Mechanical pump-in foam extinguishing agent proportioning system and method

Abstract

The application relates to the technical field of foam extinguishing agent mixing, and discloses a mechanical pump-in type foam extinguishing agent proportion mixing system and method, wherein the mechanical pump-in type foam extinguishing agent proportion mixing system comprises a main water flow pipeline, a water flow flow sensor is arranged on the main water flow pipeline, an original liquid storage tank, an electric original liquid cut-off valve, a servo metering pump and a foam flow meter are sequentially arranged on a foam original liquid supply pipeline, and a static mixer is connected to the outlets of the main water flow pipeline and the foam original liquid supply pipeline. Through concentration online detection and closed-loop feedback, the application directly monitors the concentration of the final mixed liquid, effectively avoids fire extinguishing failure caused by original liquid deterioration, dilution error or temperature drift, supports dynamic optimization of mixing ratio, proportioning according to demand, supports qualified mixed liquid recycling and reuse, and through dry rotation protection function, the pump can be automatically stopped in a liquid-free state, the metering pump is prevented from burning, and crystallization blockage is eliminated through automatic flushing.

Description

Technical Field

[0001] This invention relates to the field of foam extinguishing agent mixing technology, specifically to a mechanically pumped foam extinguishing agent proportioning mixing system and method. Background Technology

[0002] The foam proportioner is the nerve center of a foam fire extinguishing system, and its performance directly affects the success or failure of the system in extinguishing fires. Foam proportioners include pressure-type air foam proportioners, ring pump-type foam proportioners, and inline foam proportioners. Common foam proportioning systems mainly include balanced pressure type, ring pump type, and mechanical pump-in type. Mechanical pump-in type systems inject foam concentrate into the water flow through a metering pump, which has the advantages of being unaffected by water pressure fluctuations and having stable proportions. They are widely used in high-risk locations such as petrochemical plants, airports, and ships.

[0003] Existing mechanical pump-in systems suffer from insufficient mixing accuracy, mostly employ open-loop control, and cannot compensate in real time for mixing ratio deviations caused by flow fluctuations, temperature changes, or pipeline resistance. They also have low reliability, typically due to the foam concentrate easily crystallizing and clogging the pipeline, the pump being prone to damage from long-term dry running, and significant start-up delays, which affect initial fire extinguishing efficiency. Furthermore, they lack remote monitoring, fault warning, or adaptive adjustment capabilities. Summary of the Invention

[0004] This invention provides a mechanically pumped foam extinguishing agent proportioning system and method, which has the advantages of accurate proportioning and high reliability.

[0005] This invention provides the following technical solution: a mechanically pumped foam extinguishing agent proportioning system, comprising: The main water flow pipeline is equipped with a water flow sensor. A foam raw material supply pipeline is provided with a raw material storage tank, an electric raw material shut-off valve, a servo metering pump and a foam flow meter in sequence. A static mixer, which is connected to the outlet of the main water flow pipeline and the foam concentrate supply pipeline, is used to generate foam concentrate; An online concentration detection device is installed on one side of the static mixer, and the online concentration detection device is used to measure the actual concentration of foam stock solution in the foam mixture in real time. The controller is electrically connected to the water flow sensor, foam flow meter, servo metering pump, electric raw material shut-off valve and concentration online detection device respectively. The controller is used to calculate the target foam flow rate based on the dynamically optimized mixing ratio and the water flow rate measured by the water flow sensor. Drive the servo metering pump to operate at the target foam flow rate; The flow rate is adjusted in a closed loop based on the feedback from the foam flow meter, and the concentration is adjusted in a closed loop based on the actual concentration measured by the online concentration detection device, thereby dynamically correcting the operating parameters of the servo metering pump.

[0006] As an optional embodiment of the mechanically pumped foam extinguishing agent proportioning system described in this invention, it further includes: The AI ​​optimization module is used to acquire water flow parameters, foam parameters, fire type information, and fire extinguishing effect evaluation data from historical fire extinguishing events. A mixture ratio optimization model is trained based on the data; In a new fire extinguishing event, based on the current fire situation information and real-time operating conditions, the dynamic optimal mixing ratio is output as the basis for the controller to calculate the target foam flow rate; The fire type information is obtained through external sensing devices; The AI ​​optimization module adjusts the dynamic optimal mixing ratio based on the type of fire identified by the external sensing device.

[0007] As an optional embodiment of the mechanically pumped foam extinguishing agent proportioning system of the present invention, the foam flow rate is calculated as follows: in, , where is the target foam concentrate flow rate, and is the desired foam concentrate flow rate output by the metering pump from the controller; The dynamic optimal mixing ratio; The fire water flow rate is measured in real time by a water flow sensor to accurately measure the main water flow rate. The closed-loop regulation includes an inner loop flow closed-loop and an outer loop concentration closed-loop; The inner loop traffic closed loop includes: in, This is the pump speed correction amount, used to fine-tune the servo pump speed. This is the measured value from the foam flow meter. , and The proportional, integral, and derivative gain coefficients of the flow loop are obtained through on-site tuning. This is the cumulative error amount, used to eliminate steady-state deviation. This is the rate of change of error, used to suppress overshoot; The outer ring concentration closed loop includes: in, This is the concentration correction amount. To achieve the target foam concentration, For the concentration measured by the concentration sensor, , These are the PID parameters for the concentration loop; based on , and Calculate the total pump speed command synthesis: Total pump speed command synthesis = ( )+ + ...

[0008] As an optional embodiment of the mechanically pumped foam extinguishing agent proportioning system described in this invention, it further includes: An automatic rinsing and raw liquid recovery module, comprising a rinsing water valve, a three-way switching valve, and a recovery tank; The controller is also used to close the electric raw liquid shut-off valve, open the flushing water valve, and control the servo metering pump to run at low speed to flush the raw liquid pipeline and pump chamber with clean water after the fire extinguishing or test is completed. If the incident is determined to be a non-real fire and the online concentration detection device confirms that the concentration of the foam mixture is within the acceptable range, the three-way switching valve is controlled to introduce the mixture into the recovery tank.

[0009] As an optional solution of the mechanically pumped foam extinguishing agent proportioning system of the present invention, the servo metering pump is further provided with a dry run protection module. When the dry run protection module detects that the liquid-free state has lasted for more than a set time threshold, it automatically stops the servo metering pump and issues a fault alarm signal.

[0010] As an optional solution of the mechanically pumped foam extinguishing agent proportioning system of the present invention, it further includes a communication module, which uploads operating data, concentration data, equipment health status and fault information to the remote fire management platform; A remote fire management module, which is equipped with a digital twin simulation engine; The area of ​​the fire covered by foam at time t is calculated using the digital twin simulation engine. : in, This is an empirical coefficient, related to foam type, nozzle type, and floor material, and is fitted using historical data. The mixing ratio is... For water flow rate, This is an exponentially decaying term, indicating that the coverage is approaching saturation; The digital twin simulation engine is used to virtually simulate the foam coverage effect under various mixing ratio strategies based on the current fire scene parameters and the system physical model after receiving a fire alarm signal, and generate a recommended mixing strategy to be sent to the controller. Federated learning aggregates de-identified operational data from multiple deployment sites to collaboratively train and update the mixture ratio optimization model without exchanging the original sensitive data. Federated learning includes: in, These are the updated global AI model parameters from the cloud. These are the model parameters trained locally at the i-th site; The weight of the i-th site is usually allocated based on the amount of data. The total number of sites participating in federated learning; The communication module is also configured to receive the updated hybrid ratio optimization model issued by the remote fire management platform and incrementally upgrade the local AI optimization module via over-the-air download.

[0011] As an optional solution of the mechanically pumped foam fire extinguishing agent proportioning system described in this invention, the controller adopts a dual-partition firmware storage architecture, and OTA upgrades are only performed when the system is in standby mode. The new firmware is written to the inactive partition and verified by digital signature. If the new firmware fails to pass the concentration closed-loop control self-test within the set time after startup, it will automatically roll back to the original firmware version and send a rollback alarm to the remote fire management platform.

[0012] The present invention also provides a method for a mechanically pumped foam extinguishing agent proportioning system, comprising the following steps: S1. After the fire protection system is activated, the water flow rate in the main water pipe is detected by a water flow sensor. ; S2. Obtain the dynamic optimal mixing ratio The dynamic optimal mixing ratio is generated based on a mixing ratio optimization model trained on historical fire extinguishing data, combined with current fire situation awareness information and real-time operating conditions. S3. Based on the optimal mixing ratio and Calculate the target foam concentrate flow rate ; S4, drive the servo metering pump as described The system operates by injecting the foam concentrate into the main water flow. S5. Real-time acquisition of actual foam flow rate measured by the foam flow meter. The actual concentration of the foam mixture measured by the online concentration detection device. ; S6, based on and The deviation is used to perform closed-loop flow regulation, and based on the target concentration and The concentration is adjusted in a closed-loop manner to dynamically correct the operating parameters of the servo metering pump by adjusting the deviation. S7. After the fire is extinguished or the test is completed, close the electric raw liquid shut-off valve, open the flushing water valve, and control the servo metering pump to run at low speed for 30 to 60 seconds to flush the raw liquid pipeline and pump chamber with clean water. S8. If it is determined to be a non-real fire incident and the above... If the mixture is within the preset acceptable range, the three-way switching valve will be controlled to guide the foam mixture into the recovery tank. S9. Upload operational data, concentration data, equipment health status and fault information to the remote fire management platform via the communication module; S10. Receive the updated mixing ratio optimization model sent by the remote fire management platform via over-the-air download, and perform OTA upgrade using a dual-partition firmware architecture when the system is in standby mode and there is no water flow signal; if the new firmware fails to pass the concentration closed-loop control self-test within a preset time after startup, it will automatically roll back to the original firmware version and send a rollback alarm.

[0013] The present invention has the following beneficial effects: 1. This mechanically pumped foam extinguishing agent proportioning system and method directly monitors the final mixture concentration through online concentration detection and closed-loop feedback, effectively avoiding extinguishing failures caused by raw material deterioration, incorrect dilution, or temperature drift. Secondly, it supports dynamic optimization of the mixing ratio, allowing for on-demand mixing, and also supports the recycling and reuse of qualified mixtures, reducing fluorinated foam emissions and aligning with global PFAS environmental trends. Furthermore, its dry-run protection function automatically stops the pump in the absence of liquid, preventing metering pump burnout, automatic flushing prevents crystallization blockage, and predictive maintenance based on operational data provides early warnings of malfunctions.

[0014] 2. This mechanically pumped foam extinguishing agent proportioning mixing system and method upgrades from manual setting of proportions to autonomous system decision-making through unattended operation. It can integrate multi-source sensing such as video, infrared, and gas to automatically identify the fire type and match the optimal strategy. It is particularly suitable for key locations with few or no personnel, such as airports, ships, and data centers. It also has a safe and reliable remote upgrade capability, adopting dual-partition firmware, digital signature, and self-test rollback mechanism. OTA upgrades are only performed in standby mode and when there is no water flow, and will never affect the fire extinguishing function. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall operation process of the present invention.

[0016] Figure 2 This is a flowchart illustrating the dynamic optimal mixing ratio generation process of this invention.

[0017] Figure 3 This is a flowchart of the dual closed-loop regulation process of the present invention.

[0018] Figure 4 This is a flowchart of the remote management and OTA upgrade process of this invention. Detailed Implementation

[0019] 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.

[0020] Example 1 Please see Figures 1-4 One type of mechanically pumped foam extinguishing agent proportioning system includes: The main water flow pipeline is equipped with a water flow sensor. The foam raw material supply pipeline is equipped with a raw material storage tank, an electric raw material shut-off valve, a servo metering pump and a foam flow meter in sequence. A static mixer connects the outlet of the main water supply line to the foam concentrate supply line and is used to generate foam concentrate. An online concentration detection device is installed on one side of the static mixer. The online concentration detection device is used to measure the actual concentration of the foam stock solution in the foam mixture in real time. The controller is electrically connected to the water flow sensor, foam flow meter, servo metering pump, electric raw material shut-off valve and concentration online detection device. The controller is used to calculate the target foam flow rate based on the dynamically optimized mixing ratio and the water flow rate measured by the water flow sensor. Drive the servo metering pump to operate at the target foam flow rate; The flow rate is adjusted in a closed loop based on feedback from the foam flow meter, and the concentration is adjusted in a closed loop based on the actual concentration measured by the online concentration detection device, thereby dynamically correcting the operating parameters of the servo metering pump.

[0021] Specifically, through high-precision fluid control, dual closed-loop feedback, and intelligent dynamic optimization, the foam mixture concentration is accurately, stably, and adaptively output. In standby mode, the electric raw material shut-off valve is closed, the servo metering pump stops, and the controller continuously monitors the water flow sensor signal to determine whether to start. When a fire occurs and the fire pump is turned on, water flows into the main water pipeline, and the water flow sensor detects the water flow rate in real time. This information is then transmitted to the controller. The controller obtains the dynamically optimal mixing ratio. This value can come from a preset default value, or it can be dynamically generated by the AI ​​optimization module based on fire perception information, ambient temperature, and historical successful cases, and the target foam flow rate can be calculated accordingly. Subsequently, the controller activates the electric feedstock shut-off valve, driving the servo metering pump to extract foam feedstock from the feedstock storage tank. After being metered by the foam flow meter, it is thoroughly mixed with the main water flow in the static mixer to form a foam mixture. The static mixer is equipped with spiral blades or a porous structure to ensure that the water and foam feedstock are uniformly mixed within 0.5 seconds. Two closed-loop control loops are executed simultaneously to ensure accurate output concentration.

[0022] The first layer is a flow closed loop, where the foam flow meter measures the actual injected foam flow rate in real time. The controller calculates the deviation and generates a correction amount Δv1 using an algorithm, with the following formula: in, This is the pump speed correction amount, used to fine-tune the servo pump speed. This is the measured value from the foam flow meter. , and The proportional, integral, and derivative gain coefficients of the flow loop are obtained through on-site tuning. This is the cumulative error amount, used to eliminate steady-state deviation. This is the rate of change of error, used to suppress overshoot; The servo metering pump speed is dynamically adjusted to eliminate flow errors caused by pump wear or changes in the viscosity of the raw liquid.

[0023] The second layer is a concentration closed loop. After the mixed liquid flows out of the static mixer, it immediately enters the online concentration detection device (such as a conductivity sensor). This device inverts the actual mixed concentration based on the conductivity-concentration calibration curve and generates a correction amount Δv2 through another PID loop to further fine-tune the pump speed. The calculation formula is as follows: in, This is the concentration correction amount. To achieve the target foam concentration, For the concentration measured by the concentration sensor, , These are the PID parameters for the concentration loop; The final pump speed command is superimposed on the base speed. and Total pump speed command synthesis = ( )+ + This dual closed-loop mechanism ensures that even with different batches of raw materials, temperature changes, or pipeline aging, the final output foam mixture concentration remains stable within ±0.3% of the target value.

[0024] After the fire is extinguished, the water flow stops, and the controller enters the final stage, closing the electric raw material shut-off valve, opening the flushing water valve, and controlling the servo metering pump to run at low speed for 30 to 60 seconds to flush the raw material pipeline, pump chamber, and static mixer with clean water to prevent foam residue and crystallization. If the system determines that this was a test or a false start (confirmed by the fire control panel signal), and the concentration sensor confirms that the mixture is qualified (e.g., 6% ± 0.2%), it automatically switches the three-way valve to guide the mixture into the recovery tank for subsequent training, avoiding waste. In addition, all operational data (including flow rate, concentration, faults, and health status) can be uploaded to the remote fire management platform via the communication module, supporting digital twin simulations, federated learning model updates, and secure OTA policy distribution.

[0025] In summary, this example demonstrates a significant improvement in mixing accuracy. Traditional systems only control the injection ratio but cannot verify the mixing results. Through online concentration detection and closed-loop feedback, the final mixture concentration is directly monitored, effectively preventing extinguishing failures caused by raw material deterioration, incorrect dilution, or temperature drift. Secondly, it supports dynamic optimization of the mixing ratio, allowing for on-demand mixing. For example, using 5.2% instead of a fixed 6% for small-area initial fires has been measured to save on foam raw material annually. It also supports the recycling and reuse of qualified mixtures, reducing fluorinated foam emissions and aligning with global PFAS environmental trends. Furthermore, the dry-run protection function automatically stops the pump in the absence of liquid, preventing metering pump burnout. Automatic flushing prevents crystallization blockage, and predictive maintenance based on operational data provides early warnings of malfunctions. Supports unattended operation, upgrading from manual setting of the ratio to autonomous system decision-making. It can integrate multi-source sensing such as video, infrared, and gas to automatically identify the fire type and match the optimal strategy. It is particularly suitable for key locations with few or no personnel, such as airports, ships, and data centers. It also has a safe and reliable remote upgrade capability, adopting dual-partition firmware, digital signature and self-test rollback mechanism. OTA upgrades are only performed in standby mode and when there is no water flow, and will never affect the fire extinguishing function.

[0026] Example 2 This embodiment is an improvement upon embodiment 1. For details, please refer to [link / reference]. Figures 1-4 It also includes: The AI ​​optimization module is used to acquire water flow parameters, foam parameters, fire type information, and fire extinguishing effect evaluation data from historical fire extinguishing events. Based on data training, a mixture ratio optimization model is trained. In new fire extinguishing events, based on the current fire situation awareness information and real-time operating conditions, the dynamic optimal mixing ratio is output as the basis for the controller to calculate the target foam flow rate; Fire type information is obtained through external sensing devices; The AI ​​optimization module adjusts the dynamic optimal mixing ratio based on the type of fire identified by external sensing devices.

[0027] Specifically, the AI ​​optimization module is essentially a lightweight AI inference engine deployed on an edge controller. It dynamically generates an optimal foam mixing ratio that is superior to a fixed value when a fire occurs. This module does not rely on human experience or preset rules, but rather uses a machine learning model to calculate the most suitable mixing ratio for the current scenario based on historical firefighting data and current fire scene awareness information.

[0028] The module's workflow is divided into two phases: an offline training phase and an online inference phase. During the offline training phase, anonymized historical fire suppression event data was collected from multiple deployed sites. Each data record included input features and output labels. Input features primarily included water flow parameters (such as main water flow rate and water pressure fluctuation rate), foam parameters (such as concentrate type, ambient temperature, and concentrate viscosity), fire scene type information (such as combustible material type, flame area, and heat release rate), and location characteristics (such as tank diameter, hangar height, and whether it is a confined space). The output labels were the actual mixing ratio used and the corresponding fire suppression effect evaluation indicators, such as fire control time, reignition frequency, and foam coverage uniformity. This data was used to train a regression-based machine learning model, preferably a lightweight gradient boosting tree model, such as XGBoost or LightGBM, due to its high accuracy, strong interpretability, low memory footprint, and fast inference speed, making it ideal for running on resource-constrained embedded controllers. After training, the model was compressed and embedded into the AI ​​optimization module of the edge device.

[0029] During the online inference phase, when a new fire alarm is triggered, the AI ​​optimization module first receives real-time fire information from external sensing devices. These external sensing devices include infrared thermal imagers, video flame recognition cameras, and volatile organic compound (VOC) gas sensors. The infrared thermal imager provides a fire temperature distribution map to determine the fire intensity and spread trend; the video camera analyzes flame morphology, color, and dynamic characteristics using a lightweight convolutional neural network (such as MobileNetV2) to identify the type of combustible material (such as oil, solids, and polar solvents); and the gas sensors detect concentrations of carbon monoxide, hydrocarbons, etc., to help determine whether it is smoldering or open flame. This multi-source information is fused and encoded into a structured feature vector, which is then input into the local AI model. The model then outputs a dynamically optimal mixing ratio, for example, 5.8%, instead of the traditional fixed 6%. This value is directly passed to the main controller as the basis for calculating the target foam flow rate.

[0030] The AI ​​optimization module also supports continuous learning capabilities. After each fire suppression operation, the effectiveness of the strategy is evaluated. If the fire control time is shorter than the threshold and there is no reignition, the complete feature label pairs of this event are added to the local incremental training set. Periodically (e.g., weekly) or after accumulating enough new samples, the module can fine-tune the model to better suit the characteristics of the local site. In addition, through a federated learning mechanism with the remote fire management platform, each site can collaboratively update the global model without sharing original sensitive data, achieving collective intelligent evolution.

[0031] Example 3 This embodiment is an improvement upon embodiment 2. For details, please refer to [link / reference]. Figures 1-4 It also includes a communication module, which uploads operating data, concentration data, equipment health status and fault information to the remote fire management platform; The remote fire management module is equipped with a digital twin simulation engine. The digital twin simulation engine is used to virtually simulate the foam coverage effect under various mixing ratio strategies after receiving a fire alarm signal, based on current fire scene parameters (including fire area, obstacle layout, wind speed, and type of combustible material) and a system physical model (including pipeline characteristics, pump performance curves, and nozzle diffusion angle). Specifically, the engine uses the following mathematical model of foam coverage area changing over time for simulation calculations: The area of ​​the fire covered by foam at time t was calculated using a digital twin simulation engine. : in, This is an empirical coefficient, related to foam type, nozzle type, and floor material, and is fitted using historical data. The mixing ratio is... For water flow rate, This is an exponentially decaying term, indicating that the coverage is approaching saturation; The digital twin simulation engine is used to virtually simulate the foam coverage effect under various mixing ratio strategies based on the current fire scene parameters and the system physical model after receiving a fire alarm signal, and generate recommended mixing strategies to be sent to the controller. It is worth noting that the simulation engine iterates through different R values ​​to predict the time required to achieve an effective fire control area (e.g., covering 90% of the fire area) under each strategy, and then selects the recommended mixing strategy that meets the fire extinguishing time requirements and has the lowest original liquid consumption. Subsequently, it calculates the optimal mixing ratio corresponding to this strategy. The settings are sent to the on-site controller as the initial settings for this firefighting operation. The federated learning unit is used to achieve collaborative intelligent evolution across multiple sites. This unit aggregates anonymized operational data from multiple deployment sites and collaboratively trains and updates the mixture ratio optimization model without exchanging original sensitive data (such as specific site layouts, original concentration waveforms, and video images). Federated learning includes: in, These are the updated global AI model parameters from the cloud. These are the model parameters trained locally at the i-th site; The weight of the i-th site is usually allocated based on the amount of data. The total number of sites participating in federated learning; The communication module is also configured to receive updated hybrid ratio optimization models from the remote fire management platform and incrementally upgrade the local AI optimization module via over-the-air (OTA) updates. The upgrade process strictly adheres to security mechanisms. This process is executed only when the system is in standby mode and there is no water flow signal. The new firmware is written to the inactive storage partition and its integrity is verified by SM2 or RSA digital signature. After the new model is loaded, a concentration closed-loop control self-test process (such as simulating small flow operation and verifying the stability of concentration output) is required. If it fails to pass within the preset time, it will automatically roll back to the original version and send an alarm to the platform to ensure that the fire protection function is never interrupted.

[0032] The present invention also provides a method for a mechanically pumped foam extinguishing agent proportioning system, comprising the following steps: S1. After the fire protection system is activated, the water flow rate in the main water pipe is detected by a water flow sensor. ; S2. Obtain the dynamic optimal mixing ratio The dynamic optimal mixing ratio is generated based on a mixing ratio optimization model trained on historical fire extinguishing data, combined with current fire situation awareness information and real-time operating conditions. S3. Based on the optimal mixing ratio and Calculate the target foam concentrate flow rate ; S4, drive servo metering pump according to The system operates by injecting the foam concentrate into the main water flow. S5. Real-time acquisition of actual foam flow rate measured by the foam flow meter. The actual concentration of the foam mixture measured by the online concentration detection device. ; S6, based on and The deviation is used to perform closed-loop flow regulation, and based on the target concentration and The concentration is adjusted in a closed-loop manner to correct the deviation and the operating parameters of the servo metering pump are dynamically corrected. S7. After the fire is extinguished or the test is completed, close the electric raw liquid shut-off valve, open the flushing water valve, and control the servo metering pump to run at low speed for 30 to 60 seconds to flush the raw liquid pipeline and pump chamber with clean water. S8. If it is determined to be a non-real fire incident and If the mixture is within the preset acceptable range, the three-way switching valve will be controlled to guide the foam mixture into the recovery tank. S9. Upload operational data, concentration data, equipment health status and fault information to the remote fire management platform via the communication module; S10: Receive the updated mixing ratio optimization model from the remote fire management platform via over-the-air download, and perform OTA upgrade using a dual-partition firmware architecture when the system is in standby mode and there is no water flow signal; if the new firmware fails to pass the concentration closed-loop control self-test within a preset time after startup, it will automatically roll back to the original firmware version and send a rollback alarm.

[0033] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0034] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A mechanically pumped foam extinguishing agent proportioning system, characterized in that, include: The main water flow pipeline is equipped with a water flow sensor. A foam raw material supply pipeline is provided with a raw material storage tank, an electric raw material shut-off valve, a servo metering pump and a foam flow meter in sequence. A static mixer, which is connected to the outlet of the main water flow pipeline and the foam concentrate supply pipeline, is used to generate foam concentrate; An online concentration detection device is installed on one side of the static mixer, and the online concentration detection device is used to measure the actual concentration of foam stock solution in the foam mixture in real time. The controller is electrically connected to the water flow sensor, foam flow meter, servo metering pump, electric raw material shut-off valve and concentration online detection device respectively. The controller is used to calculate the target foam flow rate based on the dynamically optimized mixing ratio and the water flow rate measured by the water flow sensor. Drive the servo metering pump to operate at the target foam flow rate; The flow rate is adjusted in a closed loop based on the feedback from the foam flow meter, and the concentration is adjusted in a closed loop based on the actual concentration measured by the online concentration detection device, thereby dynamically correcting the operating parameters of the servo metering pump.

2. The mechanically pumped foam extinguishing agent proportioning system according to claim 1, characterized in that, Also includes: The AI ​​optimization module is used to acquire water flow parameters, foam parameters, fire type information, and fire extinguishing effect evaluation data from historical fire extinguishing events. A mixture ratio optimization model is trained based on the data; In a new fire extinguishing event, based on the current fire situation information and real-time operating conditions, the dynamic optimal mixing ratio is output as the basis for the controller to calculate the target foam flow rate; The fire type information is obtained through external sensing devices; The AI ​​optimization module adjusts the dynamic optimal mixing ratio based on the type of fire identified by the external sensing device.

3. The mechanically pumped foam extinguishing agent proportioning system according to claim 2, characterized in that: The foam flow rate is calculated as follows: in, , where is the target foam concentrate flow rate, and is the desired foam concentrate flow rate output by the metering pump from the controller; The dynamic optimal mixing ratio; The fire water flow rate is measured in real time by a water flow sensor to accurately measure the main water flow rate. The closed-loop regulation includes an inner loop flow closed-loop and an outer loop concentration closed-loop; The inner loop traffic closed loop includes: in, This is the pump speed correction amount, used to fine-tune the servo pump speed. This is the measured value from the foam flow meter. , and The proportional, integral, and derivative gain coefficients of the flow loop are obtained through on-site tuning. This is the cumulative error amount, used to eliminate steady-state deviation. This is the rate of change of error, used to suppress overshoot; The outer ring concentration closed loop includes: in, This is the concentration correction amount. To achieve the target foam concentration, For the concentration measured by the concentration sensor, , These are the PID parameters for the concentration loop; based on , and Calculate the total pump speed command synthesis: Total pump speed command synthesis = ( )+ + .

4. The mechanically pumped foam extinguishing agent proportioning system according to claim 1, characterized in that, Also includes: An automatic rinsing and raw liquid recovery module, comprising a rinsing water valve, a three-way switching valve, and a recovery tank; The controller is also used to close the electric raw liquid shut-off valve, open the flushing water valve, and control the servo metering pump to run at low speed to flush the raw liquid pipeline and pump chamber with clean water after the fire extinguishing or test is completed. If the incident is determined to be a non-real fire and the online concentration detection device confirms that the concentration of the foam mixture is within the acceptable range, the three-way switching valve is controlled to introduce the mixture into the recovery tank.

5. The mechanically pumped foam extinguishing agent proportioning system according to claim 1, characterized in that: The servo metering pump is also equipped with a dry run protection module. When the dry run protection module detects that the liquid-free state has lasted for more than a set time threshold, it automatically stops the servo metering pump and issues a fault alarm signal.

6. The mechanically pumped foam extinguishing agent proportioning system according to claim 5, characterized in that, It also includes a communication module, which uploads operating data, concentration data, equipment health status and fault information to the remote fire management platform; A remote fire management module, which is equipped with a digital twin simulation engine; The area of ​​the fire covered by foam at time t is calculated using the digital twin simulation engine. : in, This is an empirical coefficient, related to foam type, nozzle type, and floor material, and is fitted using historical data. The mixing ratio is... For water flow rate, This is an exponentially decaying term, indicating that the coverage is approaching saturation; The digital twin simulation engine is used to virtually simulate the foam coverage effect under various mixing ratio strategies based on the current fire scene parameters and the system physical model after receiving a fire alarm signal, and generate a recommended mixing strategy to be sent to the controller. Federated learning aggregates de-identified operational data from multiple deployment sites to collaboratively train and update the mixture ratio optimization model without exchanging the original sensitive data. Federated learning includes: in, These are the updated global AI model parameters from the cloud. These are the model parameters trained locally at the i-th site; The weight of the i-th site is usually allocated based on the amount of data. The total number of sites participating in federated learning; The communication module is also configured to receive the updated hybrid ratio optimization model issued by the remote fire management platform and incrementally upgrade the local AI optimization module via over-the-air download.

7. The mechanically pumped foam extinguishing agent proportioning system according to claim 6, characterized in that: The controller adopts a dual-partition firmware storage architecture. OTA upgrades are only performed when the system is in standby mode. The new firmware is written to the inactive partition and verified by digital signature. If the new firmware fails to pass the concentration closed-loop control self-test within the set time after startup, it will automatically roll back to the original firmware version and send a rollback alarm to the remote fire management platform.

8. The mixing method of the mechanically pumped foam extinguishing agent proportioning system according to any one of claims 1-7, characterized in that, Includes the following steps: S1. After the fire protection system is activated, the water flow rate in the main water pipe is detected by a water flow sensor. ; S2. Obtain the dynamic optimal mixing ratio The dynamic optimal mixing ratio is generated based on a mixing ratio optimization model trained on historical fire extinguishing data, combined with current fire situation awareness information and real-time operating conditions. S3. Based on the optimal mixing ratio and Calculate the target foam concentrate flow rate ; S4, drive the servo metering pump as described The system operates by injecting the foam concentrate into the main water flow. S5. Real-time acquisition of actual foam flow rate measured by the foam flow meter. The actual concentration of the foam mixture measured by the online concentration detection device. ; S6, based on and The deviation is used to perform closed-loop flow regulation, and based on the target concentration and The concentration is adjusted in a closed-loop manner to dynamically correct the operating parameters of the servo metering pump by adjusting the deviation. S7. After the fire is extinguished or the test is completed, close the electric raw liquid shut-off valve, open the flushing water valve, and control the servo metering pump to run at low speed for 30 to 60 seconds to flush the raw liquid pipeline and pump chamber with clean water. S8. If it is determined to be a non-real fire incident and the above... If the mixture is within the preset acceptable range, the three-way switching valve will be controlled to guide the foam mixture into the recovery tank. S9. Upload operational data, concentration data, equipment health status and fault information to the remote fire management platform via the communication module; S10. Receive the updated mixing ratio optimization model sent by the remote fire management platform via over-the-air download, and perform OTA upgrade using a dual-partition firmware architecture when the system is in standby mode and there is no water flow signal; if the new firmware fails to pass the concentration closed-loop control self-test within a preset time after startup, it will automatically roll back to the original firmware version and send a rollback alarm.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in claim 8.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method as described in claim 8.

Images