A method for intelligent opening and closing control of gas fire extinguishing equipment

Abstract

This invention relates to the field of equipment control technology, specifically to an intelligent opening and closing control method for a gas fire extinguishing device. The method includes the following steps: acquiring infrared images to generate temperature rise rate and edge stability parameters; controlling the valve by acquiring pressure and displacement data to construct rate data; cross-correlation to identify deviations and correction values; loading correction values ​​to locate breakpoints and calling parameter groups to generate a reconstruction sequence; and determining fire extinguishing stability after release to generate a closing command. In this invention, temperature rise rate and edge stability parameters are constructed using thermal imaging, enabling multi-dimensional hotspot feature judgment and triggering control signals. The action hysteresis time window is acquired and the drive signal is corrected, improving valve response accuracy and timing coordination, ensuring the stability and continuity of the release process. Combined with temperature decay slope regression analysis, fire extinguishing stability is judged, achieving control convergence when the fire extinguishing effect is achieved, thus improving the dynamic adaptability of the control process and the closed-loop response capability of the fire extinguishing operation.

Description

Technical Field

[0001] This invention relates to the field of equipment control technology, and in particular to an intelligent opening and closing control method for gas fire extinguishing equipment. Background Technology

[0002] The field of equipment control technology involves the manipulation and management of the operational status of various automated equipment, including the start-up and shutdown control of actuators, status detection, command response, adjustment of operating parameters, and the formulation and execution of control logic. The core aspects of this technology are: achieving automatic start-up, shutdown, adjustment, and status feedback of equipment under specific conditions through hardware drive components, electrical signal transmission paths, logic control circuits, and program control units. Its applications cover multiple areas such as industrial production, security systems, energy dispatching, and traffic control, typically involving relay control, electromagnetic drive control, motor control, time-series control, and real-time response mechanisms to sensor-acquired data. The development of this technology aims to improve control accuracy, response speed, system stability, and the flexibility of control strategies. Control methods include timing control, condition-triggered control, closed-loop feedback control, and combinational logic control.

[0003] Traditional intelligent opening and closing control methods for gaseous fire extinguishing equipment refer to the process of opening and closing valves by controlling the release and blocking of the extinguishing medium in the gaseous fire extinguishing equipment. This method typically uses a triggering electrical signal generated by a fire detection system to control an electromagnetic starting device, driving the valve to open and release the gaseous extinguishing agent to the fire area. Its control structure generally includes a trigger power circuit, a relay switching unit, an electromagnetic driver, and a linkage component for the starting mechanism. During the release process, some systems use mechanical pressure release devices or rupture discs to break the seal and use limit contacts or physical position detection elements to confirm the action status. The control process relies on unidirectional sequential logic, making it unable to make path judgments and feedback responses under multi-state conditions, and it lacks the ability to dynamically adjust control steps according to the fire extinguishing progress. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intelligent opening and closing control method for gas fire extinguishing equipment.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent opening and closing control method for a gas fire extinguishing device, comprising the following steps: S1: Collect infrared thermal images of the protected area, extract the temperature matrix and calculate the change, generate the regional temperature rise rate, extract the outline of the infrared thermal image, calculate the sealing degree and frequency, establish hot spot edge stability parameters, and generate start control commands based on the regional temperature rise rate and hot spot edge stability parameters. S2: Respond to the start control command to drive the valve, collect pressure readings and position signals, extract pressure and displacement increase values, and construct pressure change rate data and valve displacement rate data; S3: Cross-correlation between the pressure change rate data and the valve displacement rate data, identify the deviation position, construct the response lag time window, and calculate the drive signal correction amount; S4: Load the drive signal correction amount, obtain the pressure reading sequence, identify the discontinuous mode of the pressure reading sequence and locate the breakpoint, call the parameter group to perform voltage regulation or channel reconstruction according to the breakpoint, and generate the release logic reconstruction sequence. S5: Execute the release according to the release logic reconstruction sequence, collect temperature data, calculate the temperature decay slope of the temperature data, determine the fire extinguishing stability, and generate a valve closing control command.

[0006] As a further embodiment of the present invention, the start control command includes an initial drive voltage amplitude, an activation pulse duration, and an emergency trigger priority bit; the valve displacement rate data includes a valve core displacement velocity vector, a mechanical stroke acceleration factor, and a sampling point timestamp sequence; the drive signal correction includes a response lag time compensation value, a drive gain correction coefficient, and a signal phase synchronization offset; the release logic reconstruction sequence includes a dynamic pressure regulation target value, a release channel topology command, and an intermittent control duty cycle; and the valve closing control command includes a cutoff action timing parameter, a closing position check code, and a residual pressure release flag.

[0007] As a further aspect of the present invention, the step of obtaining the start control command specifically comprises: S101: Collect infrared thermal images within the protected area, analyze multiple frames of continuous image data within a preset time window, extract the corresponding two-dimensional temperature matrix from each frame, perform temporal difference operation on the temperature values ​​of each pixel coordinate at adjacent time nodes, calculate the cumulative temperature increment of each pixel area within a unit time, spatially map the pixel increment data, and construct the area temperature rise rate. S102: Call the infrared thermal image associated with the temperature rise rate of the region, use the gradient calculation method to identify the continuous region of temperature gradient distribution inside the image and extract the contour edge of the region, continuously monitor the geometric closure integrity of the contour edge in the continuous time series and the random jump frequency of the edge pixel position, perform weighted aggregation calculation on the closure integrity and jump frequency, and establish hot spot edge stability parameters. S103: Obtain the temperature rise rate of the region and the stability parameters of the hot spot edge, compare the value of the temperature rise rate of the region with the preset temperature rise rate alarm threshold, and compare the hot spot edge stability parameters with the preset morphology maintenance threshold. When both sets of parameters are determined to be abnormal states exceeding the safety limits, trigger the start decision and generate a start control command.

[0008] As a further aspect of the present invention, the setting process of the temperature rise rate alarm threshold is specifically as follows: based on the historical background temperature fluctuation data of the protected area under non-fire conditions, the statistical peak value of the fluctuation data is calculated and linearly superimposed with a preset safety margin coefficient to serve as the temperature rise rate alarm threshold. The process of setting the shape maintenance threshold is as follows: based on the edge contour change characteristics of real fire source samples in a continuous time series, the average offset and closure variance of edge pixel positions are statistically analyzed, and the lower limit of the confidence interval of the statistical results is used as the shape maintenance threshold.

[0009] As a further aspect of the present invention, the step of obtaining the valve displacement rate data specifically includes: S201: Based on the start control command, send a drive voltage to the actuator, and simultaneously monitor the status of the gas pipeline and valve. Record the instantaneous reading of the pressure sensor and the real-time position signal of the valve stroke sensor at a preset sampling frequency. Use a unified clock source to add time stamps to the collected readings and signals to generate a synchronous sampling sensor sequence. S202: Call the synchronous sampling sensor sequence, extract the pressure reading difference and stroke position difference corresponding to adjacent timestamps, calculate the gas pressure rise amplitude and valve core mechanical displacement distance within a unit time interval, map the calculated amplitude value and distance value to the corresponding time window, and generate the pressure increase value and valve displacement increase value per unit time respectively. S203: Based on the pressure increase value per unit time and the valve displacement increase value, perform normalized rate calculation, convert the pressure increase value into a pressure gradient sequence reflecting the dynamics of gas release, convert the valve displacement increase value into a displacement velocity sequence reflecting the mechanical action response, perform time-series formatting and encapsulation on the two sequences, and establish pressure change rate data and valve displacement rate data.

[0010] As a further aspect of the present invention, the step of obtaining the driving signal correction amount specifically includes: S301: Call the pressure change rate data and valve displacement rate data, perform cross-correlation operation on the discrete sequence, calculate the cross-correlation coefficient sequence within the preset sliding time shift range, locate the time shift step corresponding to the maximum value of the cross-correlation coefficient through peak retrieval logic, and generate the maximum deviation time shift amount; S302: Based on the maximum deviation time shift and the data sampling period, perform a product operation to convert the time shift step into a delay value with physical time dimensions, and combine it with the preset allowable error range boundary to define the response lag time window; S303: Obtain the response lag time window, call the valve displacement rate data to calculate the valve average speed in the current action stage, introduce the preset pipeline input pressure benchmark value, rated reference pressure value and valve nominal action speed value, and calculate to obtain the drive signal correction amount.

[0011] As a further aspect of the present invention, the step of obtaining the release logic reconstruction sequence specifically includes: S401: The drive signal correction amount is loaded into the valve controller to correct the timing of subsequent actions. During the gas release process, the pressure reading sequence is continuously acquired. The continuous characteristic of the pressure value monotonically decreasing with time in the sequence is scanned point by point. The instantaneous fluctuation variance of the release rate is calculated using a sliding window. The relationship between the variance and the preset stability threshold is compared. The time coordinate point where the variance value exceeds the limit is locked, and the breakpoint position representing the abnormal release moment is generated. S402: Extract the corresponding local pressure data segment based on the breakpoint location, calculate the pressure gradient mutation rate and high-frequency oscillation component within the segment, map the calculation results to the preset fault feature vector space for cluster matching, extract the dynamic pressure regulation coefficient or backup channel activation command corresponding to the matching result index, and establish an abnormal processing parameter group. S403: Call the voltage regulation coefficient in the abnormal handling parameter group to modulate the amplitude of the original valve control voltage, and use the backup channel activation command to modify the on / off state matrix of the multi-way valve. Reassemble and encapsulate the modulated voltage parameters and the modified on / off matrix according to the release sequence to generate a release logic reconstruction sequence.

[0012] As a further aspect of the present invention, the stability threshold is specifically set by selecting several sets of historical gas release pressure curves collected under standard fault-free operating conditions as reference samples, calculating the variance of the pressure change rate during the steady-state release stage for each sample, extracting the maximum value in the statistical distribution of variances of all samples as the baseline noise level, and multiplying the baseline noise level by a preset sensitivity redundancy coefficient as the stability threshold.

[0013] As a further aspect of the present invention, the step of obtaining the valve closing control command specifically includes: S501: Based on the release logic reconstruction sequence, dynamically adjust the opening of the gas release valve, simultaneously collect real-time temperature data through the temperature sensor array in the area, construct a time series of temperature changes over time, use the least squares method to perform linear fitting on the sequence, calculate the slope of the fitted line, quantify the temperature drop per unit time, and generate the temperature decay slope. S502: Call the temperature decay slope, calculate the standard deviation of the slope value fluctuation within a continuous sliding time window, and calculate the absolute value of the difference between the current slope value and the preset ideal fire extinguishing model benchmark slope. Perform a weighted aggregation operation on the standard deviation of fluctuation and the absolute value of the difference to establish a fire extinguishing effectiveness stability judgment value. S503: Obtain the fire extinguishing effectiveness stability judgment value and compare it with the preset fire extinguishing termination safety judgment threshold. If the fire extinguishing effectiveness stability judgment value exceeds the fire extinguishing termination safety judgment threshold, it is determined that the fire threat in the protected area has been eliminated, and the valve actuator reset and locking procedure is immediately triggered to generate a valve closing control command.

[0014] As a further aspect of the present invention, the process of setting the fire extinguishing termination safety judgment threshold is as follows: a benchmark database is constructed based on historical fire extinguishing experimental data or thermodynamic simulation models; critical time nodes in the state of complete fire source extinguishment are screened; the temperature decay slope fluctuation amplitude and model fitting deviation data corresponding to the nodes are extracted; statistical analysis is performed on the extracted dataset to determine its probability distribution interval in steady state; boundary values ​​representing high-confidence extinguishing states in the probability distribution interval are selected; and environmental thermal noise baseline parameters are superimposed as safety redundancy to construct the fire extinguishing termination safety judgment threshold.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, temperature rise rate and edge stability parameters are constructed using thermal imaging images to achieve multi-dimensional hot spot feature judgment and trigger control signals. By cross-comparison of valve displacement and pressure rate data, the action hysteresis time window is obtained and the drive signal is corrected, improving the valve response accuracy and timing coordination. During the release process, continuous time slice identification and abnormal fluctuation detection are used to locate breakpoints and reconstruct channels, ensuring the stability and continuity of the release process. Combined with temperature decay slope regression analysis, the fire extinguishing stability is judged, achieving control convergence when the fire extinguishing effect is achieved, improving the dynamic adaptability of the control process and the closed-loop response capability of the fire extinguishing operation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0017] Figure 1 This is a schematic diagram of the main steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] Please see Figure 1 This invention provides a technical solution: an intelligent start-stop control method for a gas fire extinguishing device, comprising the following steps: S1: Collect infrared thermal images of the protected area and extract the temperature matrix within the preset time window. Calculate the temperature change per unit time of each pixel area to generate the area temperature rise rate. Extract the contour of the temperature gradient area in the infrared thermal image. Calculate the edge closure degree and edge change frequency as hot spot edge stability parameters. Perform dual threshold judgment on the area temperature rise rate and hot spot edge stability parameters to generate start control command. S2: In response to the start control command, execute valve drive operation, synchronously collect the pressure sensor reading sequence and valve stroke sensor output signal in the gas pipeline, extract the pressure increase value and valve displacement increase value per unit time, construct pressure change rate data based on the pressure increase value, and construct valve displacement rate data based on the valve displacement increase value. S3: Call the pressure change rate data and valve displacement rate data, use the cross-correlation function to perform cross-sliding comparison analysis on the two sets of rate data, determine the maximum deviation position of the data curve to construct the response lag time window, and calculate the drive signal correction amount to offset the mechanical motion inertia based on the response lag time window. S4: The drive signal correction amount is applied to the valve controller to correct the timing of subsequent actions. During the gas release process, the pressure reading sequence is continuously acquired. The discontinuous pattern of the continuous drop time slice in the pressure reading sequence is identified. The release rate is detected to locate the breakpoint by detecting whether there is an unstable high fluctuation area in the continuous window. Based on the breakpoint, the preset abnormal handling parameter group is called to execute the pressure regulation command loading or release channel reconstruction operation, and a release logic reconstruction sequence containing adaptive adjustment logic is generated. S5: Perform segmented release operation according to the release logic reconstruction sequence, collect real-time temperature data in the protected area, input the real-time temperature data into the linear regression algorithm to calculate the temperature decay slope, perform fire extinguishing effectiveness stability judgment on the temperature decay slope, and generate valve closing control command when the judgment result meets the preset safety threshold.

[0024] The start control command includes the initial drive voltage amplitude, start pulse duration, and emergency trigger priority bit; the valve displacement rate data includes the valve core displacement velocity vector, mechanical stroke acceleration factor, and sampling point timestamp sequence; the drive signal correction includes the response lag time compensation value, drive gain correction coefficient, and signal phase synchronization offset; the release logic reconstruction sequence includes the dynamic pressure regulation target value, release channel topology command, and intermittent control duty cycle; and the valve closing control command includes the cut-off action timing parameters, closing position check code, and residual pressure release flag.

[0025] Please see Figure 2 The specific steps for obtaining the start control command are as follows: S101: Collect infrared thermal images within the protected area, analyze multiple frames of continuous image data within a preset time window, extract the corresponding two-dimensional temperature matrix from each frame, perform temporal difference operation on the temperature values ​​of each pixel coordinate at adjacent time nodes, calculate the cumulative temperature increment of each pixel area within a unit time, spatially map the pixel increment data, and construct a regional temperature rise rate that can digitally characterize the temperature rise rate of the regional thermal field. Relying on a highly reliable industrial control system as the core processing platform and combined with a distributed IoT control system for front-end data sensing, the process first activates the focal plane array detector of the infrared thermal imager, setting the sampling frequency to 25 Hz, i.e., acquiring 25 frames of thermal imaging data per second, with an image resolution of 640 x 512 pixels. For the acquired infrared thermal imaging video stream, a continuous 1-second time window is selected as the analysis object, containing 25 consecutive digital images. For each frame of the raw image, the image processing unit integrated within the industrial control system preprocesses the image using a thermal radiation non-uniformity correction algorithm. Then, the grayscale value of each pixel in the image is extracted and converted into a physically meaningful degree Celsius value according to the blackbody radiation calibration curve, thereby constructing a two-dimensional temperature matrix with a resolution of 640 x 512. Based on this, the process locks the coordinates of each pixel in the matrix; for example, the pixel with coordinates 320 rows and 256 columns, and extracts the time-series temperature data of that point from frame 1 to frame 25. The process performs temporal difference operations on adjacent frames, subtracting the temperature value of the first frame from the temperature value of the second frame, subtracting the temperature value of the second frame from the temperature value of the third frame, and so on, to obtain 24 instantaneous temperature changes for the pixel within a time window. Next, the process accumulates all positive increments among these 24 changes, ignoring negative or zero values ​​to filter out noise from natural cooling, thus calculating the cumulative temperature increment of the pixel within one second. For example, if a pixel experiences three 0.5-degree Celsius temperature increases between consecutive frames, while its temperature remains constant or decreases at other times, the cumulative temperature increment for that pixel is 1.5 degrees Celsius. The process repeats the above difference and accumulation operations on all 327,680 pixels of the image, generating a temperature rise rate mapping matrix with the same resolution as the original image. This matrix visually quantifies the rate of thermal rise in each tiny space within the region, completing the digital construction of the regional temperature rise rate.

[0026] S102: Call the infrared thermal image associated with the regional temperature rise rate, use the gradient calculation method to identify the continuous region of temperature gradient distribution inside the image and extract the contour edge of the region, continuously monitor the geometric closure integrity of the contour edge in the continuous time series and the random jump frequency of the edge pixel position, perform weighted aggregation calculation on the closure integrity and jump frequency, and establish hot spot edge stability parameters for quantitatively evaluating the spatiotemporal consistency of heat source morphology. For infrared images, a high-performance edge computing module of the industrial control system is used to perform convolution operations in the horizontal and vertical directions using the Sobel operator to calculate the temperature gradient amplitude of each pixel. The process sets a gradient amplitude screening threshold of 50, marking pixels with gradient amplitudes greater than 50 as edge candidates. Then, non-maximum suppression is applied to refine the edge lines, extracting the contours of high-temperature regions with continuous temperature gradient distributions. In continuous time-series images, the process continuously tracks the contour morphology at the same spatial location. For each identified hotspot contour, its geometric closure integrity is calculated. Specifically, the number of pixels constituting the contour is counted, and the Euclidean distance between the first and last pixels is detected. The ratio of this distance to the contour perimeter is used as the non-closure degree. The closure integrity value is obtained by subtracting the non-closure degree from 1, with the value ranging from 0 to 1. Simultaneously, the process monitors random jumps in the pixel positions of the contour edges. By calculating the displacement vectors of corresponding edge pixels between adjacent frames, the frequency of displacement distances exceeding 3 pixel units within 1 second is counted. For example, if a hotspot edge experiences five large positional jumps within one second, the jump frequency is recorded as 5. The process establishes the calculation logic for the hotspot edge stability parameters, setting the weighting coefficient for closure integrity to 0.6 and the weighting coefficient for the normalized jump frequency to 0.4. Specifically, the jump frequency is first normalized by dividing by the baseline frequency constant 10, then the normalized value is subtracted from 1 to obtain the position stability score. Finally, the closure integrity value and the position stability score are multiplied by their respective weights and added together. For example, if the closure integrity is 0.9, the jump frequency is 2 (normalized to 0.2), and the position stability score is 0.8, the weighted calculation result is 0.9 multiplied by 0.6 plus 0.8 multiplied by 0.4, which equals 0.54 plus 0.32, resulting in a final hotspot edge stability parameter of 0.86.

[0027] S103: Acquire the area temperature rise rate and hot spot edge stability parameters, compare the value of the area temperature rise rate with the preset temperature rise rate alarm threshold, and compare the hot spot edge stability parameters with the preset shape maintenance threshold. When both sets of parameters are determined to be abnormal states exceeding the safety limits, trigger the start decision and generate a start control command containing the action sequence information of the fire extinguishing actuator.

[0028] The specific process of setting the temperature rise rate alarm threshold is as follows: based on the historical background temperature fluctuation data of the protected area under non-fire conditions, the statistical peak value of the fluctuation data is calculated and linearly superimposed with the preset safety margin coefficient to serve as the temperature rise rate alarm threshold. The specific process of setting the shape maintenance threshold is as follows: based on the edge contour change characteristics of real fire source samples in a continuous time series, the average offset and closure variance of edge pixel positions are statistically analyzed, and the lower limit of the confidence interval of the statistical results is used as the shape maintenance threshold. The preset temperature rise rate alarm threshold and shape maintenance threshold are retrieved from the central storage unit of the industrial control system. The temperature rise rate alarm threshold is set based on historical background temperature fluctuation data of the protected area under non-fire conditions over the past 30 days. The maximum temperature rise rate per second for all pixels during this period is statistically analyzed, with an average value of 1.2 degrees Celsius per second and a standard deviation of 0.3 degrees Celsius per second. The average value plus three times the standard deviation is selected as the statistical peak value, i.e., 2.1 degrees Celsius per second. A preset safety margin factor of 0.4 degrees Celsius per second is then added, ultimately determining the temperature rise rate alarm threshold to be 2.5 degrees Celsius per second. The shape maintenance threshold is set based on edge contour data from 100 sets of real fire source samples at the initial stage of fire. The average edge offset and closure variance are statistically analyzed, and a lower limit of 0.75 for the 95% confidence interval is determined through confidence analysis. This lower limit is used as the shape maintenance threshold. In real-time monitoring, analysis is performed by the logic judgment unit of the industrial control system. If the current temperature rise rate is 3.0 degrees Celsius per second and the hot spot edge stability parameter is 0.65, a comparison of process execution values ​​reveals that 3.0 is greater than 2.5 and 0.65 is less than 0.75, indicating that both sets of parameters are in an abnormal state. At this time, the system immediately triggers the start-up decision logic, generating a start-up control command containing the specific action sequence of the fire extinguishing actuators. This command clearly specifies the opening sequence and interval time of valves 1 and 2, and sends the command to the communication gateway of the underlying control unit and the IoT control system to ensure coordinated response of the entire system. Table 1 shows the key statistical data in the threshold setting process.

[0029] Table 1. Threshold Setting Reference Data Table

[0030] Please see Figure 3 The specific steps for obtaining valve displacement rate data are as follows: S201: Based on the start control command, send the drive voltage to the actuator, and simultaneously monitor the status of the gas pipeline and valve. Record the instantaneous reading of the pressure sensor and the real-time position signal of the valve stroke sensor at a preset sampling frequency. Use a unified clock source to add time stamps to the collected readings and signals, and generate a synchronous sampling sensor sequence containing timestamp index. The industrial control system's output module sends a 24-volt DC drive voltage to the solenoid coil of the valve actuator, driving the valve to open. Simultaneously, high-frequency monitoring of the gas pipeline and valve status is initiated, with a sampling frequency set to 1000 Hz. This monitoring process relies on the perception layer architecture of the IoT control system, recording in real-time the instantaneous readings of pressure sensors installed on the pipeline (range 0-20 MPa) and the position signal from the valve stroke sensor (range 0-50 mm). The system utilizes a high-precision crystal oscillator within the industrial control system as a unified clock source, assigning microsecond-level time stamps to each acquired pressure reading and position signal. For example, at system timestamp 1000 milliseconds, a pressure value of 5.00 MPa and a valve position of 0.0 mm are acquired; at 1001 milliseconds, the pressure value is acquired at 5.01 MPa and the valve position at 0.1 mm. This data, with its strict time correspondence, is continuously written to the industrial control system's cache, generating a synchronized sampling sensor sequence containing timestamp indices to ensure strict alignment of pressure changes and mechanical actions on the timeline during subsequent analysis.

[0031] S202: Call the synchronous sampling sensor sequence, extract the pressure reading difference and stroke position difference corresponding to adjacent timestamps, calculate the gas pressure rise amplitude and valve core mechanical displacement distance within a unit time interval, map the calculated amplitude value and distance value to the corresponding time window, and generate the unit time pressure increase value and valve displacement increase value to characterize the transient change characteristics respectively. The analysis time interval is set to 10 milliseconds, meaning that data from the current moment and the moment 10 milliseconds ago are extracted and differentially calculated. This data processing task is undertaken by the industrial control system's processing unit. The process reads the pressure reading corresponding to the current timestamp as 5.50 MPa and the pressure reading 10 milliseconds ago as 5.00 MPa, performing a subtraction operation to obtain a pressure difference of 0.50 MPa. Similarly, it reads the current valve stroke position as 2.0 mm and the position 10 milliseconds ago as 0.5 mm, calculating a stroke position difference of 1.5 mm. The process maps this 0.50 MPa pressure increment and 1.5 mm displacement increment back to the current 10-millisecond time window, generating the unit time pressure increase value and valve displacement increase value within this time period. This operation is performed sliding across the entire sampling sequence, thereby converting the original cumulative data into an incremental data sequence characterizing transient changes, accurately reflecting the system's dynamic response characteristics within a very short time.

[0032] S203: Perform normalized rate calculation based on the pressure increase value and valve displacement increase value per unit time, convert the pressure increase value into a pressure gradient sequence reflecting the dynamics of gas release, and convert the valve displacement increase value into a displacement velocity sequence reflecting the mechanical action response. Perform time-series formatting and encapsulation on the two sequences to establish pressure change rate data and valve displacement rate data.

[0033] Dividing the pressure increase of 0.50 MPa by the time interval of 0.01 seconds yields a pressure change rate of 50 MPa per second, which is then stored in the pressure gradient sequence. Similarly, dividing the displacement increase of 1.5 mm by the time interval of 0.01 seconds yields a valve displacement velocity of 150 mm per second, which is stored in the displacement velocity sequence. This division operation is performed on all data points throughout the monitoring process, converting the increment of the physical quantity into a rate physical quantity. Subsequently, the generated pressure gradient sequence and displacement velocity sequence are time-series formatted and encapsulated to ensure that the two sequences have the same length and strictly correspond on the time axis. This establishes the pressure change rate data and valve displacement rate data for subsequent correlation analysis, characterizing the energy burst rate of the extinguishing gas release and the response sensitivity of the mechanical actuator, respectively.

[0034] Please see Figure 4 The specific steps for obtaining the drive signal correction amount are as follows: S301: Call the pressure change rate data and valve displacement rate data, perform cross-correlation calculation of discrete sequences, calculate the cross-correlation coefficient sequence within the preset sliding time shift range, locate the time shift step corresponding to the maximum value of the cross-correlation coefficient through peak retrieval logic, and generate the maximum deviation time shift amount; Executed by a dedicated digital signal processing unit in the industrial control system, the sliding time shift range is set to 100 sampling points forward and backward, corresponding to a time window of 0.1 seconds. The process fixes the valve displacement rate sequence and translates the pressure change rate sequence point-by-point along the time axis. At each translation step, the sum of the dot products of the overlapping portions of the two sequences is calculated to obtain the cross-correlation coefficient for that step. The process iterates through all set translation steps, generating a cross-correlation coefficient sequence. Subsequently, peak retrieval logic scans this coefficient sequence to find the location of the maximum value. For example, the retrieval reveals that the cross-correlation coefficient reaches its maximum value of 0.95 at a time shift step of 50 sampling points, indicating that the waveform of the pressure data best matches the valve displacement data after shifting it backward by 50 points. The process locks this time shift step, generating a maximum deviation of 50 points.

[0035] S302: Based on the product operation of the maximum deviation time shift and the data sampling period, the time shift step is converted into a delay value with physical time dimension. Combined with the preset allowable error range boundary, the response lag time window used to cover the mechanical action response delay range is defined. The maximum deviation time shift of 50 points is obtained and multiplied with the data sampling period of 0.001 seconds in the industrial control system. Specifically, 50 is multiplied by 0.001 seconds, converting the dimensionless step size into a physically meaningful delay value of 0.05 seconds. This value represents the physical lag time between the valve's initiation of action and a significant change in gas pressure. Next, a preset allowable error range boundary is introduced, such as ±0.005 seconds. Based on the calculated 0.05-second delay value, the lower bound of the response lag time window is determined to be 0.045 seconds, and the upper bound to be 0.055 seconds. This range is defined as the response lag time window to cover the mechanical action response delay range, precisely delineating the critical transition stage from the issuance of the electrical signal to the establishment of the fluid dynamic response.

[0036] S303: Obtain the response lag time window and call the valve displacement rate data to calculate the valve's average speed during the current action phase. It incorporates the preset pipeline input pressure benchmark value, rated reference pressure value, and valve nominal action speed value, using the following formula: ; The correction amount of the drive signal is obtained through calculation; in, This represents the amount of correction to the drive signal. This represents the duration of the response lag window, obtained by extracting values ​​from the response lag window generated in the preceding steps. The basic compensation bias coefficient is obtained by reading a preset empirical constant. The reference value for the input pressure of the pipeline network is obtained by collecting the initial static readings of the gas pipeline pressure sensor. This represents the system's rated reference pressure value, obtained by reading the rated working pressure parameter set at the factory of the equipment. This represents the average valve velocity, obtained by performing an arithmetic mean calculation on the valve displacement rate data. This represents the valve's nominal operating speed, which is obtained by reading the technical specifications of the valve actuator. The process retrieves valve displacement rate data, calculates the average valve velocity within the time window, and sets the average velocity obtained through arithmetic averaging to 100 mm / s. The process reads a pre-stored empirical constant, the basic compensation bias coefficient, set to 0.2. By acquiring the initial static reading of the gas pipeline pressure sensor in the IoT control system when the system is stationary, the pipeline input pressure reference value is obtained as 12 MPa. The rated working pressure parameters set by the equipment at the factory are read, obtaining the system rated reference pressure value as 15 MPa. The technical specifications of the valve actuator are read, obtaining the valve's nominal operating speed value as 150 mm / s. Based on preset logic, the process uses the floating-point arithmetic unit of the industrial control system to perform calculations on the drive signal correction: first, it calculates the ratio of the pipeline input pressure reference value of 12 MPa to the system rated reference pressure value of 15 MPa, i.e., 12 divided by 15 equals 0.8; then, it performs a square root operation on this ratio of 0.8, obtaining the square root of the pressure ratio, which is approximately 0.894. Simultaneously, the ratio of the valve's average velocity of 100 mm / s to its nominal operating velocity of 150 mm / s is calculated, i.e., 100 divided by 150 is approximately 0.667. Adding 1 to this ratio gives 1.667, and calculating its natural logarithm yields approximately 0.511. Then, the square root of the pressure ratio (0.894) is multiplied by the velocity logarithm (0.511), resulting in 0.457. This product is added to the basic compensation bias coefficient (0.2), yielding the comprehensive coefficient (0.657). Finally, the response lag time window length (0.05 seconds) is multiplied by the comprehensive coefficient (0.657), calculating the drive signal correction to be 0.03285 seconds, or 32.85 milliseconds. This correction will be used to advance or delay subsequent control signals to compensate for physical delays. Table 2 lists the key parameters and values ​​involved in this calculation step.

[0037] Table 2 Calculation Parameters for Drive Signal Correction

[0038] Please see Figure 5 The specific steps for obtaining the release logic reconstruction sequence are as follows: S401: The drive signal correction amount is applied to the valve controller to correct the timing of subsequent actions. During the gas release process, the pressure reading sequence is continuously acquired. The continuous characteristic of the pressure value monotonically decreasing with time in the sequence is scanned point by point. The instantaneous fluctuation variance of the release rate is calculated using a sliding window. The relationship between the variance and the preset stability threshold is compared. The time coordinate point where the variance value exceeds the limit is locked, and the breakpoint position representing the abnormal release moment is generated. The stability threshold is set by selecting several sets of historical gas release pressure curves collected under standard fault-free operating conditions as reference samples, calculating the variance of the pressure change rate in the steady-state release stage for each sample, extracting the maximum value in the statistical distribution of variance of all samples as the baseline noise level, and multiplying the baseline noise level by the preset sensitivity redundancy coefficient as the stability threshold. The drive signal correction is applied to the valve controller of the industrial control system to provide feedforward compensation for the original valve opening command timing, for example, advancing the opening time by 32.85 milliseconds. During gas release, the process continuously acquires pressure reading sequences through the real-time data link of the IoT control system, scanning the continuous characteristic of the monotonically decreasing pressure value over time point by point. Using a sliding window with a width of 50 milliseconds, the instantaneous fluctuation variance of the pressure release rate within the window is calculated. The process sets 20 sets of historical gas release pressure curves under standard fault-free operating conditions as reference samples, calculates the variance of the pressure change rate of these samples in the steady-state release phase, extracts the maximum value of its statistical distribution as 0.02, and multiplies this maximum value by a preset sensitivity redundancy coefficient of 1.5, setting the stability threshold to 0.03. In real-time monitoring, if the industrial control system calculates an instantaneous fluctuation variance of 0.045 at a certain moment, and the process compares and finds that 0.045 is greater than 0.03, it is determined that the variance value at that moment exceeds the limit. The system locks the time coordinate point and generates the breakpoint position representing the moment of abnormal release, indicating that unexpected turbulence or stagnation occurred in the gas flow at this time.

[0039] S402: Extract the corresponding local pressure data segment based on the breakpoint location, calculate the pressure gradient mutation rate and high-frequency oscillation component within the segment, map the calculation results to the preset fault feature vector space for cluster matching, extract the dynamic pressure regulation coefficient or backup channel activation command corresponding to the matching result index, and establish an abnormal handling parameter group for correcting the release state. Local pressure data segments of 100 milliseconds before and after the location were extracted. Numerical differentiation was performed on the segment data to obtain the pressure gradient, and the abrupt change rate of the gradient was calculated, with the result set to 5% per millisecond. Simultaneously, a Fast Fourier Transform was performed on the segment data to extract the energy proportion of the high-frequency oscillation component, with the result set to 15%. These two feature values ​​were combined to form a feature vector 5,15, which was mapped to a pre-defined fault feature vector space in the industrial control system database. This space pre-stores cluster centers for various fault modes, such as centers 4,14 for "nozzle blockage" and centers 1,2 for "pipeline leakage." The Euclidean distance between the real-time feature vector and each cluster center was calculated, finding that the distance to the "nozzle blockage" center was the closest. Based on this matching result, the system extracted the dynamic pressure regulation coefficient 1.1 corresponding to the fault index and the backup channel activation command "open backup valve 3". These data together established an abnormal handling parameter set for correcting the release state. Table 3 lists some reference data for fault feature cluster matching.

[0040] Table 3 Fault Feature Clustering Matching Reference Table

[0041] S403: Call the voltage regulation coefficient in the exception handling parameter group to modulate the amplitude of the original valve control voltage, and use the backup channel activation instruction to modify the on / off state matrix of the multi-way valve. Reassemble and encapsulate the modulated voltage parameters and the modified on / off matrix according to the release sequence to generate a release logic reconstruction sequence containing adaptive adjustment logic. The industrial control system extracts a voltage regulation coefficient of 1.1 and modulates the original valve control voltage of 24 volts. A multiplication operation is then performed to obtain a new control voltage of 26.4 volts, enhancing the valve's driving force to overcome blockage resistance. Next, the on / off state matrix of the multi-valve system is modified using a backup channel activation command, changing the element representing backup valve number 3 from "0" (closed) to "1" (open). The process then reassembles and encapsulates the modulated voltage parameter of 26.4 volts and the modified on / off matrix according to the original release sequence, generating a release logic reconstruction sequence containing adaptive adjustment logic. Once generated, this sequence is immediately sent to the underlying execution unit in the IoT control system, which then drives the main valve with the new high voltage and simultaneously opens the backup valve, achieving real-time correction of abnormal release states.

[0042] Please see Figure 6 The specific steps for obtaining the valve closing control command are as follows: S501: Based on the release logic reconstruction sequence, dynamically adjust the opening of the gas release valve, simultaneously collect real-time temperature data through the temperature sensor array in the area, construct a time series of temperature changes over time, use the least squares method to perform linear fitting on the sequence, calculate the slope of the fitted line, quantify the temperature drop per unit time, and generate the temperature decay slope. Real-time temperature data is collected by an IoT control system temperature sensor array deployed within the protected area, with a sampling interval of 0.5 seconds. A time series of temperature changes over time is constructed; for example, the collected sequence might be: temperature 200 degrees Celsius at time 0 seconds, 190 degrees Celsius at time 0.5 seconds, and 180 degrees Celsius at time 1.0 seconds. Utilizing the computing power of the industrial control system, the least squares method is used to perform linear fitting on this sequence, aiming to find a straight line equation that minimizes the sum of the squared distances from all measured temperature points to this line. The slope of the fitted line is calculated to be -20. The absolute value of this slope, 20, is quantified as the temperature decrease rate per unit time, generating a temperature decay slope, which directly reflects the current fire suppression and cooling efficiency.

[0043] S502: Call the temperature decay slope, calculate the standard deviation of the slope value fluctuation within a continuous sliding time window, and calculate the absolute value of the difference between the current slope value and the preset ideal fire extinguishing model benchmark slope. Perform a weighted aggregation operation on the standard deviation of fluctuation and the absolute value of the difference to establish a fire extinguishing effectiveness stability judgment value for quantifying the thermodynamic trend stability of the fire extinguishing process. Collect slope data within a continuous sliding time window of 5 seconds, for example, collect a set of slope values. The process calculates the standard deviation of the slope values, resulting in 0.63. Simultaneously, a preset ideal fire extinguishing model with a baseline slope of -25 is retrieved from the industrial control system's pre-set model library. The absolute value of the difference between the current average slope -20 and the baseline slope -25 is calculated, which is 5. The process establishes the calculation logic for the fire extinguishing effectiveness stability judgment value, setting the stability judgment value to equal the baseline constant 100 minus the weighted sum of the standard deviation and the absolute value of the difference. The standard deviation weight is set to 10, and the difference weight to 5. Specifically, the calculation is: 100 minus (0.63 multiplied by 10 plus 5 multiplied by 5), i.e., 100 minus (6.3 plus 25), resulting in 68.7. This value of 68.7 is the established fire extinguishing effectiveness stability judgment value, used to quantify the comprehensive stability of the thermodynamic trend of the fire extinguishing process. A higher value indicates that the fire extinguishing process is closer to an ideal and stable state.

[0044] S503: Obtain the stability judgment value of fire extinguishing effectiveness and compare it with the preset fire extinguishing termination safety judgment threshold. If the stability judgment value of fire extinguishing effectiveness exceeds the fire extinguishing termination safety judgment threshold, it is determined that the fire threat in the protected area has been eliminated, and the valve actuator reset and locking procedure is immediately triggered to generate a valve closing control command containing mechanical shut-off action timing parameters. The specific process of setting the fire extinguishing termination safety judgment threshold is as follows: a benchmark database is constructed based on historical fire extinguishing experimental data or thermodynamic simulation models; critical time nodes in the state of complete fire source extinguishment are screened; the temperature decay slope fluctuation amplitude and model fitting deviation data corresponding to the nodes are extracted; statistical analysis is performed on the extracted dataset to determine its probability distribution interval in steady state; boundary values ​​representing high confidence extinguishing state in the probability distribution interval are selected; and environmental thermal noise baseline parameters are superimposed as safety redundancy to construct the fire extinguishing termination safety judgment threshold. The preset fire extinguishing termination safety threshold is retrieved. This threshold is set based on historical fire extinguishing experiment data, selecting critical time node data under the condition of complete fire source extinguishment. The model fitting deviation and slope fluctuation characteristics corresponding to these nodes are extracted. Statistical analysis determines that the probability distribution range of the above stability judgment value under steady-state extinguishing conditions is 85% to 95%. The high confidence boundary value of 85 in this range is selected, and the environmental thermal noise baseline parameter 5 is superimposed as a safety redundancy to construct a fire extinguishing termination safety judgment threshold of 90. In the current logical comparison, the decision core of the industrial control system compares the calculated judgment value of 68.7 with the threshold of 90. Since 68.7 is less than 90, the system determines that the current fire extinguishing effect has not yet reached the safety standard of complete extinguishing and needs to continue to maintain the release. It is set that after a period of continuous suppression, the judgment value rises to 92. At this time, 92 exceeds the threshold of 90, and the process determines that the fire threat in the protected area has been completely eliminated. The system then triggers the start decision, generates control commands including valve closing sequence, the actuator resets and locks, and ends the fire extinguishing process. This logic ensures that the injection stops only when the thermodynamic characteristics are extremely stable and close to the ideal extinction model, thus preventing reignition.

[0045] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0046] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0047] It should be understood that, in various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0048] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0049] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0050] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0051] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0052] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0053] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of protection of the described technical solutions.

Claims

1. A method for intelligent opening and closing control of a gas fire extinguishing device, characterized in that, Includes the following steps: S1: Collect infrared thermal images of the protected area, extract the temperature matrix and calculate the change, generate the regional temperature rise rate, extract the outline of the infrared thermal image, calculate the sealing degree and frequency, establish hot spot edge stability parameters, and generate start control commands based on the regional temperature rise rate and hot spot edge stability parameters. S2: Respond to the start control command to drive the valve, collect pressure readings and position signals, extract pressure and displacement increase values, and construct pressure change rate data and valve displacement rate data; S3: Cross-correlation between the pressure change rate data and the valve displacement rate data, identify the deviation position, construct the response lag time window, and calculate the drive signal correction amount; S4: Load the drive signal correction amount, obtain the pressure reading sequence, identify the discontinuous mode of the pressure reading sequence and locate the breakpoint, call the parameter group to perform voltage regulation or channel reconstruction according to the breakpoint, and generate the release logic reconstruction sequence. S5: Execute the release according to the release logic reconstruction sequence, collect temperature data, calculate the temperature decay slope of the temperature data, determine the fire extinguishing stability, and generate a valve closing control command.

2. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 1, characterized in that, The start control command includes the initial drive voltage amplitude, the start pulse duration, and the emergency trigger priority bit. The valve displacement rate data includes the valve core displacement velocity vector, the mechanical stroke acceleration factor, and the sampling point timestamp sequence. The drive signal correction includes the response lag time compensation value, the drive gain correction coefficient, and the signal phase synchronization offset. The release logic reconstruction sequence includes the dynamic pressure regulation target value, the release channel topology command, and the intermittent control duty cycle. The valve closing control command includes the cut-off action timing parameters, the closing position check code, and the remaining pressure release flag.

3. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 2, characterized in that, The specific steps for obtaining the startup control command are as follows: S101: Collect infrared thermal images within the protected area, analyze multiple frames of continuous image data within a preset time window, extract the corresponding two-dimensional temperature matrix from each frame, perform temporal difference operation on the temperature values ​​of each pixel coordinate at adjacent time nodes, calculate the cumulative temperature increment of each pixel area within a unit time, spatially map the pixel increment data, and construct the area temperature rise rate. S102: Call the infrared thermal image associated with the temperature rise rate of the region, use the gradient calculation method to identify the continuous region of temperature gradient distribution inside the image and extract the contour edge of the region, continuously monitor the geometric closure integrity of the contour edge in the continuous time series and the random jump frequency of the edge pixel position, perform weighted aggregation calculation on the closure integrity and jump frequency, and establish hot spot edge stability parameters. S103: Obtain the temperature rise rate of the region and the stability parameters of the hot spot edge, compare the value of the temperature rise rate of the region with the preset temperature rise rate alarm threshold, and compare the hot spot edge stability parameters with the preset morphology maintenance threshold. When both sets of parameters are determined to be abnormal states exceeding the safety limits, trigger the start decision and generate a start control command.

4. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 3, characterized in that, The specific process for setting the temperature rise rate alarm threshold is as follows: based on the historical background temperature fluctuation data of the protected area under non-fire conditions, the statistical peak value of the fluctuation data is calculated and linearly superimposed with the preset safety margin coefficient to serve as the temperature rise rate alarm threshold. The process of setting the shape maintenance threshold is as follows: based on the edge contour change characteristics of real fire source samples in a continuous time series, the average offset and closure variance of edge pixel positions are statistically analyzed, and the lower limit of the confidence interval of the statistical results is used as the shape maintenance threshold.

5. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 4, characterized in that, The specific steps for obtaining the valve displacement rate data are as follows: S201: Based on the start control command, send a drive voltage to the actuator, and simultaneously monitor the status of the gas pipeline and valve. Record the instantaneous reading of the pressure sensor and the real-time position signal of the valve stroke sensor at a preset sampling frequency. Use a unified clock source to add time stamps to the collected readings and signals to generate a synchronous sampling sensor sequence. S202: Call the synchronous sampling sensor sequence, extract the pressure reading difference and stroke position difference corresponding to adjacent timestamps, calculate the gas pressure rise amplitude and valve core mechanical displacement distance within a unit time interval, map the calculated amplitude value and distance value to the corresponding time window, and generate the pressure increase value and valve displacement increase value per unit time respectively. S203: Based on the pressure increase value per unit time and the valve displacement increase value, perform normalized rate calculation, convert the pressure increase value into a pressure gradient sequence reflecting the dynamics of gas release, convert the valve displacement increase value into a displacement velocity sequence reflecting the mechanical action response, perform time-series formatting and encapsulation on the two sequences, and establish pressure change rate data and valve displacement rate data.

6. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 5, characterized in that, The specific steps for obtaining the drive signal correction amount are as follows: S301: Call the pressure change rate data and valve displacement rate data, perform cross-correlation operation on the discrete sequence, calculate the cross-correlation coefficient sequence within the preset sliding time shift range, locate the time shift step corresponding to the maximum value of the cross-correlation coefficient through peak retrieval logic, and generate the maximum deviation time shift amount; S302: Based on the maximum deviation time shift and the data sampling period, perform a product operation to convert the time shift step into a delay value with physical time dimensions, and combine it with the preset allowable error range boundary to define the response lag time window; S303: Obtain the response lag time window, call the valve displacement rate data to calculate the valve average speed in the current action stage, introduce the preset pipeline input pressure benchmark value, rated reference pressure value and valve nominal action speed value, and calculate to obtain the drive signal correction amount.

7. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 6, characterized in that, The specific steps for obtaining the release logic reconstruction sequence are as follows: S401: The drive signal correction amount is loaded into the valve controller to correct the timing of subsequent actions. During the gas release process, the pressure reading sequence is continuously acquired. The continuous characteristic of the pressure value monotonically decreasing with time in the sequence is scanned point by point. The instantaneous fluctuation variance of the release rate is calculated using a sliding window. The relationship between the variance and the preset stability threshold is compared. The time coordinate point where the variance value exceeds the limit is locked, and the breakpoint position representing the abnormal release moment is generated. S402: Extract the corresponding local pressure data segment based on the breakpoint location, calculate the pressure gradient mutation rate and high-frequency oscillation component within the segment, map the calculation results to the preset fault feature vector space for cluster matching, extract the dynamic pressure regulation coefficient or backup channel activation command corresponding to the matching result index, and establish an abnormal processing parameter group. S403: Call the voltage regulation coefficient in the abnormal handling parameter group to modulate the amplitude of the original valve control voltage, and use the backup channel activation command to modify the on / off state matrix of the multi-way valve. Reassemble and encapsulate the modulated voltage parameters and the modified on / off matrix according to the release sequence to generate a release logic reconstruction sequence.

8. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 7, characterized in that, The stability threshold is set by selecting several sets of historical gas release pressure curves collected under standard fault-free operating conditions as reference samples, calculating the variance of the pressure change rate during the steady-state release stage for each sample, extracting the maximum value in the statistical distribution of variance of all samples as the baseline noise level, and multiplying the baseline noise level by a preset sensitivity redundancy coefficient as the stability threshold.

9. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 8, characterized in that, The specific steps for obtaining the valve closing control command are as follows: S501: Based on the release logic reconstruction sequence, dynamically adjust the opening of the gas release valve, simultaneously collect real-time temperature data through the temperature sensor array in the area, construct a time series of temperature changes over time, use the least squares method to perform linear fitting on the sequence, calculate the slope of the fitted line, quantify the temperature drop per unit time, and generate the temperature decay slope. S502: Call the temperature decay slope, calculate the standard deviation of the slope value fluctuation within a continuous sliding time window, and calculate the absolute value of the difference between the current slope value and the preset ideal fire extinguishing model benchmark slope. Perform a weighted aggregation operation on the standard deviation of fluctuation and the absolute value of the difference to establish a fire extinguishing effectiveness stability judgment value. S503: Obtain the fire extinguishing effectiveness stability judgment value and compare it with the preset fire extinguishing termination safety judgment threshold. If the fire extinguishing effectiveness stability judgment value exceeds the fire extinguishing termination safety judgment threshold, it is determined that the fire threat in the protected area has been eliminated, and the valve actuator reset and locking procedure is immediately triggered to generate a valve closing control command.

10. The intelligent start-stop control method for gas fire extinguishing equipment according to claim 9, characterized in that, The process of setting the fire extinguishing termination safety judgment threshold is as follows: a benchmark database is constructed based on historical fire extinguishing experimental data or thermodynamic simulation models; critical time nodes in the state of complete fire source extinguishment are screened; the temperature decay slope fluctuation amplitude and model fitting deviation data corresponding to the nodes are extracted; statistical analysis is performed on the extracted dataset to determine its probability distribution interval in steady state; boundary values ​​representing high confidence extinguishing state in the probability distribution interval are selected; and environmental thermal noise baseline parameters are superimposed as safety redundancy to construct the fire extinguishing termination safety judgment threshold.

Images