligned time-series signal; Adaptive weighted fusion submodule: Adaptively allocates weights based on the information entropy of each feature vector and calculates the comprehensive health index by fusion; Diagnostic Decision Submodule: Determines the status level based on the health index and generates control feedback or early warning decisions.

[0025] The multidimensional data acquisition module shall simultaneously acquire timing signals from at least the following four channels: flare igniter operating voltage signal, ion flow signal reflecting the degree of flame ionization, spectral intensity signal reflecting specific combustion chemical components, and combustion chamber pressure signal.

[0026] The real-time fault self-diagnosis module performs timing alignment in the following way: it adopts a dynamic time warping algorithm to align the timing data collected in this session of each channel with its pre-stored reference timing template on a non-linear time axis, so as to eliminate the influence of tiny timing jitter in the ignition process.

[0027] The real-time fault self-diagnosis module performs feature extraction in the following way: for the time series data of each channel after alignment, calculate the deviation of its deviation from the corresponding reference data in multiple consecutive sub-time periods within the entire ignition observation time window, and form the feature vector of that channel.

[0028] The specific method by which the real-time fault self-diagnosis module performs adaptive weighted fusion calculation is as follows: First, calculate the information entropy of each channel's feature vector based on the uniformity of the distribution of deviation in each channel's feature vector, and then assign adaptive fusion weights to each channel according to the reciprocal relationship of the information entropy, with channels that have a more concentrated distribution of deviation being given higher weights. By combining the overall deviation of each channel feature vector from the ideal benchmark, the adaptive fusion weights, and the preset deviation normalization benchmark, a comprehensive health index with a value between zero and one is calculated. The closer the index is to one, the higher the health level.

[0029] The decision feedback executed by the real-time fault self-diagnosis module includes: uploading the comprehensive health index and diagnostic judgment results to the remote control center via a wireless data link; and storing the calculated comprehensive health index value locally as the basis for adjusting the ignition energy parameters when the next ignition command is generated.

[0030] Please see Figure 2 As shown, this invention is a remote-controlled torch ignition method based on wireless commands and fault self-diagnosis function, comprising the following steps: Step S1: The remote control center generates and sends an encrypted ignition command packet; Step S2: The field control unit receives, decrypts, and executes the instruction packet, and controls the valves to open and the igniter to be powered on in sequence; Step S3: During the ignition process, synchronously acquire multi-dimensional physical quantity timing signals; Step S4: Based on the collected time-series signals, perform real-time fault self-diagnosis, calculate the comprehensive health index, and complete the status diagnosis; Step S5: Based on the diagnostic results, perform decision feedback operations, including data feedback, parameter adjustment, or safety interlock.

[0031] The specific steps for performing real-time fault self-diagnosis in step S4 include: Step S41: Align the timing signals acquired from each channel with their reference timing template; Step S42: Extract feature vectors from the aligned data of each channel; Step S43: Calculate the adaptive fusion weights based on the distribution characteristics of the deviation of the feature vectors of each channel; Step S44: Combine the normalized bias and adaptive weights of the feature vectors of each channel to calculate the comprehensive health index; Step S45: Compare the overall health index with the preset threshold to determine whether the igniter is currently in good condition, under warning, or in fault condition.

[0032] One specific application of this embodiment is: System initialization and parameter preset The main control computer at the remote control center establishes an encrypted wireless data link with the field control unit installed on the engine test stand (e.g., using a dedicated wireless data transmission radio with anti-interference capabilities). The main control computer loads a reference parameter package for the specific engine model and flare igniter, including: Reference ignition command sequence time nodes: (Instruction issued) (Fuel valve open) (Oxidant valve open) (Igniter is powered on) (Theoretical ignition timing) (Stable combustion detection point); Reference timing data templates for four monitoring channels: reference voltage curves Reference ion current curve Spectral intensity curves at a specific reference wavelength (e.g., OH* radical spectrum) Reference combustion chamber pressure rise curve The time windows are all ,in This is the pre-trigger acquisition time offset.

[0033] Health Index Calculation parameters: initial weights of each feature vector (satisfy ), and health threshold , , .

[0034] Control strategy mapping table: based on The value maps to the parameter adjustment amount for the next ignition command (such as the igniter power-on duration). (Compensation value).

[0035] Step S1: Ignition command generation, encryption, and wireless transmission The operator triggers the start command for this ignition from the remote control center interface. The main control computer operates according to the following procedure: Query the value calculated in the previous loop (initial value for the first ignition). And based on the control strategy mapping table, determine the parameters of this ignition command. For example, if the last... If the circuit is in the warning zone, the duration of ignition power-on will be [duration to be specified]. ,in This is the positive compensation amount derived from an empirical formula.

[0036] Generate a sequence of valve action timestamps, encrypted verification codes, and valve action sequence timestamps. , (,...) and the minor adjustments made this time Encrypted ignition command packets containing information such as...

[0037] The encrypted command packet is sent to the test bench control unit via the wireless communication module.

[0038] Step S2: Receiving, decrypting, and executing on-site instructions Field control unit: Receives and decrypts wireless command packets, verifies the validity of the checksum and timestamp. (Absolute time) (Ensuring high-precision synchronous clock operation), execution begins strictly according to the instruction sequence: at The fuel control valve should be kept open at all times. The oxidant control valve should be kept open at all times. At a certain time, electrical energy is applied to the torch igniter for a duration of [duration missing]. The ignition process is initiated.

[0039] Step S3: Synchronous Acquisition and Preprocessing of Multi-Dimensional Time Series Data Simultaneously with the ignition command execution, the on-site data acquisition system synchronously triggers and acquires timing data from the following four channels at a high sampling rate (e.g., 10kHz), with a time window of [time window value missing]. : Channel A (Voltage): Monitors the operating voltage of the igniter. .

[0040] Channel B (Ion Flow): Monitors the ion flow signal generated by combustion using a probe installed near the flame. .

[0041] Channel C (Spectrum): Monitors the spectral intensity of a specific wavelength (e.g., OH*308nm) in the torch combustion region using a fiber optic spectrometer. .

[0042] Channel D (Pressure): Monitors combustion chamber pressure .

[0043] Raw data collected Immediately perform local preprocessing, including low-pass filtering for noise reduction, power frequency interference elimination, and normalization to the same dimensional range as the reference template.

[0044] Step S4: Temporal alignment and feature vector extraction based on Dynamic Time Warping (DTW) Preprocessed data It is sent to the diagnostic algorithm module built into the field control unit.

[0045] Step S41, Timing Alignment: Due to the slight time jitter in the actual physical process of each ignition (such as differences in mixing rate), direct point-to-point comparison is inaccurate. The algorithm uses dynamic time warping technology to align the timing of each ignition. With the corresponding benchmark template Alignment is performed to find the optimal curved path, thereby eliminating nonlinear drift on the time axis and obtaining aligned time series data. .

[0046] Step S42, Feature Extraction: From each aligned time-series data point, extract a feature vector (length [length missing]) that can characterize the ignition quality of this test. Taking voltage channels as an example, the feature vector Each element Calculated in the following way: In the formula, The first characteristic vector of the voltage channel One element; This is the index of the feature vector element, with a value of ,in It is the pre-defined feature vector length (i.e., the number of time segments). For the first time intervals The number of sampling points within; For discrete time points, within the sampling interval of the data acquisition system, the time window... Discretized into a series of time points, Represent one of them; For the first This is a time interval, which is the entire analysis time window. Average score The first continuous subinterval obtained after the nth... Sub-intervals; For at a certain point in time The voltage measurement value after dynamic time-normalized alignment at that location; For at a certain point in time The corresponding value of the reference voltage template at that location.

[0047] Similarly, the ion current characteristic vector is obtained. Spectral eigenvectors Pressure eigenvector .

[0048] Step S43, Adaptive weight calculation: It is the first The adaptive weights of each channel are calculated as follows:

[0049]

[0050] In the formula, For the first Adaptive weighting of each channel in the health index calculation; For the first The reciprocal of the information entropy of each channel; For the summation index, representing the channel, iterate through the set. ; For the first In the feature vector of the i-th channel, the i-th Normalized absolute value weights (probabilities) of each element; For channel indexing, These represent four monitoring channels: voltage, ion current, spectrum, and pressure, respectively. For the first The th channel feature vector of the th channel One element; The length of the feature vector is a pre-defined positive integer representing the number of segments into which the time window is divided, and it also determines the length of the feature vector. The dimension; This weighting strategy is based on information entropy; if the feature vector of a channel... If the deviation distribution is very "concentrated" (low entropy value) across different time periods, it indicates that the channel exhibited a significant anomaly at a specific stage during this ignition, and should be given a higher weight, because such concentrated deviation is likely to correspond to a specific fault mode (e.g., voltage anomalies in the early stages of ignition may correspond to insulation problems). Conversely, if the deviation distribution is very "dispersed" (high entropy value), it may be random noise or a global weak degradation, and its weight should be reduced. This strategy enables the fusion process to adaptively focus on the feature channels that best reveal the problem during this ignition. Step S44, Adaptive weighted fusion of multidimensional feature vectors and calculation of health index: Calculate the comprehensive health index. The formula is as follows: In the formula, A comprehensive health index; The ideal benchmark for complete health; For the first The feature vector of each channel is a 3D column vector, i.e. ; for Zero-dimensional vector, i.e. This represents the ideal state that perfectly matches the benchmark template; For vectors of Norm (Euclidean norm); For the first The feature vector of each channel under the known worst-case fault condition of Norm. This is a scalar constant used for normalization, its value being calculated from the feature vectors of the most severe failure cases identified in historical data. The norm is obtained, which represents the scale of the maximum possible deviation of the channel from the baseline.

[0051] Step S45: Based on the calculated The system automatically performs diagnosis and decision-making: Diagnostic determination: If If it is judged as "healthy"; if If it is determined to be "sub-health (warning)"; if It was determined to be a "malfunction".

[0052] Step S5, Decision and Feedback: Regardless of the state, all raw data and feature vectors from this ignition process The values ​​and judgment results are transmitted back to the remote control center for recording and display via wireless link.

[0053] At the same time, the field control unit will calculate the results of this operation. The value is temporarily stored in local non-volatile memory for adjustment when the next ignition command is generated. The basis for this is as follows. For example, if the condition is "sub-healthy", the ignition energy will be increased next time; if the condition is "healthy", the baseline parameters will be used.

[0054] If a "fault" is detected, the field control unit will immediately and automatically block subsequent continuous ignition commands (if specified in the test plan) while transmitting data back, and issue the highest level audible and visual alarm, prompting that a physical inspection must be performed immediately.

[0055] Loop iteration: For continuous repeated ignition tests, the system repeatedly executes steps S1 to S5. Each ignition is both an execution process and a diagnostic process. Historical trend charts (displayed in a remote center) can clearly show the performance degradation trajectory of the torch igniter, enabling precise "on-demand maintenance" and intervention by adjusting parameters or issuing warnings before potential failures lead to serious consequences.

[0056] Through the above steps, this invention deeply couples remote ignition control with embedded advanced diagnostics, solving the problem of difficulty in early detection of gradual faults in specific scenarios, and improving the safety, intelligence and economy of testing.

[0057] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0058] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A remote-controlled torch ignition system based on wireless command and fault self-diagnosis functions, characterized in that, The system includes a remote control center and a field control unit; The remote control center is used to generate an encrypted ignition command package containing an ignition command sequence and adjustable parameters and send it through the wireless data link, while receiving and displaying control feedback and diagnostic data from the field. The field control unit is deployed on the engine test bench and is electrically connected to the torch igniter and its associated fuel valve, oxidizer valve and ignition power supply. It is used to receive, decrypt and execute the encrypted ignition command packet and control the corresponding valves and ignition power supply to operate in sequence. The system also includes a multi-dimensional data acquisition module, which is used to synchronously acquire multi-channel physical quantity timing signals reflecting the working status of the igniter during the ignition process; The field control unit includes a real-time fault self-diagnosis module, which is configured to generate a comprehensive index characterizing the current health of the torch igniter based on the time-series signals synchronously acquired by the multi-dimensional data acquisition module through time alignment, feature extraction and adaptive weighted fusion calculation, and complete status diagnosis and decision feedback based on the index.

2. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 1, characterized in that, The remote control center includes: Command generation module: Generates encrypted ignition commands containing adjustable parameters at the remote control center; User interface display module: Provides the operator with an ignition control interface and displays the returned diagnostic data and system status; The first wireless communication module is responsible for sending encrypted commands from the remote control center to the field and receiving data from the field. The field control unit also includes: Data storage module: Locally stores diagnostic results to provide a basis for adaptive parameter adjustment for the next ignition command; The second wireless communication module: The field side is responsible for receiving remote commands and uploading locally collected diagnostic data; Command receiving and decryption module: Receives wireless commands on-site, performs decryption and verification to ensure command validity; Timing execution control module: Strictly controls valve opening and closing and ignition of the igniter according to the time nodes of the instruction sequence.

3. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 1, characterized in that, The real-time fault self-diagnosis module includes: Timing alignment submodule: Uses dynamic time warping algorithm to eliminate time axis jitter between the acquired signals of each channel and the reference template; Feature extraction submodule: Extracts feature vectors that characterize the degree of deviation from the benchmark from each aligned time-series signal; Adaptive weighted fusion submodule: Adaptively allocates weights based on the information entropy of each feature vector and calculates the comprehensive health index by fusion; Diagnostic Decision Submodule: Determines the status level based on the health index and generates control feedback or early warning decisions.

4. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 1, characterized in that, The multidimensional data acquisition module synchronously acquires timing signals from at least the following four channels: flare igniter operating voltage signal, ion flow signal reflecting the degree of flame ionization, spectral intensity signal reflecting specific combustion chemical components, and combustion chamber pressure signal.

5. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 1, characterized in that, The specific method for the real-time fault self-diagnosis module to perform timing alignment is as follows: a dynamic time warping algorithm is used to nonlinearly align the timing data collected in this session of each channel with its pre-stored reference timing template to eliminate the influence of minute timing jitter during the ignition process.

6. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 5, characterized in that, The specific method for the real-time fault self-diagnosis module to perform feature extraction is as follows: for the time series data of each channel after alignment, calculate the deviation of its deviation from the corresponding reference data in multiple consecutive sub-time periods within the entire ignition observation time window, and form the feature vector of that channel.

7. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to claim 6, characterized in that, The specific method by which the real-time fault self-diagnosis module performs adaptive weighted fusion calculation is as follows: First, calculate the information entropy of each channel's feature vector based on the uniformity of the distribution of deviation in each channel's feature vector, and then assign adaptive fusion weights to each channel according to the reciprocal relationship of the information entropy, with channels that have a more concentrated distribution of deviation being given higher weights. By combining the overall deviation of each channel feature vector from the ideal benchmark, the adaptive fusion weights, and the preset deviation normalization benchmark, a comprehensive health index with a value between zero and one is calculated. The closer the index is to one, the higher the health level.

8. The remote-controlled torch ignition system based on wireless command and fault self-diagnosis function according to any one of claims 1 to 7, characterized in that, The decision feedback executed by the real-time fault self-diagnosis module includes: uploading the comprehensive health index and diagnostic judgment results to the remote control center via a wireless data link; and storing the calculated comprehensive health index value locally as the basis for adjusting the ignition energy parameters when the next ignition command is generated.

9. A remote-controlled torch ignition method based on wireless command and fault self-diagnosis functions, characterized in that, The method includes the following steps: Step S1: The remote control center generates and sends an encrypted ignition command packet; Step S2: The field control unit receives, decrypts, and executes the instruction packet, and controls the valves to open and the igniter to be powered on in sequence; Step S3: During the ignition process, synchronously acquire multi-dimensional physical quantity timing signals; Step S4: Based on the collected time-series signals, perform real-time fault self-diagnosis, calculate the comprehensive health index, and complete the status diagnosis; Step S5: Based on the diagnostic results, perform decision feedback operations, including data feedback, parameter adjustment, or safety interlock.

10. The remote-controlled torch ignition method based on wireless command and fault self-diagnosis function according to claim 9, characterized in that, The specific steps for performing real-time fault self-diagnosis in step S4 include: Step S41: Align the timing signals acquired from each channel with their reference timing template; Step S42: Extract feature vectors from the aligned data of each channel; Step S43: Calculate the adaptive fusion weights based on the distribution characteristics of the deviation of the feature vectors of each channel; Step S44: Combine the normalized bias of the feature vectors of each channel with the adaptive weights to calculate the comprehensive health index; Step S45: Compare the comprehensive health index with a preset threshold to determine whether the current status of the igniter is healthy, under warning, or faulty.

Images