A ground flare long-lasting lamp group energy-saving control system and method

Abstract

The present application relates to the technical field of industrial flare combustion and fuel gas control, and particularly relates to a ground flare long-lasting lamp group energy-saving control system and method, which comprises: the first long-lasting lamp is continuously and effectively supplied with gas and can be superimposed with incremental gas supply, and at least two second long-lasting lamps are put into use according to needs after emission triggering. A processor determines the conservative remaining arrival time of the gas to be treated according to the emission delivery state calibrated on site, and determines the ignition response margin in combination with the upper limit of the flame building response time of each second long-lasting lamp; when the started long-lasting lamp does not form a flame, only when the other second long-lasting lamp that has not been started still meets the input feasible condition in the complete flame building response time upper limit calculated from the never-started state, the two long-lasting lamps are started simultaneously. The effective flame of the second long-lasting lamp is confirmed according to the local continuous change of the background signal before gas supply and the fuel gas after arrival. The first long-lasting lamp maintains enhanced gas supply before at least one second long-lasting lamp forms a flame, and completes the flame handover in the reverse order when the emission ends.

Description

Technical Field

[0001] This invention relates to the field of industrial flare combustion and fuel gas control technology, and in particular to an energy-saving control system and method for ground flare continuous lamp groups. Background Technology

[0002] Flare systems are used to handle flammable gases emitted from oil, natural gas, liquefied natural gas, and chemical production plants during start-up, shutdown, and emergency venting. Ground-based flares typically have multiple permanent lamps installed near the flare burner assembly to provide an ignition flame when the emitted gases arrive.

[0003] To ensure ignition reliability during sudden emissions, multiple pilot lights typically burn continuously at a predetermined flow rate. Even when the flare is not emitting gas, each pilot light continues to consume fuel gas, and the burner, ignition components, and surrounding heat-resistant parts remain in a high-temperature environment for extended periods. One way to reduce normal gas consumption is to keep some pilot lights burning and activate the remaining ones upon emission detection. Existing controls typically issue start-up commands based on pressure or flow rates in the flare emission pipeline reaching set values. This detection result reflects the emission status at the detection location and does not directly provide the remaining time for the emitted gas to reach the flare burner assembly. When the emission rate changes, the same triggering conditions may correspond to different ignition preparation times. After receiving the start-up command, the pilot lights must undergo processes such as gas supply channel response, fuel gas delivery, ignition, and flame establishment. Each process is affected by gas supply pressure, channel conditions, and environmental conditions, and the actual start-up time for different pilot lights may also differ. Using a uniform start-up time or a fixed retry sequence makes it difficult to simultaneously consider the emission propagation rate and the actual response status of each pilot light.

[0004] Within the ground flare combustion zone, the existing mandrel flame, the adjacent mandrel flame, and the main flare flame may coexist simultaneously. Flame detection signals are easily affected by background radiation and ignition transients. Valve opening, igniter activation, or momentary changes in the detection signal do not independently indicate that the activated mandrel has formed a stable flame. If the gas supply to the existing mandrel is reduced before the activated mandrel actually forms a flame, insufficient ignition capacity may occur during ignition source switching; if multiple mandrels are continuously maintained in combustion to avoid this problem, it is difficult to effectively reduce fuel gas consumption. At the end of the emission period, it is also necessary to confirm that the normally maintained mandrels have resumed stable combustion before stopping the mandrels activated on demand.

[0005] Therefore, it is still necessary to improve the on-demand start-up and gas supply control of the ground flare permanent lamp group, so as to reduce the consumption of normal fuel gas, ensure sufficient ignition capacity before the emission gas arrives, and avoid flame interruption during ignition source switching. Summary of the Invention

[0006] The present invention aims to provide an energy-saving control system and method for ground flare continuous lamp groups, which makes the conservative remaining arrival time of the gas to be treated correspond to the flame-building response capability of each second continuous lamp, and constrains the gas supply state conversion with the actual flame-building confirmation result after the arrival of fuel gas, thereby reducing the normal fuel gas consumption and avoiding insufficient ignition capability caused by improper emission arrival time estimation, delayed start-up of second continuous lamps, or misjudgment of flame status.

[0007] To solve the above-mentioned technical problems, the present invention adopts the following complementary technical solutions.

[0008] A first aspect of the present invention provides an energy-saving control system for a ground flare continuous-burning lamp group, used to control a first continuous-burning lamp and at least two second continuous-burning lamps. The system includes an emission status detection module, a first flame status data interface, second flame status data interfaces corresponding to each of the second continuous-burning lamps, a gas supply and ignition status interface, and at least one processor; the first continuous-burning lamp has a continuously effective flame-preserving gas supply control channel and an incremental gas supply control channel capable of superimposed gas supply, and each second continuous-burning lamp has an independently controllable second start-up channel.

[0009] The processor determines the conservative remaining arrival time of the gas to be treated to the flare burner assembly based on the emission delivery status calibrated on-site. For each second pilot light, it determines the ignition response margin by using the upper bound of the flame build-up response time required to establish a verifiable flame from the current state and the reserved safety time. The ignition response margin represents the time margin after not overestimating the arrival time and taking into account the full response load of the corresponding pilot light.

[0010] After emission is triggered, the processor maintains the flame-keeping gas supply of the first permanent lamp and adds incremental gas supply, then selects the first candidate permanent lamp based on the ignition response margin of each second permanent lamp. Both the first candidate permanent lamp and the additional candidate permanent lamps play a controlling role for the second permanent lamps in a single emission event. Parallel activation indicates that the gas supply and ignition processes of at least two second permanent lamps overlap at least partially in time, but the parallel upgrade request is distinct from the actual deployment of the additional candidate permanent lamps.

[0011] For each activated second permanent light, the processor uses the flame sensing signal before gas supply as background and only confirms a valid flame based on local and continuous signal changes after the fuel gas reaches its combustion zone. If the first candidate permanent light fails to form a valid flame and its ignition response margin is less than the second margin threshold, the processor selects an additional candidate permanent light from the remaining unactivated second permanent lights. Using the current moment as the calculation reference, the processor independently determines the ignition response margin of the additional candidate permanent light according to the upper bound of the complete flame-building response time required for the additional candidate permanent light to form a confirmable valid flame from an unactivated state, without deducting the start-up phase already completed by the first candidate permanent light. The additional candidate permanent light is activated only if the ignition response margin is not less than the feasible lower limit; otherwise, the first permanent light is maintained with enhanced gas supply and a fault is output.

[0012] In some implementations, for the first [unclear] that has not yet issued a start command The ignition response margin of the second permanent lamp is determined according to the following formula:

[0013]

[0014] In the formula, For the first The ignition response margin of the second night lamp The remaining arrival time for the exhaust gases to reach the flare burner assembly. To correspond to the predicted flame-building time of the second continuous lamp from receiving the start command to the formation of a verifiable flame, To allow for a safety margin, for the second permanent flame that has been issued a start command but whose effective flame has not yet been confirmed, the remaining response time after deducting the completed phases and incorporating feedback delay corrections will be used instead of the full predicted flame-building time.

[0015]

[0016] in, This corresponds to the remaining response time required for the second continuously lit lamp at the current moment. Therefore, the gas supply or ignition time already elapsed will not be counted again, while stages that are not completed as expected will still be included in subsequent strategy judgments.

[0017] In some implementations, the emission delivery status is determined by flow time series, pressure time series, or a combination of both obtained from at least two detection locations along the flare discharge pipeline. Flow data can be combined with pipeline effective volume and gas state parameters to determine material delivery time; pressure change characteristics can be used for emission triggering, pipeline storage status correction, or as time characteristics mapped after field calibration, but the propagation time difference of pressure disturbances is not directly equivalent to the material delivery time difference of the gas to be treated. When multiple results are available simultaneously, the shorter remaining arrival time is used. The phased response upper bounds of each second permanent lamp are stored separately and updated according to the actual flame-starting time or startup failure result.

[0018] In some implementations, single-lamp startup is performed when the ignition response margin of the first candidate lamp is not less than a first margin threshold and not less than the feasible lower limit for activation; when its ignition response margin is less than the first margin threshold and not less than a second margin threshold, the gas supply or ignition level of at least one of the first lamp and the first candidate lamp is increased; when its ignition response margin is less than the second margin threshold, a parallel upgrade request is generated. Additional candidate lamps are only activated when the ignition response margin determined by the upper bound of the complete flame-building response time calculated from their unactivated state is not less than the feasible lower limit for activation. The strategy only shifts towards increasing ignition capability until at least one second lamp forms an effective flame.

[0019] In some implementations, the flame-maintaining gas supply and the incremental gas supply of the first pilot lamp are controlled separately. During the flame-maintaining gas supply state, the enhanced gas supply state, and the switching between the two, the flame-maintaining gas supply remains effective, and the incremental gas supply is changed to enable the first pilot lamp to form an enhanced flame or restore the flame-maintaining flame. After the emission state meets the release condition, at least one second pilot lamp that has formed an effective flame continues to burn; after confirming that the first pilot lamp has restored an effective flame, the second pilot lamp is stopped and the first pilot lamp is restored to the flame-maintaining gas supply state.

[0020] A second aspect of the present invention provides an energy-saving control method for a ground flare permanent light assembly, executed by at least one processor, including data acquisition, ignition response margin determination, second permanent light selection, start-up strategy determination, effective flame confirmation, and gas supply status control corresponding to the aforementioned control system. A third aspect provides a non-transient computer-readable storage medium, wherein a computer program stored thereon, when executed by at least one processor, causes the processor to execute the control method.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects:

[0022] Based on the calibrated emission and delivery status, a conservative remaining arrival time is determined, and combined with the upper limit of the individual flame-building response time of each second lamp, a corresponding ignition response margin is formed. This allows the selection of the lamp to establish a definite data connection with the current emission and delivery conditions, channel feedback, and actual flame-building capacity, avoiding the direct misinterpretation of pressure disturbance velocity as the velocity of the gas material to be processed.

[0023] When sufficient time is available, single-lamp start-up is used; when the first candidate lamp responds late, a parallel upgrade request is generated, and the complete response burden of the additional candidate lamp is independently verified. Therefore, the system does not need to activate all second lamps every time emissions occur, nor does it activate an additional lamp that has not had time to form an effective flame simply because the margin of the first candidate lamp has decreased.

[0024] Each second pilot light has its own start-up response parameters, which are updated based on the actual flame build-up time and start-up results. This allows it to reflect response differences caused by different gas supply channels, ignition components, and operating states, providing a basis for selecting pilot lights during subsequent emission processes. Confirming whether a valid flame has formed based on the flame background before the second pilot light is started and the changes in the detection signal after the fuel gas arrives reduces the influence of the original pilot light flame, adjacent flames, and ignition transients on the confirmation results. Changes in gas supply status are based on the actual flame formation confirmation results, avoiding the misinterpretation of start-up commands or gas supply channel opening status as ignition completion.

[0025] While maintaining a continuous and effective flame-keeping gas supply, the first pilot lamp increases or decreases its incremental gas supply and maintains an enhanced gas supply until the second pilot lamp effectively forms a flame. After emissions are completed, the second pilot lamp, which has already formed a flame, is maintained until the first pilot lamp regains a stable flame. Therefore, when the second pilot lamp is activated, the incremental gas supply from the first pilot lamp will not be removed before the second pilot lamp actually forms a flame; when the second pilot lamp is deactivated, its flame will not be extinguished before the first pilot lamp is confirmed to have resumed flame, thus reducing the possibility of flame interruption during ignition source handover. When there are no emissions, the first pilot lamp maintains the flame with a flame-keeping gas supply, and the second pilot lamp stops supplying gas; ignition resources are only activated as needed during emissions. Compared to the continuous combustion of multiple pilot lamps, this reduces fuel gas consumption and the cumulative high-temperature operating time of the second pilot lamp and related components.

[0026] 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

[0027] 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.

[0028] Figure 1 This is a schematic diagram of the functional module architecture of the energy-saving control system for the ground torch continuous lamp group provided in an embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram showing the relationship between the gas supply, ignition, and flame feedback control interfaces provided in an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram illustrating the data processing relationship between propagation analysis, response prediction, and activation strategy provided in an embodiment of the present invention;

[0031] Figure 4 This is a schematic diagram of the application environment and control object topology of the ground torch provided in an embodiment of the present invention;

[0032] Figure 5 This is a flowchart of the normal start-up and differential flameout confirmation process after emission triggering provided in an embodiment of the present invention;

[0033] Figure 6 This is a flowchart of the parallel start-up upgrade and reverse recovery interlock after the emission is completed, provided by an embodiment of the present invention. (a) is the parallel upgrade control during start-up delay, and (b) is the reverse recovery control after the emission is released. Detailed Implementation

[0034] 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.

[0035] The following reference Figures 1 to 6Exemplary embodiments of the present invention are described below. Taking a first permanent light 100, which performs normal flame maintenance and transition enhancement functions, and two second permanent lights 200a and 200b, which are activated on demand, as examples, the processor 500 configures ignition resources according to the gas delivery process and the actual starting capability of each second permanent light. The normal shutdown of the second permanent lights only applies to situations where their corresponding combustion zones are not in operation and the project's safety design allows for on-demand activation; for in-service emission paths that require continuous ignition sources according to the engineering safety design, the required continuous ignition sources are not removed due to energy-saving control. The quantities, signals, operation cycles, and thresholds listed are for illustrative purposes only and do not constitute a numerical limitation on the scope of protection.

[0036] The emission delivery status referred to in this embodiment is a status quantity used to provide the conservative remaining arrival time of the gas to be treated, determined jointly by the time series of different detection locations on the flare emission pipeline 11, the pipeline position relationship, and the on-site calibration relationship. The pressure value or valve position signal at a single moment can be used as a trigger input; the propagation time difference of pressure disturbance is only used as the pipeline storage change characteristic or as an input after calibration mapping, and is not directly used as the delivery time difference of the gas material front.

[0037] The start-up response characteristics referred to in this embodiment are the gas supply control response, fuel gas delivery, ignition establishment, and flame confirmation time stored separately for each second lamp. Second lamps 200a and 200b use their own response parameters, and the processor 500 does not use a single fixed value to represent the start-up capability of all second lamps.

[0038] The ignition response margin, as referred to in this embodiment, represents the time margin remaining before the gas to be treated conservatively reaches the flare burner group 12, after deducting the upper limit of the flame-building response time required for the corresponding second permanent lamp to form a verifiable effective flame from its current state, as well as the reserved safety time. The ignition response margin can be positive, zero, or negative.

[0039] In this embodiment, an effective flame is defined as the flame whose amplitude, normalized intensity, and duration all satisfy the following conditions when the flame sensing signal changes relative to the background baseline before startup, after the fuel gas has reached the combustion zone of the second permanent lamp. Valve opening, igniter activation, or a momentary rise in the flame sensing signal are not directly equivalent to an effective flame.

[0040] In this embodiment, parallel startup refers to the gas supply and ignition processes of at least two second permanent flames overlapping at least partially in time. When the ignition response margin of the first candidate permanent flame is less than a second margin threshold, only a parallel upgrade request is generated. The processor 500 also needs to determine the ignition response margin of the additional candidate permanent flame based on the upper bound of the complete flame-building response time from the unstarted state. Additional startup is only executed when this ignition response margin is not less than the lower limit of feasibility and the corresponding startup channel is available. Parallel startup aims to shorten the latest establishment time of at least one effective flame.

[0041] Example 1

[0042] Reference Figure 4 The ground flare 10 includes a flare discharge pipeline 11, a flare burner assembly 12, and a fireproof and windproof enclosure structure 13. The production unit discharge interface 20 is connected to the flare discharge pipeline 11; the first automatic lamp 100, the second automatic lamp 200a, and the second automatic lamp 200b are set in the ignition area corresponding to the flare burner assembly 12. Figure 4 It is used to represent the topological relationship between the application environment and the controlled object, and does not limit the mechanical installation dimensions of each controlled object.

[0043] An upstream detection point is installed on flare discharge pipeline 11. and downstream detection location . and The pipelines are installed sequentially along the direction of gas propagation, and the center-line distance between them is denoted as . , The remaining propagation distance to the flare burner assembly 12 is denoted as The emission status detection module 400 sends the timestamped data obtained from the two locations to the processor 500.

[0044] Reference Figure 1 The processor 500 executes the data acquisition module 510, the propagation analysis module 520, the response prediction module 530, the strategy determination module 540, the flame confirmation module 550, the output control module 560, and the history update module 570. The aforementioned "modules" represent the data processing functions executed by the processor, and it is not required to set up independent hardware processing units.

[0045] Emissions Status Data Interface 401 Receive , The pressure time series or flow time series; the first flame status data interface 402 receives the flame status of the first lamp 100; the second flame status data interface 403 receives the flame sensing signals of the second lamps 200a and 200b respectively; the gas supply and ignition status interface 404 receives the command status and feedback status of the controlled channel.

[0046] Reference Figure 1 and Figure 2 The output control module 560 outputs a flame-preserving gas supply control command, an incremental gas supply control command, and a second nightlight start command, respectively. The flame-preserving gas supply control command and the incremental gas supply control command act on the flame-preserving gas supply control channel 311 and the incremental gas supply control channel 312, respectively, via the first nightlight gas supply and ignition control interface 601; the second nightlight start command acts on the second start channels 320a and 320b, respectively, via the gas supply and ignition interface 602 of the second nightlight 200a and the gas supply and ignition interface 603 of the second nightlight 200b.

[0047] The gas supply of the first permanent lamp 100 is formed by the superposition of the flame-preserving component and the incremental component. Indicates the amount of gas supplied for fire protection. This indicates the incremental gas supply, specifically the total gas supply for the first continuous light (100). satisfy:

[0048]

[0049] During standby, enhanced, parallel boot, and recovery processes, the processor 500 remains The corresponding fire protection gas supply control is effective, only by changing The first permanent flame 100 is switched between a flame-preserving gas supply state and an enhanced gas supply state. Thus, the enhancement of the first permanent flame 100 is not achieved by first shutting off the flame-preserving channel and then re-igniting, but by increasing the fuel gas while the existing flame continues to exist.

[0050] The control cycle of processor 500 is denoted as When the field controller executes at a fixed cycle, it can be set to 20 milliseconds to 100 milliseconds; the following computational example uses 50 milliseconds. The specific values ​​used in this embodiment to illustrate formula transfer and state transitions are parametric calculation values ​​and do not represent actual field measurement results; the values ​​for engineering deployment should be determined by the project's pipeline, instrumentation, and continuous lighting start-up tests. Each sampled value is accompanied by a timestamp generated by a unified clock.

[0051] Each input setting has a validity flag. or When the data is invalid, the propagation analysis module 520 does not use the data to increase the remaining arrival time; when the flame status data is invalid, the flame confirmation module 550 does not output a valid flame confirmation; when the gas supply or ignition feedback is invalid, the output control module 560 maintains the current relatively safe gas supply state and generates a fault record.

[0052] Data acquisition module 510 with sampling period Read detection position , pressure value , ,in Here is the sampling sequence number. To suppress sensor quantization noise without significantly increasing trigger delay, first-order recursive filtering is applied to both sequences:

[0053]

[0054] In the formula, For the first Filtering pressure at each detection location; These are the filter weights. Noise tests under emission-free conditions determined that the optimal peak-to-peak value after filtering should not exceed half of the original peak-to-peak value, while the step response delay should not exceed one control cycle. This implementation method adopts... =0.35.

[0055] The propagation analysis module 520 determines the pressure change rate based on the filtered pressure. :

[0056]

[0057] During the emission-free calibration period, the average filter pressure was obtained respectively. Standard deviation and the standard deviation of the rate of change Pressure trigger threshold Set as and · The sum of the rates of change threshold Set as · ; and Based on the allowable false trigger probability, this implementation method uses 4 for all values.

[0058] No. Wavefront arrival time at each detection location Take as the first time simultaneously satisfying Not less than and Not less than Sampling time:

[0059]

[0060] To exclude single-point spikes, the aforementioned conditions should be met within a length of At least one short window -1 sampling point is required to establish the system. This implementation method takes... =4, meaning that at least 3 out of 4 consecutive sampling points satisfy the condition. If The wavefront arrival time is no later than ,or - If the fluctuation exceeds the historical propagation time difference range of the pipeline, it will not be considered a valid emission event propagating toward flare burner group 12.

[0061] Reference Figure 3 The propagation analysis module 520 reads emission time series data 410 and pipeline location and distance data 420. For the flow front, or the state characteristics for which a calibrated relationship has been established between the pressure change characteristic time difference and the actual arrival time of the gas to be treated through controlled emissions in the field, after applying a conservative correction to prevent the arrival time from being overestimated, the upper bound of the equivalent transport velocity used for time budgeting is determined according to the following formula:

[0062]

[0063] When the pressure waveform broadens during propagation, causing instability at the threshold crossing moment, the propagation analysis module 520 includes a wavefront length of... Calculate normalized cross-correlation between two data windows :

[0064]

[0065] In the formula, x1(k) and x2(k) are the filtered sequences within the windows corresponding to detection positions P1 and P2, respectively; μ1 and μ2 are the mean values ​​of the corresponding windows; and m is the number of candidate time-shift samples. Let C 12 (m) The candidate time shift that achieves the maximum value is denoted as m*, and the propagation feature time difference is taken as m*Δt. The maximum correlation coefficient reaching the threshold and the time shift being positive only indicates that the two detection positions have obtained identifiable state features in the same direction; when the feature comes from the pressure sequence, it must also be converted through on-site calibration relationship and conservatively corrected, and the acoustic or pressure disturbance velocity cannot be directly used as the material conveying velocity of the gas to be treated.

[0066] Using the detection location as the time reference, and employing the calibrated and conservatively corrected upper bound of the equivalent transport velocity, the lower bound of the initial remaining arrival time of the gas to be treated to reach the flare burner group 12 is:

[0067]

[0068] The first after detecting valid delivery characteristics At any given time, the lower bound of the remaining arrival time decreases according to the elapsed time:

[0069]

[0070] If a new effective equivalent conveying speed is obtained in a continuous control cycle, the maximum value among the three most recent effective values ​​is used for updating, thus ensuring that the remaining arrival time is shortened in a timely manner when the conveying speed is increased. Using the maximum value instead of the average value is to avoid overestimating available time based on slower historical values.

[0071] When both pressure and flow sequences are available, the remaining arrival time corresponding to the pressure is determined based on the actual material arrival at the calibrated pressure sequence, and the remaining arrival time corresponding to the flow rate is determined based on the flow sequence and material balance results; additionally, a conservative remaining arrival time is determined based on the most unfavorable historical discharge conditions. Processor 500 uses the minimum value among the three:

[0072]

[0073] When a signal is lost, the correlation coefficient is below a threshold, the calculation result exceeds the calibration range, or the pressure characteristics have not yet established a calibration relationship with the actual material arrival, the propagation analysis module 520 will not use the remaining arrival time derived from the abnormal or uncalibrated data, but will instead... Set to the non-negative value obtained by subtracting the elapsed time from the shortest arrival time calibrated on-site. This degradation process prevents unreasonable ignition grace caused by sensor malfunctions or pressure wave velocities.

[0074] Figure 3 The propagation analysis module 520 outputs the conservative remaining arrival time and sends it to the ignition response margin determination module 541. Figure 3 The so-called propagation analysis includes material transport estimation and calibrated state characteristic processing; Figure 4 This only indicates the topological location of the detection location, flare discharge line 11, and flare burner group 12.

[0075] The response prediction module 530 saves the startup response characteristics 531 for the second continuous lights 200a and 200b respectively. Based on the second permanent light, the startup process is divided into four stages: gas supply control response, fuel gas delivery, ignition establishment, and flame confirmation. The corresponding times are denoted as follows: , , and .

[0076] No. The second permanent flame lamp's predicted flame-building time from receiving the start command to the formation of a verifiable flame is based on the initial flame-building time. for:

[0077]

[0078] From the moment the start command for the second start channel 320i is written to the output control module 560, until the channel returns to the allowable gas supply state, the measured time is used when valve position feedback is available; when there is only a switch command and no continuous valve position feedback, the nearest measured time is used. The upper quantile of the valve action time during each effective start-up.

[0079] This represents the time required for the fuel gas to propagate from the controlled position to the corresponding combustion zone of the second pilot light. The effective volume of the second start-up channel is known. and current projected volumetric flow rate hour, Calculate using the following formula:

[0080]

[0081] The pressure is determined based on the gas supply pressure, allowable opening degree, and channel flow coefficient. If pressure arrival feedback is installed near the second nightlight, the measured value from the time the command is issued to the time the pressure arrives is used instead. / When both calculated and measured values ​​exist, the larger of the two values ​​shall be used. From the moment the fuel gas reaches the permissible ignition zone until the flame sensing signal first shows a sustained change related to ignition; The flame build-up time begins from the occurrence of this change until the flame confirmation window meets the confirmation conditions. Therefore, the predicted flame build-up time includes the flame confirmation process, rather than considering the start-up complete at the first discharge of the igniter.

[0082] For the The second ever-burning lamp is recently Each effective start-up, average actual flame-building time and standard deviation They are respectively:

[0083]

[0084]

[0085] In recent When the first valid start-up sample meets the predetermined statistical requirements, the response prediction module 530 determines the first effective start-up sample based on the statistical upper quantile characteristics of the actual flame-starting time. Upper bound of the flame response time of the second permanent lamp In one implementation, the upper bound of the flame-building response time is estimated according to the following formula:

[0086]

[0087] In the formula, The confidence coefficient is set to 2 in this implementation method; This is a non-negative correction applied when the current fuel gas pressure is lower than the rated pressure, the ambient temperature is lower than the rated temperature, or the wind speed is higher than the rated wind speed. Insufficient field data. Next, the results were obtained through phased calibration. Alternative The upper bound of the flame response time is a statistically conservative estimate used to control the time budget, and does not imply that the actual startup time will necessarily not exceed this value.

[0088] After a single startup, the historical update module 570 updates the average response value of the corresponding second ever-burning lamp according to the actual flame-starting time. Recursive weighting is used. hour:

[0089]

[0090] This implementation method takes =0.2. When startup fails, instead of averaging the failed samples as normal successful times, the predicted flame-up time of the corresponding second permanent light is increased by a failure penalty. And reduce its selection priority; maintain and recalibrate successfully before canceling the penalty amount.

[0091] Figure 3 The ignition response margin determination module 541 in the middle simultaneously receives , and allow for safety time For the second permanent lamp that has not yet received a start command, the remaining response time required for it to form a verifiable flame from the current moment. equal .

[0092] For already at the moment The second continuous lamp, which has received a start command but whose effective flame has not yet been confirmed, is only in full use. It will repeatedly calculate the startup time that has already been elapsed. Therefore, Update according to the following formula:

[0093]

[0094] In the formula, This is the feedback correction amount. When the allowed gas supply feedback or fuel gas arrival feedback does not occur at the predicted time, Take the current value corresponding to the delay; when the feedback occurs on time, Set it to zero. Thus, the remaining response time not only deducts the completed stages but also reflects the actual startup delay.

[0095] No. Ignition response margin of the second permanent lamp in the current control cycle for:

[0096]

[0097] Allow time for safety It is not used to compensate for known and measurable valve or fuel gas delays, but rather to cover propagation estimation errors, control cycle jitter, and unmodeled disturbances. The upper limit of the absolute value of the difference between the predicted arrival time and the measured arrival time is determined by multiple controlled emission tests, and an additional control cycle is added.

[0098] The strategy determination module 540 sets a first margin threshold and a second margin threshold that is less than the first margin threshold, and also sets a feasible lower limit for investment that is not less than zero. The first and second margin thresholds are used to determine whether an enhancement or parallel upgrade request is generated; the feasible lower limit for investment is used to verify whether the additional candidate continuous lights that have not yet been started still have investment significance after being calculated based on the upper limit of the complete flame-building response time from their unstarted state. Each threshold is determined through engineering time budgeting and on-site start-up calibration.

[0099] like Figure 3 As shown, when the ignition response margin of the first candidate lamp is not less than the first margin threshold, a single-lamp normal start is adopted; when it is less than the first margin threshold but not less than the second margin threshold, an enhanced single-lamp start is adopted; when it is less than the second margin threshold, a parallel upgrade request is generated. After the parallel upgrade request is generated, the processor 500 calculates the ignition response margin of each unstarted candidate lamp, and only invests the additional candidate lamps whose ignition response margin is not less than the lower limit of feasible investment and whose start-up channel is in an available state.

[0100] A one-way escalation strategy is employed within the same emission event. The strategy can be upgraded from routine single-lamp start-up to enhanced single-lamp start-up, and then to parallel start-up; until at least one second pilot lamp forms an effective flame, the escalation is not affected by [other factors]. It briefly recovers and then automatically degrades. This process avoids frequent back-and-forth between gas supply and ignition commands caused by slight fluctuations in propagation velocity estimation.

[0101] If both the second permanent lights 200a and 200b meet the single-lamp start-up conditions, the one with the larger ignition response margin is selected first; if the difference is less than one control cycle, the one with the higher recent historical success rate is selected. Unselected permanent lights remain on standby; when parallel upgrades are required, the current moment is used as the calculation benchmark, and the ignition response margin is re-checked based on the upper bound of the complete flame-building response time of the unactivated permanent light from its unactivated state, without using the remaining response time of the activated permanent lights.

[0102] Example 2

[0103] The flame confirmation module 550 processes the sensing signals of the second night lamps 200a and 200b respectively through the second flame status data interface 403. The second continuous light is obtained during the baseline sampling period before the output start command. Sample values The baseline sampling period should be during the time when the second nightlight is not supplied with gas.

[0104] No. Baseline of the flame background of the second eternal lamp and background standard deviation They are respectively:

[0105]

[0106]

[0107] The baseline sampling period should cover at least one major fluctuation cycle of the flame sensing signal; if no defined cycle is available, Not less than 20. This implementation method adopts... =40, corresponding to a 2-second baseline at a 50-millisecond sampling period. If there isn't enough time to resample after an emission trigger, the nearest sample from the pre-trigger loop cache is used. One valid data point.

[0108] The second eternal light after it was turned on The absolute change of the flame sensing signal relative to the background baseline at any given time. and normalized change for:

[0109]

[0110]

[0111] A small positive number is set to prevent the background standard deviation from approaching zero; this can be one-thousandth of the full scale of the flame sensing channel. Change threshold. The absolute change threshold was determined through calibration tests of the corresponding second lamp ignited individually, and was selected between the stable change range after flame formation and the change range without flame background. Independent calibration tests verified that the false confirmation rate and false negative rate corresponding to this threshold met the project requirements. The normalized threshold was then used. In this implementation method, we take 4.

[0112] In length of Within the confirmation window, the definition satisfies Not less than and Not less than sampling ratio :

[0113]

[0114] In the formula, I{·} is the indicator function. Only after the expected arrival time of the fuel gas, and Not less than the confirmation ratio Only then is it confirmed that the corresponding second permanent lamp has formed a valid flame. This embodiment takes... =10、 =0.8, meaning that at least 8 out of the most recent 10 sampled values ​​meet the double threshold condition.

[0115] Signal rises occurring before the arrival of fuel gas are identified as external radiation or ignition interference and are not used for flame confirmation. When multiple second flame detection channels show approximately proportional changes simultaneously, the flame confirmation module 550 also compares the difference between the activated channel and the non-activated channel; the confirmation result is maintained only when there are local additional changes in the activated channel.

[0116] The valid flame confirmation result is simultaneously sent to the output control module 560 and the history update module 570. Based on this, the output control module 560 releases the restriction on the first permanent light 100 exiting the enhanced gas supply state; the history update module 570 subtracts the start command time from the confirmation time to obtain... Update the startup response characteristics used for the next emission event.

[0117] Example 3

[0118] Reference Figure 5 During the standby phase, where the project's safety design allows for the second permanent light to be activated on demand, the flame-preserving gas supply control channel 311 remains active, and the first permanent light 100 maintains its flame; the incremental gas supply control channel 312 and the second activation channels 320a and 320b are in a non-activation state. The processor 500 continuously updates the emission background and the flame background baseline of the second permanent light.

[0119] when , When valid propagating wavefronts are detected sequentially, processor 500 first maintains Unchanged and increased This causes the first permanent light 100 to enter an enhanced gas supply state; subsequently, the propagation analysis module 520, response prediction module 530, and ignition response margin determination module 541 complete the process. Figure 3 The calculation is shown.

[0120] In a parametric chemical condition used only to illustrate the calculation relationships, the upper limit of the equivalent conveying speed after actual material arrival and conservative correction is set to 50 m / s; the remaining distance from the detection position to the flare burner group 12 is 420 m, and the initial conservative remaining arrival time is 8.40 seconds. This value does not represent the acoustic velocity of pressure disturbance. The gas supply control response time, fuel gas delivery time, ignition setup time, and flame confirmation time of the second permanent lamp 200a are 0.35 seconds, 1.10 seconds, 0.80 seconds, and 0.85 seconds, respectively. It takes 3.10 seconds. A safety margin is allowed. The time is 1.50 seconds, corresponding to the initial ignition response margin. It takes 3.80 seconds. If For 3.00 seconds, If it is 1.20 seconds, then Not less than The strategy determination module 540 selects the second permanent light 200a to perform a single-lamp normal start. The output control module 560, while maintaining the enhanced gas supply to the first permanent light 100, enables the second start channel 320a to perform gas supply and ignition.

[0121] Baseline average value before the second continuous light 200a is activated The background standard deviation is 0.18. The value was 0.02; after startup, the stable sampling value was 0.46. 0.28 It is 14. If 0.12 If the threshold value is 4, and 9 out of the most recent 10 sampling points satisfy the double threshold condition, then... The value is 0.9, and the flame confirmation module 550 confirms the formation of a valid flame. Only after the above valid flame confirmation is the output control module 560 allowed... Lower or set to zero to deactivate the enhanced gas supply mode of the first night light 100; This remains valid. This sequence corresponds to... Figure 5 End-point judgment and Figure 2 The diagram shows the continuous relationship of the fire-preserving gas supply. If the actual flame-building time is 3.60 seconds, the original... For 3.10 seconds If it is 0.2, then the updated The duration is 3.20 seconds. In the next emission event, the response prediction module 530 recalculates using the updated mean and standard deviation. Instead of continuing to use the initial calibration value of 3.10 seconds.

[0122] The aforementioned values ​​are used to illustrate the transfer relationships between the formulas and are computational parameters rather than field test conclusions. In specific engineering projects, the dimensions of distance, time, threshold, gas state parameters, and sensing signals should be determined by the field pipelines, fuel gas conditions, instrument configuration, and continuous lamp start-up tests, and a traceable calibration record should be formed.

[0123] Example 4

[0124] Reference Figure 6 (a) When the second permanent flame 200a has been activated but a valid flame has not yet been confirmed, the processor 500 updates the remaining response time and ignition response margin of 200a in each control cycle. If its ignition response margin is less than the second margin threshold, the processor 500 simultaneously calculates the ignition response margin of 200b independently, based on the upper bound of the complete flame-building response time required for a confirmed flame to be formed in the unactivated state of 200b.

[0125] In the aforementioned computational example, if accelerated emission delivery reduces the ignition response margin of 200a to 0.60 seconds, which is less than the second margin threshold, the strategy determination module 540 generates a parallel upgrade request. This 0.60 seconds only reflects the remaining time budget of 200a and is not directly used as the basis for the deployment of 200b; 200b is only identified as an additional candidate for continuous illumination when its ignition response margin is not less than the lower limit of feasible deployment and the second start-up channel 320b is in a valid state.

[0126] When 200b passes the aforementioned ignition response margin check, the output control module 560 maintains the current state of the flame-keeping gas supply control channel 311, the incremental gas supply control channel 312, and the second start channel 320a, and puts the second start channel 320b into operation, thereby forming a parallel gas supply and ignition process that overlaps in time; when 200b fails the check, it does not issue an invalid additional start command, but maintains the first permanent lamp enhanced gas supply and outputs a fault state that there is no feasible additional candidate permanent lamp.

[0127] The second mandatory lamps 200a and 200b each use their own background baseline, fuel gas arrival time, change threshold, and confirmation window. The flame sensing change of the second mandatory lamp 200a cannot replace the flame formation confirmation of the second mandatory lamp 200b, and vice versa.

[0128] Once any second lamp has formed a valid flame, the flame confirmation module 550 outputs a valid flame status. Based on this, the output control module 560 allows the first lamp 100 to exit the enhanced gas supply state; the other second lamp that has not yet formed a valid flame can stop ignition and delay stopping the gas supply according to the on-site purging requirements.

[0129] If neither of the two second permanent lights produces a valid flame, or if there is no additional candidate permanent light that meets the lower limit of feasible deployment, the first permanent light 100 maintains the enhanced gas supply level allowed by the project, the deployed second start-up channel continues to perform the permitted ignition action, the processor 500 outputs a fault status and enters the backup handling specified in the project safety design. Second permanent lights that only have a start command or channel opening feedback but no valid flame confirmation are not counted as established ignition sources.

[0130] Startup delays also serve as input for historical updates. If a delay occurs but success is ultimately achieved, the actual flameout time is used for updates. and Apply when startup fails Therefore, the strategy escalation in the current emission event and the initial choice of keeping lights on in subsequent emission events are linked by the same set of actual response data.

[0131] Reference Figure 6 (b) When the emission propagation status meets the release condition, the processor 500 does not simultaneously stop the incremental gas supply to the second permanent light and the first permanent light 100. The release condition adopts a release threshold lower than the emission trigger threshold and requires continuous satisfaction of the preset release period to avoid round-trip switching caused by emission boundary fluctuations.

[0132] Upon entering the recovery state, at least one of the second lamps that has formed an effective flame continues to burn. The first flame status data interface 402 acquires the flame sensing signal of the first lamp 100, and the processor 500 confirms whether the first lamp 100 maintains an effective flame according to the same window persistence principle as the second lamp.

[0133] If the first permanent light 100 fails to meet the recovery confirmation conditions, the processor 500 keeps the second permanent light burning, maintaining or increasing its operation. And ignite the first permanent lamp 100; It remains effective throughout the process.

[0134] Only after the confirmation result of the first permanent flame 100 maintaining an effective flame has been continuously maintained for the recovery confirmation period, will the output control module 560 stop the second start channel 320a and / or 320b; thereafter, it will reduce... This restores the first ever-burning lamp 100 to its flame-keeping and gas-supplying state.

[0135] The forward start-up process is constrained by "first lamp enhancement - second lamp activation - second lamp actual flame confirmation", while the reverse recovery process is constrained by "second lamp retention - first lamp flame confirmation - second lamp deactivation". Both directions use the actual flame state, rather than the commanded state, as the permissible condition for changing the gas supply state.

[0136] When a valid wavefront propagating toward the flare burner group 12 is detected again during the recovery process, the processor 500 cancels the recovery state, maintains the current enhanced gas supply to the second pilot lamp flame and the first pilot lamp 100, and returns to normal. Figure 5 The start-up control process is shown.

[0137] Example 5

[0138] Before the system is put into operation, the pipeline distances between detection locations and from the detection locations to the flare burner group 12 should be measured. Under controlled operating conditions covering the predetermined emission range, the flow and pressure time series at each detection location and the actual arrival time of the gas to be treated at the flare burner group 12 should be recorded to establish or verify the calibration relationship between the state characteristic time difference and the actual material arrival time, the conservative correction amount, and the minimum arrival time degradation value.

[0139] The second permanent lights 200a and 200b should also undergo startup tests that meet the project's statistical requirements, recording the gas supply control response, fuel gas delivery, ignition establishment, flame confirmation, and startup failure results. The test scope should cover permissible fuel gas pressure, ambient temperature, and specified adverse wind conditions to determine the upper limit of the phased response, environmental correction, and startup failure penalty; the number of tests is not replaced by the computational example in this embodiment.

[0140] Flame confirmation calibration includes at least four states: all second mandrels are not supplied with gas, the first mandrel 100 is enhanced, only the second mandrel 200a is lit, and only the second mandrel 200b is lit. This is achieved by comparing the performance under each state. , and ,Sure , , and And verify that the adjacent flame will not cause the second unsupplied lamp to be identified as a flame.

[0141] The first margin threshold, the second margin threshold, and the lower limit of feasible implementation are derived by working backward from the time budget. The lower limit of feasible implementation is used to determine whether there is still a positive safety margin after the additional candidate continuous light has completed full gas supply, ignition, and confirmation from standby state; the two upgrade thresholds are used to determine the trigger range of enhancement and parallel requests. After calibration, each threshold and the reserved safety time are managed as controlled safety parameters, and the version is recorded when modifications are made.

[0142] After parameter calibration, the following conditions were verified sequentially: normal single-lamp start-up, propagation acceleration, valve delay, failure of a second nightlight to ignite, external main flame interference, emission release, and recovery failure. For each verification, the processor 500's state transitions, output commands, valid flame confirmation, and historical update results were checked against the actual conditions. Figure 3 , Figure 5 and Figure 6 Consistent.

[0143] The technical effect of this implementation method is not generated by the isolated calculation of propagation time, flame detection, or activation of the backup continuous lamp, but by a closed-loop data system: and Together , The ignition resource configuration is determined, the background differential confirms whether the first permanent flame is allowed to exit the enhanced state, and the actual flame build-up time is then used for updates. The basis for prediction.

[0144] Therefore, when there is sufficient available time, it is not necessary to activate all the second permanent lights at the same time; when the propagation speed is accelerated or the start-up delay causes the time margin to be shortened, the system can bypass the fixed retry sequence to increase the parallel ignition resources, while avoiding false flame confirmation caused by radiation from the main flame or adjacent flames.

[0145] It should be noted that the data acquisition, propagation analysis, response prediction, policy, and security output tasks executed by processor 500 use the same monotonic clock. The data acquisition task writes the time series to a circular buffer; the other tasks only read data with complete timestamps and true validity flags.

[0146] The control states include at least standby flame preservation, emission pre-triggering, single-lamp normal start, enhanced single-lamp start, parallel start, effective flame maintenance, and restoration. Each state transition is determined by emission propagation conditions, Conditions or flame confirmation conditions are triggered, but state transition conditions are not replaced by a single valve position feedback.

[0147] If the propagation analysis task or response prediction task is not completed within one control cycle, the safety output task will use a more conservative approach from the previous cycle. or longer When the flame confirmation data is invalid, a valid flame status is not generated; when communication is interrupted, the field controller maintains the flame of the first continuous lamp 100 or enhances the gas supply.

[0148] The program and parameters for performing the above processing can be stored in a non-transient computer-readable storage medium. When the program is executed by at least one processor, it completes... Figure 1 The module functions shown and Figure 3 , Figure 5 and Figure 6 The data processing and status control are shown.

[0149] The program can be deployed on a single PLC or industrial computer, or flame confirmation and safety output can be handled by a field ignition controller, while propagation analysis and historical updates can be handled by a DCS. Distributed deployment does not change... , , And the data dependency between the valid flame confirmation results.

[0150] In other embodiments, the number of second permanent lamps can be more than two. The processor 500 maintains the start-up response characteristics and flame background parameters of each second permanent lamp, and selects a single lamp to start based on ignition response margin, channel validity, and historical success rate. When parallel upgrades are required, additional candidate permanent lamps are selected only from those non-started permanent lamps whose ignition response margin is still not less than the lower limit of feasible deployment and whose start-up channels are available, after calculating the upper limit of the complete flame-building response time from the non-started state.

[0151] Emission time series can be formed from flow rate, pressure, or a combination of both. Flow rate series can be used for material balance; pressure series can be used for triggering, pipe inventory correction, and conservative arrival time determination after calibration based on actual material arrival. When there is only a unified time reference and detection location distance, but no calibration relationship between pressure characteristics and actual material arrival time, the pressure disturbance time difference is not directly converted into the remaining arrival time of the gas to be treated.

[0152] Flame sensing signals can be ultraviolet, infrared, ion current, or image features. Different detection principles use corresponding units of measurement. , and However, backgrounds were obtained when the second lamp was not supplied with gas, and differential and window confirmations were performed after the fuel gas arrived.

[0153] Without changing Figure 1 The functional relationships shown are Figure 3 Given the time budget relationship shown, the filter, statistical estimation method, and threshold value can be adjusted according to the on-site noise and risk level. These adjustments do not alter the technical concept that the emission propagation state and individual initiation response jointly determine the ignition strategy, and that the actual flame formation result limits the gas supply state transition.

[0154] Example 6

[0155] Combination Figure 3 , Figure 5 and Figure 6This embodiment demonstrates the field verification of the emission propagation detection, continuous lamp activation, effective flame confirmation, and gas supply handover logic of the system of the present invention. The following field test data are derived from the controller's historical database and event log files, and the test period was from June 15th to 17th, 2026. Content separately marked as "logic recalculation" is only used to verify the corresponding control branches and is not considered as complete field test results. The system is configured with one first continuous lamp 100 and two second continuous lamps 200a and 200b; in the field tag numbering, the first continuous lamp 100 is designated as Pilot-1, and the second continuous lamps 200a and 200b are designated as Pilot-2A and Pilot-2B, respectively.

[0156] The first permanent flame lamp 100 has a continuous gas supply channel during the test, while the incremental gas supply channel can be independently activated or deactivated depending on the control status. The second permanent flame lamps 200a and 200b each have independent fuel gas supply and ignition start-up channels. The fuel gas for the permanent flame lamps is piped natural gas with a methane volume fraction of not less than [amount missing]. Low calorific value The gas supply pressure is (gauge pressure).

[0157] Considering the safety of on-site testing, materials with a purity of not less than [a certain value] were used. The unit produces nitrogen as a byproduct to simulate the gas to be treated, and the upstream storage tank pressure is... (Gas gauge pressure), the simulated emission intensity is changed by adjusting the valve. Since the simulated gas does not participate in combustion, this embodiment verifies the emission propagation detection, time budgeting, continuous lamp activation, flame confirmation, and gas supply handover logic, and does not use this test to evaluate the flare's ability to handle combustible emissions.

[0158] Detection location , Pressure transmitters were installed separately, and the pressure transmitters used were Rosemount 3051S type pressure transmitters with a range of [missing information]. Output Signal. The second permanent flame lamp uses a Honeywell C7027A ultraviolet flame detector. The detector signal is converted into a 0–10 V DC signal that the controller can collect via a matching flame signal processing unit; the flame sensing signal values ​​mentioned below are all output values ​​of the signal processing unit. A gas analyzer is installed near the flare burner group 12. The sampling point of the gas analyzer is set near the flare burner group 12. The time when the oxygen volume fraction drops from the background value before emission to below 20.5% and remains there for a predetermined confirmation time is used as the criterion for determining whether the nitrogen front has arrived at the sampling point. The response delay of the gas analyzer and sampling pipeline is deducted from the recording time according to the pre-calibrated value.

[0159] Detection location To the detection location Distance along the centerline of the pipeline for Detection location Distance from the center of the flame area of ​​flare burner group 12 along the pipeline centerline for The above distances were determined based on the completed pipeline data of the project. For the measurable pipe sections on site, laser rangefinders and measuring tapes were used for segmented verification.

[0160] The raw sampling period for the pressure transmitter signal is 0.1 s, and historical trend data is stored at 1-second intervals. Pressure threshold exceedances, start-up commands, valve position feedback, fuel gas arrival feedback, and flame confirmation events are recorded with a time resolution of no more than 0.1 s. The sampling period for the flame sensing signal is... All detection signals and control events are time-stamped using the same controller clock. The main controlled parameters used in this experiment are shown in the table below:

[0161] parameter numerical values Parameter meaning Pressure trigger threshold Allow time for safety First margin threshold for single-lamp normal start and enhanced start The second margin threshold for generating parallel upgrade requests Feasible lower bound for additional candidate continuous lamps No. The threshold of absolute change of the second permanent light No. The normalized change threshold of the second permanent light Number of sampling points contained in the flame confirmation window Confirm the minimum proportion that meets the conditions in the window

[0162] Single-lamp start-up test under normal transport conditions

[0163] Slowly open the simulated emission control valve to create a low-flow emission condition, simulating an emission gas temperature of 25 °C. Detection location. , The following is an excerpt from the pressure record.

[0164] Pre-emission status The pressure started to rise The pressure trigger threshold has not yet been reached. — First time exceeding , recorded as The pressure characteristics continue to spread — First time exceeding , recorded as — Emission status continues

[0165] The same pressure change characteristics were detected at the location. , propagation time difference for:

[0166]

[0167] Gas analyzer in A nitrogen leading edge was detected, therefore, in this experiment, the simulated emission gas was emitted from the detection location. Measured delivery time to flare burner assembly 12 for:

[0168]

[0169] Corresponding single-cycle calibrated equivalent conveying speed for:

[0170]

[0171] To avoid overestimating the remaining arrival time, the upper bound of the equivalent conveying speed is set. It should not be less than the measured equivalent conveying speed. During this review, the upper limit of the equivalent conveying speed was adjusted upwards to:

[0172]

[0173] Corresponding conservative initial remaining arrival time for:

[0174]

[0175] The determined Less than the measured delivery time of this test The remaining arrival time of the simulated emissions was not overestimated. The engineering operation adopted... The determination should be based on the results of multiple on-site calibrations covering the predetermined emission range, and a single test result should not replace the complete calibration.

[0176] Startup command issued — Valve opening feedback Fuel gas reaches the combustion zone Valid flame confirmation Complete flame response time —

[0177] Processor 500 A start command is sent to the second permanent light 200a. The start-up process of the second permanent light 200a is recorded as follows.

[0178] The actual flame-building time of the second eternal lamp 200a for:

[0179]

[0180] A post-event review of the time budget based on the actual construction time of the flame project yielded the following results:

[0181]

[0182]

[0183] because:

[0184]

[0185] The subsequent verification results are consistent with the system's control actions of selecting the second permanent light 200a to perform a normal single-lamp start and the second permanent light 200b to remain in standby mode. The effective flame confirmation time of the second permanent light 200a is earlier than the actual arrival time of the simulated emission gas front, and the time margin between the two is:

[0186]

[0187] Before the second lamp 200a receives a gas supply command, the processor 500 uses the 40 most recent valid sampled values ​​in the pre-supply command buffer to determine the flame background baseline and background standard deviation of the second lamp 200a; with a 50 ms sampling period, the 40 sampled values ​​correspond to a 2 s baseline sampling period. The background mean is calculated based on the 40 sampled values.

[0188]

[0189] And background standard deviation:

[0190]

[0191] After the fuel gas from the second lamp 200a reaches the combustion zone, The processor 500 acquires the 10 most recent flame sensing signal samples and uses these 10 samples to form an confirmation window. The average value of the flame sensing signal within this confirmation window is:

[0192]

[0193] Absolute change for:

[0194]

[0195] Normalized change for:

[0196]

[0197] in, A preset positive number used to avoid a denominator of zero, its value is relative to the value in this experiment. It can be ignored.

[0198] Therefore, we can conclude that:

[0199]

[0200] All 10 sampling points within the confirmed window meet the aforementioned dual threshold conditions, therefore the window satisfies the proportion requirement. for:

[0201]

[0202] because:

[0203]

[0204] And at this time, the conditions for the arrival of fuel gas for the second lamp 200a have been met, and the processor 500 is in... It was confirmed that the second permanent lamp 200a had formed an effective flame.

[0205] Before a valid flame is confirmed, both the flame-preserving gas supply channel and the incremental gas supply channel of the first permanent lamp 100 remain engaged. After confirming that the second permanent lamp 200a has formed a valid flame, the processor 500 removes the incremental gas supply to the first permanent lamp 100, restoring the first permanent lamp 100 to its flame-preserving gas supply state. Throughout the entire startup and gas supply handover process, the flame-preserving gas supply to the first permanent lamp 100 remains uninterrupted.

[0206] Additional candidate continuous lamps put into calibration test under accelerated conveying conditions

[0207] Quickly fully open the simulated discharge control valve to simulate a large discharge flow rate. Detection location. exist When the pressure is detected to reach Detection location exist When the pressure is detected to reach .

[0208] The propagation time difference of the pressure characteristic under this operating condition is:

[0209]

[0210] The propagation time difference is smaller than that in the normal transport test. This indicates that the emission transport process is accelerating. Based on the field calibration relationship corresponding to this operating condition, the upper limit of the equivalent transport speed used by processor 500 is:

[0211]

[0212] Therefore, in The conservative initial remaining arrival time is determined as follows:

[0213]

[0214] The actual arrival time of the nitrogen leading edge, as recorded by the gas analyzer, was: That is, from the detection location The measured delivery time to torch burner assembly 12 is:

[0215]

[0216] because:

[0217]

[0218] The adopted The conservative estimate of the remaining arrival time did not overestimate the actual delivery time.

[0219] Processor 500 A start command was issued to the first candidate pilot light 200a. Due to instantaneous fluctuations in fuel gas pressure, the second pilot light 200a failed to form a valid flame as originally predicted. At that time, the processor 500 calculates the ignition response margin of the second night lamp 200a as follows:

[0220]

[0221] because:

[0222]

[0223] Processor 500 generates a parallel upgrade request.

[0224] exist When the pressure change characteristics reach the detection position The process has already begun:

[0225]

[0226] The current conservative remaining arrival time is:

[0227]

[0228] The upper bound of the complete flame-building response time required for the second permanent light 200b, which has not yet been activated, to complete gas supply control response, fuel gas delivery, ignition establishment, and flame confirmation from standby state is:

[0229]

[0230] The ignition response margin of the second permanent lamp 200b is:

[0231]

[0232] because:

[0233]

[0234] The processor 500 determines that the second permanent light 200b cannot generate a valid positive time margin after taking into account its full flame-building response burden and reserved safety time. Therefore, the system does not issue a start command to the second permanent light 200b, but instead outputs a fault status indicating that there is no feasible additional candidate permanent light, switches the first permanent light 100 to the highest level of enhanced gas supply, and continues the start-up process that the second permanent light 200a has already started.

[0235] The second eternal lamp 200a was finally in use. When an effective flame is formed, the confirmed time of the effective flame is earlier than the actual arrival time of the simulated emission gas front.

[0236]

[0237] During the test, the valve and ignition channel of the second permanent light 200b did not receive a start command; after the test, an inspection of its valve and ignition system revealed no equipment malfunction. This test shows that a parallel upgrade request is not directly equivalent to starting all remaining second permanent lights, and additional candidate permanent lights must independently pass the feasibility verification according to their own complete flame-building response time upper limit.

[0238] Parallel input via offline logic recalculation of branches

[0239] The existing records do not include the complete pressure time series corresponding to this branch, the actual arrival records of simulated exhaust gases, and the continuous event records of each start-up channel. Therefore, the following content is only based on offline logic recalculation of the saved control nodes to illustrate the control actions when the additional candidate continuous lamp meets the conditions for deployment, and is not intended as field test results or performance verification data.

[0240] In this recalculated operating condition, the pressure variation characteristics are as follows: Arrive at the detection location The conservative initial remaining arrival time determined by processor 500 is:

[0241]

[0242] Processor 500 The second emergency light (200A) will then be activated. At that time, the second lamp 200a had not yet formed an effective flame, and the processor 500 generated a parallel upgrade request.

[0243] At this point, the self-pressure change characteristic reaches the detection position. The process has already begun:

[0244]

[0245] The current conservative remaining arrival time is:

[0246]

[0247] The upper bound of the complete flame-building response time of the second permanent lamp 200b is:

[0248]

[0249] Therefore, the ignition response margin of the second permanent lamp 200b is:

[0250]

[0251] because:

[0252]

[0253] The second permanent light 200b meets the conditions for being put into operation. According to this control branch, the processor 500 maintains the startup process of the second permanent light 200a and simultaneously starts the second permanent light 200b.

[0254] In the saved control nodes, the second continuous light 200b is... First, an effective flame is formed, and then the second continuous lamp 200a is in... When an effective flame is formed, that is:

[0255]

[0256] Processor 500 uses the first formed effective flame of the second eternal lamp as a condition for gas supply handover permission. The increased gas supply to the first ever-burning lamp 100 was then removed.

[0257] This recalculation, together with the aforementioned failure of the input verification, illustrates that for the additional candidate permanent light 200b, the system always calculates its own ignition response margin according to the upper limit of the complete flame-building response time required for the light to form a verifiable effective flame from an unstarted state, without mistakenly including the start-up phase already completed by the second permanent light 200a in the calculation of the second permanent light 200b.

[0258] False confirmation suppression test under external radiation interference conditions

[0259] When the system is in standby mode and the second permanent lamp 200a is not supplied with fuel gas, a temporary acetylene flame is set up in a safe area outside the flare enclosure structure to generate external radiation interference to the ultraviolet flame detector of the second permanent lamp 200a.

[0260] Before the interference, the background mean value of the flame sensing signal of the second permanent lamp 200a was:

[0261]

[0262] The background standard deviation is:

[0263]

[0264] The peak value of the flame sensing signal during the interference period is:

[0265]

[0266] Maximum absolute change for:

[0267]

[0268] The corresponding maximum normalized change for:

[0269]

[0270] Therefore, we can conclude that:

[0271]

[0272] Based on the recalculation of the original values, it was confirmed that the number of sampling points within the window that simultaneously meet both the absolute change threshold and the normalized change threshold is 0, and the corresponding window satisfaction ratio is:

[0273]

[0274] therefore:

[0275]

[0276] Meanwhile, the start-up channel of the second permanent light 200a remained in a gas-free state, and the conditions for its fuel gas arrival were not met. Throughout the entire interference process, the processor 500 did not generate a valid flame confirmation signal for the second permanent light 200a, nor did it change the gas supply status of the first permanent light 100 due to external radiation interference.

[0277] This experiment verified the combined suppressive effect of fuel gas arrival conditions, absolute change threshold, normalized change threshold, and window duration conditions on false positives of external radiation.

[0278] Reverse recovery interlock test under emission release conditions

[0279] With the second permanent lamp 200a confirmed to have formed a valid flame and the first permanent lamp 100 in flame-preserving gas supply mode, the simulated emission regulating valve is slowly closed to bring the detection position to a safe position. The pressure continues to decrease.

[0280] When detection location The pressure is below the emission release threshold and remains continuously during the release confirmation period:

[0281]

[0282] Subsequently, processor 500 determines that the emission release condition has been met and enters the recovery state. The state transitions during the recovery process are shown in the table below.

[0283] Emissions removal conditions met Unstable flame signal Valid flame confirmed Maintain the gas supply and combustion of the second night lamp 200A. The first ever-burning lamp has been restored. Valid flame conditions not met Continue to burn Maintain gas supply for fire prevention and increase gas supply. The flame of the first eternal lamp was restored. The flame signal re-satisfies the confirmation window conditions. Continue burning Simultaneously keep both lights on and begin resuming the confirmation timer. Continue to meet recovery conditions Effective flame duration Continue to burn Recovery confirmation time reached The second eternal lamp was deactivated. Stable combustion has been confirmed. Stop gas supply and ignition First, exit the second continuous light 200a Restore fire protection status Ensure continuous and effective fire and gas supply Exited Then remove the increased gas supply from the first continuously lit lamp 100.

[0284] Among them, the recovery confirmation period for:

[0285]

[0286] While the first permanent flame 100 has not yet met the valid flame confirmation condition again, the processor 500 does not stop the second permanent flame 200a, but maintains the existing flame of the second permanent flame 200a and strengthens the first permanent flame 100 by increasing the gas supply. This only applies if the valid flame confirmation result of the first permanent flame 100 is continuously maintained.

[0287]

[0288] Then, the system first stops the gas supply and ignition of the second pilot light 200a, and then removes the incremental gas supply to the first pilot light 100. Throughout the entire restoration process, at least one confirmed ignition flame is maintained.

[0289] The on-site records and logical recalculation results of this embodiment are summarized as follows.

[0290] T-01 Normal delivery single lamp start Field test The second permanent lamp 200a forms a flame before the simulated exhaust gases arrive. T-02 Accelerated transport and additional candidate continuous light verification Field test ,but Invalid parallel startup not executed T-03 Parallel input via branch Offline logic recalculation The recalculation results show that the second permanent light 200b meets the parallel start-up conditions. T-04 External radiation interference suppression Field test , and No misconfirmation occurred. T-05 Emissions removal and reverse recovery Field test The first continuous light confirmation time has reached [time]. Then, the second continuous light and the increased gas supply were turned off in sequence.

[0291] The above field records verified normal single-lamp start-up, additional candidate continuous lamp activation verification, external radiation interference suppression, and reverse recovery control, respectively; offline logic recalculation demonstrated the parallel start-up branch when the additional candidate continuous lamp meets the activation feasibility conditions. The two types of results have different evidentiary natures and are not interchangeable. In the above field verification and offline logic recalculation, the processor 500 will use the actual gas to reach the calibrated conservative remaining arrival time. The upper bound of the flame response time of each of the second eternal lamps Based on this, determine the ignition response margin for the corresponding second permanent light:

[0292]

[0293] For the second permanent lamp that has been activated but has not yet been confirmed to form a valid flame, its remaining response time (#imgpt400#) is used to determine the appropriate parameters:

[0294] #imgpt401#

[0295] When the first candidate continuous lamp is delayed in starting, whether the additional candidate continuous lamp is put into operation depends on its own complete flame response burden, rather than on the start-up time already experienced by the first candidate continuous lamp.

[0296] Simultaneously, the removal of incremental gas supply from the first mandatory lamp 100 is conditional upon at least one second mandatory lamp actually forming an effective flame; when emissions are released, the second mandatory lamp is deactivated only if the first mandatory lamp 100 reforms and maintains an effective flame. Thus, emission propagation status, individual flame-building response, local continuous flame confirmation, and forward and reverse gas supply handover form a continuous control data chain.

[0297] In summary, this invention uses the calibrated conservative remaining arrival time and the upper bound of the individual flame-building response time of each second pilot lamp to form a candidate-specific time budget. When the first candidate pilot lamp fails to ignite, the additional candidate pilot lamps must pass the input verification based on the upper bound of their complete flame-building response time calculated from their unstarted state to avoid invalid parallel operation. The gas supply state transition is also constrained by the local continuous flame-building confirmation after the arrival of fuel gas. The first pilot lamp maintains its enhancement during forward start-up until the second pilot lamp actually ignites, and retains the second pilot lamp until the first pilot lamp recovers and is confirmed to have ignited when emissions are released. This reduces normal fuel gas consumption and maintains flame handover continuity within the on-demand input range allowed by the project's safety design.

[0298] The above description represents exemplary embodiments of the present invention. Those skilled in the art can make adjustments to the component form, connection relationships, control thresholds, and action timing without departing from the inventive concept. The scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. An energy-saving control system for a ground flare continuous-burning lamp group, used to control a first continuous-burning lamp and at least two second continuous-burning lamps, wherein the first continuous-burning lamp includes a continuously supplied flame-preserving gas supply channel and an incremental gas supply channel arranged parallel to the flame-preserving gas supply channel, and each of the second continuous-burning lamps includes an independently controllable start-up channel, the system including an emission status detection module, a first flame status data interface, second flame status data interfaces corresponding to each of the second continuous-burning lamps, a gas supply and ignition status interface, and at least one processor; characterized in that: The at least one processor is programmed to: The emission delivery status in the flare emission pipeline output by the emission status detection module, the flame sensing signal of the first lamp output by the first flame status data interface, the flame sensing signal of the corresponding second lamp output by each of the second flame status data interfaces, and the command status and feedback status of each of the start channels are obtained. Based on the emission delivery status calibrated on-site, the conservative remaining arrival time of the gas to be treated to the flare burner group is determined. For each second lamp, the ignition response margin is determined based on the upper limit of the flame-building response time required for the second lamp to form a verifiable effective flame from the current state and the reserved safety time. In response to the emission delivery state meeting the emission triggering condition, while maintaining the continuous gas supply of the fire-preserving gas supply channel, the incremental gas supply channel is put into operation. The first candidate lamp is selected from the second lamps whose ignition response margin is not less than the preset feasible lower limit and whose activation channel is in an available state, and the gas supply and ignition of the first candidate lamp are controlled. The flame background baseline of the first candidate lamp is determined based on the flame sensing signal obtained before the first candidate lamp is supplied with gas. Only after the fuel gas reaches the combustion area of ​​the first candidate lamp, the formation of an effective flame is confirmed based on the local additional change of its flame sensing signal relative to the flame background baseline and the persistence of the local additional change. When the first candidate lamp has been started but has not yet been confirmed to form an effective flame, and its ignition response margin is less than the second margin threshold, an additional candidate lamp is selected from the remaining second lamps that have not yet been started. Using the current time as the calculation benchmark, the ignition response margin of the additional candidate lamp is independently determined according to the upper bound of the complete flame-building response time required for the additional candidate lamp to form a confirmed effective flame from an unstarted state. When determining the ignition response margin of the additional candidate lamp, the start-up phase already completed by the first candidate lamp is not deducted, and the start-up process of the first candidate lamp is maintained and the additional candidate lamp is started only when the ignition response margin of the additional candidate lamp is not less than the preset feasible lower limit and its start-up channel is available. The incremental gas supply to the incremental gas supply channel may be removed only after at least one second permanent lamp has been confirmed to have formed a valid flame; if there is no additional candidate permanent lamp that meets the lower limit of the feasible activation, the incremental gas supply channel shall be maintained in operation and a fault status shall be output; wherein, the activation command or opening feedback of the activation channel shall not be used as a substitute confirmation of the valid flame.

2. The energy-saving control system for the ground torch continuous lighting group according to claim 1, characterized in that: The at least one processor is also programmed to: Determine the conservative remaining arrival time , No. Upper bound of the flame-building response time required for the second eternal flame to form a confirmed effective flame from the current state. and allow for safety time ; The number is determined according to the following formula. The ignition response margin of the second lamp: in, The numbering of the second impermeable lamp; for the second impermeable lamp that has received the start command but has not yet confirmed the formation of a valid flame, the remaining response time is calculated after deducting the completed phase and incorporating feedback delay corrections. Replaces the upper bound of the flame response time. .

3. The energy-saving control system for the ground torch continuous lighting group according to claim 2, characterized in that: The emission delivery status includes at least one of a flow rate time series and a pressure time series obtained at at least two detection locations on the flare discharge pipeline, and the at least one processor is further programmed to: When using the flow time series, the first remaining arrival time is determined based on the flow rate, the effective volume of the pipeline between the corresponding detection location and the flare burner group, and the emission gas state parameters. When using the pressure time series, the propagation time difference of the same pressure change feature between at least two detection locations is determined, and the propagation time difference is used as the input of the field calibration relationship. Based on the correspondence between the propagation time difference obtained during controlled emission and the actual arrival time of the gas to be treated, a conservatively corrected upper limit of the equivalent transport speed is determined. Then, based on the upper limit of the equivalent transport speed and the remaining propagation distance from the corresponding detection location to the flare burner group, a second remaining arrival time is determined. The propagation time difference is not directly used as the material transport time difference of the gas to be treated. When both the first remaining arrival time and the second remaining arrival time are available, the shorter one is selected as the conservative remaining arrival time; when the detection signal is invalid or exceeds the applicable range of the calibration relationship, the shortest arrival time calibrated on-site is used.

4. The energy-saving control system for the ground torch continuous lighting group according to claim 2, characterized in that: The upper limit of the flame-starting response time includes the phased upper limits of the gas supply control response time, fuel gas delivery time, ignition establishment time and flame confirmation time corresponding to the second permanent lamp. The at least one processor is further programmed to correct the upper bound of the phased time based on gas supply feedback, fuel gas arrival feedback, and at least one of fuel gas pressure, ambient temperature, and wind speed, and to determine the remaining response time required for the corresponding second permanent light from the current state based on the corrected upper bound of the phased time. After the corresponding second lamp completes a valid start, the actual flame-building time of that start is used as a historical start sample to update the upper bound of the flame-building response time of the second lamp; if the corresponding second lamp fails to start, the failure penalty of the second lamp is increased and its selection priority in subsequent emission events is reduced.

5. The energy-saving control system for the ground torch continuous lighting group according to claim 2, characterized in that: The at least one processor is also programmed to: When the ignition response margin of the first candidate lamp is not less than the first margin threshold and not less than the lower limit of the feasible input, the first candidate lamp is controlled to perform gas supply and ignition independently. When the ignition response margin of the first candidate lamp is less than the first margin threshold and not less than the second margin threshold, the fuel gas supply level or ignition level of at least one of the first lamp and the first candidate lamp is increased. A parallel upgrade request is generated when the first candidate lamp has not yet formed an effective flame and its ignition response margin is less than the second margin threshold. The first candidate lamp and the additional candidate lamp are controlled to perform gas supply and ignition in parallel only when the ignition response margin determined by the upper bound of the complete flame-building response time of the additional candidate lamp calculated from its unstarted state is not less than the lower limit of the feasible input. The second margin threshold is less than the first margin threshold.

6. The energy-saving control system for the ground torch continuous lighting group according to claim 1, characterized in that: The at least one processor is also programmed to: During the baseline sampling period prior to supplying fuel gas to the selected second lamp, the flame background baseline and background standard deviation are determined based on the flame sensing signal corresponding to the second lamp. After the fuel gas reaches the combustion zone of the second lamp, the absolute change of the flame sensing signal relative to the flame background baseline and the change normalized according to the background standard deviation are determined, and the change of the flame sensing signal of the activated second lamp relative to its flame background baseline is compared with the change of the flame sensing signal of the unactivated second lamp relative to their respective flame background baseline. The second lamp is confirmed to form an effective flame only when both the absolute change and the normalized change reach their respective thresholds, the activated second lamp has local additional changes, and the local additional changes meet the persistence condition within the confirmation window.

7. The energy-saving control system for the ground torch continuous lighting group according to claim 1, characterized in that: The fire-preservation gas supply channel and the incremental gas supply channel are independently controlled; The at least one processor is also programmed to maintain continuous gas supply to the fire-preserving gas supply channel during standby, enhanced gas supply, second night lamp activation and recovery, and to switch the first night lamp between fire-preserving gas supply state and enhanced gas supply state by adjusting the incremental gas supply channel.

8. The energy-saving control system for the ground torch continuous lighting group according to claim 7, characterized in that: The at least one processor is also programmed to: In response to the emission delivery status meeting the emission release condition, while keeping at least one second pilot lamp with an effective flame burning, it is confirmed whether the first pilot lamp maintains an effective flame; If it is not confirmed that the first lamp is maintaining an effective flame, the second lamp is kept burning, and the flame-preserving gas supply channel is kept in continuous gas supply and the incremental gas supply channel is put into operation. Only after confirming that the first lamp maintains a valid flame and continues to maintain the recovery confirmation period, stop the gas supply and ignition of the second lamp, and then remove the incremental gas supply from the incremental gas supply channel.

9. A method for energy-saving control of ground torch continuous-burning lamp groups, characterized in that: Executed by at least one processor, the method controls a first permanent light and at least two second permanent lights. The first permanent light has a continuously supplied flame-keeping gas channel and an incremental gas supply channel capable of supplying gas in conjunction with the flame-keeping gas channel. Each of the second permanent lights has an independently controllable start-up channel. The method includes: The emission delivery status within the flare emission pipeline, the flame sensing signal of the first permanent lamp, the flame sensing signals of each of the second permanent lamps, and the command status and feedback status of each of the start-up channels are acquired. The conservative remaining arrival time of the gas to be treated is determined based on the emission and delivery status calibrated on site, and the ignition response margin of each second lamp is determined based on the upper limit of the flame-building response time and the reserved safety time of each second lamp. In response to the emission triggering condition, while maintaining the continuous gas supply of the flame-preserving gas supply channel, the incremental gas supply channel is put into operation. The first candidate lamp is selected from the second lamps whose ignition response margin is not less than the preset feasible lower limit and whose activation channel is in an available state, and the gas supply and ignition of the first candidate lamp are controlled. The flame background baseline of the first candidate lamp is determined based on the flame sensing signal obtained before the first candidate lamp is supplied with gas. Only after the fuel gas reaches the combustion area of ​​the first candidate lamp, the formation of an effective flame is confirmed based on the local additional change of its flame sensing signal relative to the flame background baseline and the persistence of the local additional change. When the first candidate lamp has been activated but has not yet been confirmed to form a valid flame, and its ignition response margin is less than the second margin threshold, an additional candidate lamp is selected from the remaining second lamps that have not yet been activated. Using the current moment as the calculation benchmark, the ignition response margin of the additional candidate lamp is independently determined according to the upper bound of the complete flame-building response time required for the additional candidate lamp to form a confirmed valid flame from an unactivated state. When determining the ignition response margin of the additional candidate lamp, the activation phase already completed by the first candidate lamp is not deducted. The activation process of the first candidate lamp is maintained and the additional candidate lamp is activated only when the ignition response margin of the additional candidate lamp is not less than the preset feasible lower limit and the activation channel of the additional candidate lamp is available. If there is no additional candidate lamp that meets the feasible lower limit, the incremental gas supply channel is maintained and a fault status is output. The incremental gas supply to the incremental gas supply channel may be removed only after it is confirmed that at least one of the second pilot lights has formed an effective flame; in response to the emission release condition, the second pilot lights that have formed an effective flame continue to burn, and the flame-preserving gas supply channel and the incremental gas supply channel are maintained in operation until it is confirmed that the first pilot light has maintained an effective flame; only after it is confirmed that the first pilot light has maintained an effective flame, the gas supply and ignition of the second pilot lights are stopped, and then the incremental gas supply to the incremental gas supply channel is removed.

10. A non-transient computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by at least one processor, the at least one processor performs the energy-saving control method for the ground torch continuous lamp group as described in claim 9.

Images