Intelligent control method, system, electronic equipment and storage medium for smoke generator based on preheating state recognition

Abstract

This application relates to the technical field of smoke generator control, and provides an intelligent control method, system, electronic device, and storage medium for smoke generators based on preheating state recognition, to solve the problem of low accuracy in determining the preheating state of smoke generators in existing technologies. The method includes: acquiring multi-dimensional data of the smoke generator and its surrounding environment; extracting three key feature parameters from the multi-dimensional data: preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature; subsequently using a decision tree classification algorithm integrated into a PLC to determine whether the current preheating state of the smoke generator is in an unready, transitional, or stable-ready state, obtaining a state determination result; finally, based on this result, adaptively adjusting the control parameters of the smoke generator using a PID control algorithm. This application can achieve intelligent smoke output control that accurately adapts to the preheating state of the smoke generator.

Description

Technical Field

[0001] This application relates to the technical field of smoke generator control, and in particular to an intelligent control method, system, electronic device and storage medium for a smoke generator based on preheating state recognition. Background Technology

[0002] In flight simulator training, smoke generators are needed to simulate smoke effects in scenarios such as engine malfunctions to ensure the realism of the training. Furthermore, during cockpit airtightness testing, a stable output of smoke is required to accurately assess sealing performance. Therefore, these scenarios all require precise matching between the smoke output and concentration and the preheating state of the smoke generator to avoid abnormal smoke output due to insufficient or excessive preheating, which could affect the accuracy of the scenario simulation or test results.

[0003] Currently, some existing smoke generator control technologies complete the preheating process by preset a fixed heating time. After the set time is reached, smoke liquid and heating power are supplied according to fixed parameters. Some use temperature sensors to monitor a single temperature value of the preheating chamber. When the temperature reaches a preset threshold, smoke output is started. Others change the smoke supply rate through a simple gear adjustment mechanism. Some other technologies control the heating process through a preset temperature curve to maintain a fixed preheating power and smoke liquid supply ratio.

[0004] However, the main drawback of existing technology is that the determination of the preheating state of the smoke generator is based on a single dimension, relying only on a fixed time or a single temperature index. This cannot fully reflect the actual efficiency of heat accumulation and material energy exchange during the preheating process, resulting in a lack of targeted adjustment of control parameters. It is difficult to achieve dynamic adaptation between smoke output and real-time preheating state, leading to problems such as unstable smoke concentration and delayed output response. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent control method, system, electronic device and storage medium for a smoke generator based on preheating status recognition, so as to solve the problem of low accuracy in determining the preheating status of a smoke generator in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an intelligent control method for a smoke generator based on preheating state recognition, comprising:

[0007] Acquire multi-dimensional data about the smoke generator and its surrounding environment;

[0008] Key feature parameters reflecting preheating efficiency are extracted from the multi-dimensional data.

[0009] Based on the key feature parameters, a decision tree classification algorithm integrated into the PLC is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state, and the state determination result is obtained.

[0010] Based on the state determination result, a PID control algorithm is used to adaptively adjust the control parameters of the smoke generator to achieve smoke output control that matches the preheating state.

[0011] Optionally, the step of adaptively adjusting the control parameters of the smoke generator using a PID control algorithm based on the state determination result includes:

[0012] Based on the control parameter records in historical training rounds, establish control mapping relationships corresponding to different preheating states;

[0013] Based on the control mapping relationship, the initial control parameters corresponding to the state determination result are determined;

[0014] Based on the initial control parameters, a composite adjustment mechanism including feedforward and feedback is adopted. Combining the heat accumulation index and the response coefficients of flow rate and temperature, the real-time adjustment amount of preheating power and the dynamic compensation amount of flue gas supply rate are determined. Through the multivariate coordination mechanism of PID control algorithm, the real-time adjustment amount and the dynamic compensation amount are converted into intermediate control parameters.

[0015] Based on the preheating status evaluation data from historical training rounds, a target optimization function is constructed that includes the cumulative consumption of smoke liquid, the stability of preheating temperature, and the residual concentration of exhaust smoke. Based on the target optimization function, the intermediate control parameters are optimized to obtain preliminary optimized parameters.

[0016] By combining the smoke emission index and preheating temperature data, the preliminary optimization parameters are adaptively adjusted to obtain the adaptively adjusted control parameters.

[0017] Optionally, determining the initial control parameters corresponding to the state determination result based on the control mapping relationship includes:

[0018] Based on the state determination result, the corresponding basic control parameters are selected from the control mapping relationship;

[0019] Based on the current environmental conditions, the basic control parameters are modified to adapt to environmental conditions, resulting in the parameters to be adjusted.

[0020] Based on a preset domain knowledge base, the parameters to be adjusted are adjusted to obtain initial control parameters.

[0021] Optionally, based on the initial control parameters, a composite adjustment mechanism including feedforward and feedback is adopted. Combining the heat accumulation index and the response coefficients of flow rate and temperature, the real-time adjustment of preheating power and the dynamic compensation of the flue gas supply rate are determined. Through the multivariate coordination mechanism of the PID control algorithm, the real-time adjustment and the dynamic compensation are converted into intermediate control parameters, including:

[0022] Based on the initial control parameters, a feedforward prediction mechanism is used to process the changing trend of the heat accumulation index to obtain the feedforward control quantity of the preheating power.

[0023] Based on the response coefficients of flow rate and temperature, the deviation between the current preheating state and the target preheating state is processed through a feedback correction mechanism to obtain the feedback compensation amount of preheating power.

[0024] The feedforward control quantity and the feedback compensation quantity are weighted and fused to obtain the real-time adjustment quantity of the preheating power;

[0025] Based on the real-time adjustment amount, the dynamic compensation amount of the e-liquid supply rate is obtained by deriving the demand change of the e-liquid supply rate through the preset dynamic characteristic model of the supply system.

[0026] A multivariate coordination mechanism using a PID control algorithm is employed to coordinate the real-time adjustment and dynamic compensation quantities to eliminate mutual interference between parameters. Based on the results of the coordinated processing, intermediate control parameters are generated.

[0027] Optionally, the step of combining smoke emission index and preheating temperature data to adaptively adjust the preliminary optimization parameters to obtain adaptively adjusted control parameters includes:

[0028] By cross-coupling analysis of smoke emission indicators and preheating temperature data, an influence relationship model was established;

[0029] Based on the aforementioned influence relationship model, the initial optimization parameters are adaptively adjusted to obtain the adaptively adjusted control parameters.

[0030] Optionally, based on the key feature parameters, a decision tree classification algorithm integrated into the PLC is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state, and the state determination result is obtained, including:

[0031] Based on the key feature parameters, an input feature vector for decision tree classification is constructed. The key feature parameters include preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature.

[0032] The input feature vector is fed into a pre-generated decision tree classification model integrated into the PLC. The following classification logic is then executed sequentially through the decision tree classification model:

[0033] At the root node, the preheating temperature difference is compared with a first threshold. If the first comparison result is that the preheating temperature difference is lower than the first threshold, the branch goes to the leaf node representing the not-ready state. If the first comparison result is that the preheating temperature difference is higher than or equal to the first threshold, the branch goes to the next internal node.

[0034] In the next internal node, compare the temperature rise rate with the second threshold: if the second comparison result is that the temperature rise rate is lower than the second threshold, then branch to the leaf node representing the transition state; if the second comparison result is that the temperature rise rate is higher than or equal to the second threshold, then go to the final internal node.

[0035] At the final internal node, compare the coordination coefficient between the flow rate and temperature with a third threshold: if the coordination coefficient between the flow rate and temperature is lower than the third threshold, then branch to the leaf node representing the transition state; if the coordination coefficient between the flow rate and temperature is higher than or equal to the third threshold, then branch to the leaf node representing the stable and ready state.

[0036] The state corresponding to the reached leaf node is determined as the state determination result.

[0037] Optionally, the key characteristic parameters include preheating temperature difference, temperature rise rate, and the synergy coefficient between flow rate and temperature;

[0038] The key feature parameters extracted from the multi-dimensional data to reflect preheating efficiency include:

[0039] Calculate the difference between the outlet temperature of the preheating chamber and the real-time temperature to obtain the preheating temperature difference;

[0040] The rate of temperature rise is obtained by performing a differential operation on the change of the real-time temperature in a continuous time series.

[0041] The correspondence between instantaneous flow rate and real-time temperature in the multi-dimensional data at the same time is analyzed to obtain the coordination coefficient between flow rate and temperature;

[0042] By integrating the preheating temperature difference, the temperature rise rate, and the synergy coefficient between the flow rate and temperature, key characteristic parameters are obtained.

[0043] Secondly, this application provides an intelligent control system for a smoke generator based on preheating state recognition, comprising:

[0044] The acquisition module is used to acquire multi-dimensional data about the smoke generator and its surrounding environment.

[0045] The extraction module is used to extract key feature parameters reflecting preheating efficiency from the multi-dimensional data;

[0046] The determination module is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state based on the key feature parameters and using a decision tree classification algorithm integrated into the PLC, and obtain the state determination result.

[0047] The adjustment module is used to adaptively adjust the control parameters of the smoke generator based on the state determination result and using a PID control algorithm to achieve smoke output control that matches the preheating state.

[0048] Thirdly, this application provides an electronic device, comprising:

[0049] Memory, used to store computer programs;

[0050] A processor is configured to execute the computer program to implement the steps of the intelligent control method for a smoke generator based on preheating state recognition as described in the first aspect above.

[0051] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the intelligent control method for a smoke generator based on preheating state recognition as described in the first aspect above.

[0052] The technical solution of this application has the following beneficial effects:

[0053] This application comprehensively acquires operational status information of the smoke generator and related systems, providing multi-dimensional and accurate data support for subsequent analysis of preheating efficiency, thus avoiding judgment bias caused by missing information. Next, by transforming fragmented multi-dimensional data into key characteristic parameters that directly reflect preheating efficiency, it clearly presents heat transfer, heating rate, and material-energy balance, providing a foundation for accurate preheating status determination. Then, relying on PID control algorithms for in-depth analysis of key characteristic parameters, it accurately identifies the real-time preheating status of the smoke generator, avoiding misjudgments based on a single indicator and ensuring reliable status determination results. Finally, it enables control parameters to dynamically change with the preheating status, achieving precise matching between smoke output and current preheating efficiency, avoiding problems such as smoke concentration fluctuations or output lag.

[0054] Furthermore, this application provides a reliable basis for initial control parameters by constructing a control mapping relationship based on historical data. The composite adjustment mechanism, which includes feedforward and feedback, ensures that the real-time adjustment and compensation quantities match the actual needs. Then, the multivariate coordination mechanism of the PID control algorithm can avoid mutual interference between parameters. Moreover, the objective optimization function takes into account the requirements of smoke liquid conservation, temperature stability, and low residue. Combined with the final adjustment of smoke emission indicators and temperature, the adaptability of parameters is further improved. Ultimately, the control parameters of the smoke generator are accurately and dynamically optimized, thereby ensuring stable smoke output that meets the requirements of the scenario. Attached Figure Description

[0055] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart illustrating an intelligent control method for a smoke generator based on preheating state recognition, provided in an embodiment of this application;

[0057] Figure 2 A schematic diagram illustrating a specific implementation of an intelligent control method for a smoke generator based on preheating state recognition, provided in an embodiment of this application;

[0058] Figure 3 A schematic diagram of the structure of an intelligent control system for a smoke generator based on preheating state recognition provided in this application embodiment;

[0059] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0060] In scenarios that rely on smoke generators, such as flight simulation training and cockpit airtightness testing, the stability of smoke output directly affects the training or testing results. However, existing smoke generator control technologies often judge the preheating status by only fixing the heating time or a single temperature value. This single-dimensional judgment method cannot fully capture the actual state during the preheating process, such as whether the heat accumulation is sufficient and whether the smoke liquid and heat are matched. This results in a lack of targeted adjustment of control parameters, often leading to problems such as fluctuating smoke concentration and delayed output response, making it difficult to meet the precise requirements of the scenario for smoke effects.

[0061] To address the aforementioned issues, this application proposes an intelligent control method for a smoke generator based on preheating state recognition. This method first comprehensively collects multi-dimensional data from equipment such as the smoke liquid supply pipeline, preheating chamber, control cabin, and exhaust fan, extracting key feature parameters reflecting the preheating effect from this data. Then, it analyzes these key feature parameters using a decision tree classification algorithm to accurately determine whether the smoke generator is in an unready, transitional, or stable ready state. Finally, it dynamically adjusts the smoke liquid supply rate and preheating power based on the state determination results. This multi-dimensional perception, accurate state judgment, and adaptive parameter adjustment approach overcomes the shortcomings of existing technologies with their single-dimensional judgment, enabling precise matching of control parameters with the real-time preheating state, thereby effectively ensuring stable smoke output that meets scenario requirements.

[0062] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0063] The core of this application is to provide an intelligent control method for a smoke generator based on preheating state recognition, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0064] S101. Acquire multi-dimensional data of the smoke generator and its surrounding environment;

[0065] The multi-dimensional data includes: instantaneous flow rate, cumulative consumption and supply pressure of the smoke liquid supply pipeline, real-time temperature of the preheating chamber wall, outlet temperature of the preheating chamber, smoke concentration in the cockpit, exhaust fan speed and exhaust duration, and temperature response delay time.

[0066] Furthermore, the instantaneous flow rate of the smoke liquid supply pipeline refers to the amount of smoke liquid flowing through the pipeline per unit time, the cumulative consumption is the total amount of smoke liquid used within a certain period, and the supply pressure is the pushing pressure of the smoke liquid in the pipeline; the real-time temperature of the preheating chamber wall refers to the current temperature of the inner wall of the preheating chamber, the outlet temperature of the preheating chamber is the temperature at the smoke discharge location of the preheating chamber, the temperature response delay time is the time difference between the change in the preheating chamber wall temperature and the change in the outlet temperature, the smoke concentration in the cockpit refers to the concentration of smoke that has not been discharged in the cockpit, the exhaust fan speed is the rotation speed of the exhaust fan blades, and the exhaust duration is the continuous working time of the exhaust fan after it is started.

[0067] For example, in a practical application, when an airline's flight simulator conducts engine failure simulation training, the data acquisition process is triggered simultaneously with the activation of the smoke generator. The flow sensor in the smoke liquid supply pipeline continuously collects data within 10 seconds. At the 5th second, the instantaneous flow rate is 0.5 L / min. The cumulative consumption within the first 10 seconds is calculated by accumulating the instantaneous flow rate at each time point, i.e., the cumulative consumption. The pressure sensor collected a supply pressure of 0.3 MPa; the temperature sensor of the preheating chamber wall collected a real-time temperature of 85℃ at the 3rd second, 90℃ at the 5th second, and 95℃ at the 8th second; the outlet temperature of the preheating chamber was 70℃ at the 3rd second, 75℃ at the 6th second, and 88℃ at the 8th second; the temperature response delay time was recorded as 2 seconds for the chamber wall temperature to rise from 85℃ to 90℃ and 3 seconds for the outlet temperature to rise from 70℃ to 75℃, with the difference being 1 second; the concentration sensor in the cockpit collected a smoke concentration of 0.1 mg / m³; the exhaust fan speed sensor collected a speed of 1500 r / min; the timer recorded an exhaust duration of 20 seconds; and all collected data were aggregated to the control unit as input data for subsequent determination of the preheating status using a decision tree classification algorithm.

[0068] In this embodiment of the invention, step S101 enables the comprehensive capture of key data from the entire process of the smoke generator, from smoke liquid supply and preheating to smoke discharge. This avoids the limitations of data collection from a single system or a single indicator, and provides real basic data support for the subsequent extraction of preheating efficiency characteristics and accurate judgment of the preheating status, ensuring that subsequent analysis and control do not deviate from the actual operation of the smoke generator.

[0069] S102. Extract key feature parameters from the multi-dimensional data to reflect the preheating efficiency.

[0070] Among them, the key characteristic parameters refer to the core indicators extracted from multi-dimensional data that can directly reflect the preheating effect of the smoke generator. The key characteristic parameters include preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature. The preheating temperature difference is the difference between the outlet temperature of the preheating chamber and the real-time temperature of the chamber wall, which is used to reflect the heat transfer in the preheating chamber. The temperature rise rate is the rate of change of the real-time temperature of the preheating chamber wall over a continuous period of time, which is used to reflect the heating efficiency. The coordination coefficient between flow rate and temperature is the corresponding indicator of the instantaneous flow rate of the smoke liquid and the real-time temperature of the preheating chamber wall at the same moment, which is used to reflect the matching state between the supply of smoke liquid and the preheating heat.

[0071] In one specific implementation, step S102 includes:

[0072] Step 1021: Calculate the difference between the outlet temperature of the preheating chamber and the real-time temperature to obtain the preheating temperature difference.

[0073] In this embodiment, the preheating temperature difference is obtained by using a difference calculation method based on the outlet temperature of the preheating cavity and the real-time temperature of the preheating cavity wall. The formula used in the difference calculation method can be... ,in, To preheat the temperature difference, The outlet temperature of the preheating chamber. This is the real-time temperature of the preheating chamber wall.

[0074] For example, in practical applications, when an airline's flight simulator conducts engine failure simulation training, it extracts key feature parameters based on multi-dimensional data collected by S101. First, it extracts the preheating chamber outlet temperature of 88℃ and the chamber wall real-time temperature of 95℃ from the data, according to the formula... Calculate, where, To preheat the temperature difference, The outlet temperature of the preheating chamber. To calculate the real-time temperature of the preheating cavity wall, the data was substituted into the calculation. The negative sign indicates the state of heat transfer from the cavity wall to the outlet.

[0075] Step 1022: Perform differential calculation on the change of the real-time temperature in the continuous time series to obtain the temperature rise rate.

[0076] In this embodiment, real-time temperature data of the preheating cavity wall from a continuous time series is extracted from multi-dimensional data. The temperature change rate is calculated using a differential operation method to obtain the temperature rise rate, as shown in the formula: ,in, For the rate of temperature rise, It represents the temperature change over a continuous time interval. This represents the interval duration for the corresponding time range.

[0077] For example, based on the cavity wall temperature of 85°C at the 3rd second and 95°C at the 8th second, the temperature change is 10°C, the time interval is 5 seconds, and then according to the formula... Calculate the rate of temperature rise, where, For the rate of temperature rise, The change in temperature The time interval is calculated by substituting the data. .

[0078] Step 1023: Analyze the correspondence between the instantaneous flow rate and the real-time temperature in the multi-dimensional data at the same time to obtain the coordination coefficient between flow rate and temperature.

[0079] In this embodiment, instantaneous flow rate of the flue gas liquid and real-time temperature of the preheating chamber wall at the same moment are selected from multi-dimensional data. The ratio calculation method is used to analyze the correspondence between the two to obtain the synergy coefficient between flow rate and temperature. The formula is as follows: ,in, For the synergy coefficient, This refers to the instantaneous flow rate of the vapor liquid.

[0080] For example, based on the instantaneous flow rate of the vapor liquid at the 8th second of 0.5 L / min and the real-time temperature of the cavity wall at 95°C, and according to the formula... Calculate the synergy coefficient, where, For the synergy coefficient, The instantaneous flow rate of the vapor liquid was calculated by substituting the data. .

[0081] Step 1024: Integrate the preheating temperature difference, the temperature rise rate, and the synergy coefficient between the flow rate and temperature to obtain key characteristic parameters.

[0082] In this embodiment of the invention, by transforming scattered multi-dimensional data into core indicators that focus on preheating efficiency, the matching state of heat transfer, heating rate and material energy can be clearly presented, providing a targeted and intuitive analytical basis for subsequent accurate determination of the preheating state, and avoiding analytical bias caused by the complexity of the original data.

[0083] S103. Based on the key feature parameters, a decision tree classification algorithm integrated into the PLC is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state, and the state determination result is obtained.

[0084] Among them, the decision tree classification model refers to an intelligent classification tool that is pre-trained based on historical operating data and integrated into the PLC, and outputs classification results through layer-by-layer logical judgment.

[0085] In one specific implementation, step S103 includes:

[0086] Step 1031: Based on the key feature parameters, construct an input feature vector for decision tree classification. The key feature parameters include preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature.

[0087] The input feature vector refers to a data carrier formed by structuring and combining key feature parameters in a preset format to adapt to the input requirements of the decision tree classification algorithm; and the input feature vector includes at least: preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature.

[0088] In step 1031, the preheating temperature difference, temperature rise rate, and coordination coefficient between flow rate and temperature extracted in S102 are structurally integrated according to the input specifications preset by the decision tree classification algorithm. The data format is unified and arranged in a fixed order of "preheating temperature difference, temperature rise rate, and coordination coefficient between flow rate and temperature" to form an input feature vector that can be directly recognized and processed by the algorithm, ensuring that the transmission and parsing of feature parameters are accurate.

[0089] For example, in practical applications, when an airline's flight simulator conducts engine failure simulation training, it constructs an input feature vector based on the key feature parameters obtained from S102. Specifically, following the order of "preheating temperature difference, temperature rise rate, and coordination coefficient between flow rate and temperature", the preheating temperature difference of -7℃, the temperature rise rate of 2℃ / s, and the coordination coefficient between flow rate and temperature of 0.0053L / (min•℃) are structurally integrated and formed into an input feature vector after unifying the data format.

[0090] Step 1032: Input the input feature vector into the pre-generated decision tree classification model integrated into the PLC. Through the decision tree classification model, the following classification logic is executed sequentially:

[0091] Step a1: At the root node, compare the preheating temperature difference with the first threshold. If the first comparison result is that the preheating temperature difference is lower than the first threshold, then branch to the leaf node representing the not-ready state. If the first comparison result is that the preheating temperature difference is higher than or equal to the first threshold, then proceed to the next internal node.

[0092] Step a2: In the next internal node, compare the temperature rise rate with the second threshold: if the second comparison result is that the temperature rise rate is lower than the second threshold, then branch to the leaf node representing the transition state; if the second comparison result is that the temperature rise rate is higher than or equal to the second threshold, then go to the final internal node.

[0093] Step a3: At the final internal node, compare the coordination coefficient between the flow rate and temperature with the third threshold: if the coordination coefficient between the flow rate and temperature is lower than the third threshold, then branch to the leaf node representing the transition state; if the coordination coefficient between the flow rate and temperature is higher than or equal to the third threshold, then branch to the leaf node representing the stable and ready state.

[0094] Step a4: Determine the state corresponding to the reached leaf node as the state determination result.

[0095] In the above scheme, the first threshold, the second threshold, and the third threshold refer to the judgment criteria preset based on the historical operating data of the smoke generator and the scenario requirements. They are used to measure whether the preheating temperature difference, the temperature rise rate, and the coordination coefficient between flow rate and temperature have reached the corresponding state requirements. The root node refers to the initial judgment level of the decision tree classification model, the internal nodes are the subsequent progressive judgment levels, and the leaf nodes refer to the final state output results of each judgment path. The three together constitute the hierarchical judgment structure of the decision tree.

[0096] For example, the preset first threshold, second threshold, and third threshold are all derived from the historical stable operation data of this model of smoke generator. The first threshold is set to -10℃ to distinguish between the unready state and other states; the second threshold is set to 1.5℃ / s to distinguish between the heating rate standard of the transition state and the stable ready state; and the third threshold is set to 0.005L / (min•℃) to distinguish between the material-energy matching standard of the transition state and the stable ready state.

[0097] Then, the input feature vector is input into the decision tree classification model integrated into the PLC through the data transmission channel. After the model starts, the first threshold is called at the root node level. The absolute value of the preheating temperature difference of -7℃ is compared with the first threshold of -10℃. Since -7℃ is greater than -10℃, it is determined that the heat transfer meets the basic requirements and the process moves to the next internal node. At the next internal node level, the second threshold is called. The temperature rise rate of 2℃ / s is compared with the second threshold of 1.5℃ / s. Since 2℃ / s is greater than 1.5℃ / s, it is determined that the heating rate meets the standard and the process moves to the final internal node.

[0098] At the final internal node level, the third threshold is called, and the coordination coefficient 0.0053L / (min•℃) is compared with the third threshold 0.005L / (min•℃). Since 0.0053L / (min•℃) is greater than 0.005L / (min•℃), it is determined that the supply of smoke liquid and the preheating heat have reached a dynamic balance, and the branch is sent to the leaf node corresponding to the stable and ready state.

[0099] Finally, the PLC control program reads the status flag of the leaf node, determines that the status judgment result is a stable and ready state, stores the result in the control unit data area, and sends a status signal to step S104.

[0100] In this embodiment of the invention, the multi-feature hierarchical progressive judgment logic avoids misjudgment of the state caused by a single indicator, and relies on the decision tree classification algorithm integrated into the PLC to achieve rapid and accurate identification of the preheating state, providing a reliable state basis for the adaptive adjustment of subsequent control parameters, and ensuring that the control strategy is highly consistent with the actual preheating performance of the smoke generator.

[0101] S104. Based on the state determination result, a PID control algorithm is used to adaptively adjust the control parameters of the smoke generator to achieve smoke output control that matches the preheating state.

[0102] In one specific implementation, such as Figure 2 As shown, step S104 includes:

[0103] Step 1041: Based on the control parameter records in the historical training rounds, establish control mapping relationships corresponding to different preheating states.

[0104] Among them, the control mapping relationship refers to the set of corresponding rules that associate different preheating states with corresponding control parameter combinations, which is used to quickly match and adapt the initial control scheme.

[0105] In step 1041, a large amount of historical training rounds of operation data of the smoke generator are collected, the preheating state and corresponding control parameter records of each round are extracted, and the above data are classified and organized according to different preheating states. Then, the adaptation range of control parameters and the correlation between operation effect under various states are analyzed to establish the correspondence between each preheating state and the corresponding optimal control parameter combination, and finally the control mapping relationship is formed and stored in the control unit.

[0106] For example, collect a large amount of historical training data from smoke generators of the same model, classify and organize the corresponding control parameters according to the ready state, transition state, and stable ready state, establish control mapping relationships and store them.

[0107] Step 1042: Based on the control mapping relationship, determine the initial control parameters corresponding to the state determination result.

[0108] The initial control parameters refer to the initial supply rate of the aerosol liquid and the control setpoint of the preheating power determined based on the preheating state, which serve as the basis for parameter adjustment. The initial control parameters include the control setpoint of the initial supply rate of the aerosol liquid and the preheating power.

[0109] As a specific implementation method, step 1042 may specifically include the following steps: based on the state determination result, select the corresponding basic control parameters from the control mapping relationship; based on the current environmental condition parameters, perform environmental adaptability correction on the basic control parameters to obtain the parameters to be adjusted; based on the preset domain knowledge base, adjust the parameters to be adjusted to obtain the initial control parameters.

[0110] It should be understood that the current environmental condition parameters are the relevant parameters of the real-time environment in which the smoke generator operates or conducts training tasks. The types of parameters mainly revolve around the influence of the environment on the preheating effect of the smoke generator, the supply of smoke liquid, and the diffusion of smoke. The current environmental condition parameters may include ambient temperature, such as the real-time temperature in the cockpit or simulation training chamber. This parameter may also include parameters that affect the heat loss rate of the preheating chamber, ambient air pressure, power supply system parameters, such as the voltage stability of the input devices, parameters that affect the accuracy of preheating power output, ambient humidity, and the spatial volume and basic ventilation conditions if the equipment is in a confined space.

[0111] It should also be understood that the domain knowledge base refers to a database that stores the operating experience and adaptation rules of storage devices; the dynamic characteristic model of the supply system refers to a model that describes the operating rules of the aerosol supply system.

[0112] For example, the basic control parameters corresponding to the stable and ready state are selected from the mapping relationship. The initial supply rate of the aerosol liquid is 0.6L / min, and the preheating power control setting is 800W. Then, the room temperature is collected from the current environmental condition parameters as 25℃. The preset room temperature standard range in the domain knowledge base is 20℃ to 30℃. If the temperature exceeds the range, the supply rate is adjusted by 1% for every 5℃. The current room temperature is within the standard range and no adjustment is required. Finally, the initial control parameters are determined to be the initial supply rate of the aerosol liquid of 0.6L / min and the preheating power control setting of 800W.

[0113] Step 1043: Based on the initial control parameters, a composite adjustment mechanism including feedforward and feedback is adopted. Combining the heat accumulation index and the response coefficients of flow rate and temperature, the real-time adjustment amount of preheating power and the dynamic compensation amount of flue gas liquid supply rate are determined. Through the multivariate coordination mechanism of the PID control algorithm, the real-time adjustment amount and the dynamic compensation amount are converted into intermediate control parameters.

[0114] Among them, the heat accumulation index refers to the index that reflects the degree of heat accumulation in the preheating chamber, and the flow rate and temperature response coefficient refers to the index that reflects the efficiency of the interaction between the flow rate of the flue gas liquid and the preheating temperature.

[0115] Real-time adjustment refers to the dynamic adjustment value of preheating power, dynamic compensation refers to the adaptive correction value of the smoke liquid supply rate, the multivariate coordination mechanism of the PID control algorithm refers to the collaborative processing method to eliminate the mutual interference between the adjustment of preheating power and smoke liquid supply rate, and intermediate control parameters refer to the parameters used for preliminary control after feedforward feedback adjustment and collaborative processing.

[0116] As a specific implementation, step 1043 may include the following steps: based on the initial control parameters, a feedforward prediction mechanism is used to process the changing trend of the heat accumulation index to obtain the feedforward control quantity of the preheating power; based on the response coefficients of flow rate and temperature, a feedback correction mechanism is used to process the deviation between the current preheating state and the target preheating state to obtain the feedback compensation quantity of the preheating power; the feedforward control quantity and the feedback compensation quantity are weighted and fused to obtain the real-time adjustment quantity of the preheating power; based on the real-time adjustment quantity, the demand change of the smoke liquid supply rate is derived through a preset dynamic characteristic model of the supply system to obtain the dynamic compensation quantity of the smoke liquid supply rate; a multivariate coordination mechanism of the PID control algorithm is used to coordinate the real-time adjustment quantity and the dynamic compensation quantity to eliminate mutual interference between parameters, and intermediate control parameters are generated based on the coordination result.

[0117] Among them, the feedforward prediction mechanism refers to the adjustment method that predicts control demand in advance based on the trend of parameter changes, while the feedback correction mechanism refers to the adjustment method that corrects the control quantity in real time based on the deviation between the actual state and the target state.

[0118] For example, by activating the composite regulation mechanism, the trend of heat accumulation index changes is analyzed through the feedforward prediction mechanism, and it is predicted that the heat will remain stable, resulting in a feedforward control amount of 0W for preheating power; then, based on the response coefficients of flow rate and temperature, the deviation between the current preheating state and the target state is compared to 5%, and the feedback correction mechanism is used to calculate a preheating power feedback compensation amount of 40W.

[0119] Then, a weighted fusion was performed according to the ratio of 30% for the feedforward control quantity and 70% for the feedback compensation quantity, using the formula... ,in, This is the real-time adjustment amount of preheating power. This is the feedforward control variable. To obtain the feedback compensation amount, substitute the numerical values ​​for calculation. ;

[0120] Finally, based on the correlation between the real-time adjustment amount and the vapor liquid supply rate, the dynamic compensation amount of the vapor liquid supply rate of 0.03L / min was derived through the dynamic characteristic model of the supply system. The multivariate coordination mechanism of the PID control algorithm was activated to eliminate parameter interference, and the intermediate control parameters were generated as preheating power of 828W and vapor liquid supply rate of 0.63L / min.

[0121] Step 1044: Combining the preheating status evaluation data from historical training rounds, construct a target optimization function that includes the cumulative consumption of smoke liquid, preheating temperature stability, and residual smoke concentration. Based on the target optimization function, optimize the intermediate control parameters to obtain preliminary optimized parameters.

[0122] The objective optimization function refers to a comprehensive optimization tool that takes into account the cumulative consumption of flue gas liquid, the stability of preheating temperature, and the residual concentration of flue gas. The objective optimization function can be a weighted summation formula. The specific expression used in this application embodiment is not specifically limited, and can be set according to the actual situation. The preliminary optimization parameters refer to the optimization parameters obtained after processing by the objective optimization function.

[0123] For example, by collecting preheating status evaluation data from multiple consecutive historical training rounds, a target optimization function was constructed that takes into account the cumulative consumption of smoke liquid, the stability of preheating temperature, and the residual concentration of exhaust smoke. The intermediate control parameters were then substituted into the function for calculation, and the preliminary optimization parameters were obtained as follows: preheating power 820W and smoke liquid supply rate 0.62L / min.

[0124] Step 1045: Combining the smoke exhaust index and preheating temperature data, the preliminary optimization parameters are adaptively adjusted to obtain the adaptively adjusted control parameters.

[0125] The adaptively adjusted control parameters refer to the control parameters that are finally calibrated by combining the smoke exhaust index and the preheating temperature data, and are used to precisely regulate the operation of the equipment. The control parameters are used to synchronously adjust the valve opening of the smoke liquid supply pipeline and the heating power of the preheating chamber.

[0126] As a specific implementation method, step 1045 may specifically include the following steps: establishing an influence relationship model through cross-coupling analysis of smoke emission index and preheating temperature data; and adaptively adjusting the preliminary optimization parameters based on the influence relationship model to obtain the adaptively adjusted control parameters.

[0127] Among them, the influence relationship model refers to the model that reflects the mutual influence between the smoke emission index and the preheating temperature.

[0128] In step 1045, exhaust gas index and preheating temperature data during the current operation are collected, and cross-coupled analysis is performed on the two types of data to explore the correlation between exhaust gas index and preheating power, and between preheating temperature and smoke liquid supply rate. The trend of exhaust gas index changing with preheating power and the law of preheating temperature changing with smoke liquid supply rate are clarified, and an influence relationship model reflecting these correlations is established. Subsequently, based on the model, reasonable ranges for exhaust gas index and preheating temperature under different preheating states are preset. The analysis is conducted to determine whether the exhaust gas index and preheating temperature corresponding to the preliminary optimization parameters are within reasonable ranges, and the parameter adaptability is judged. If the exhaust gas index is too high and the preheating temperature is close to the upper limit, the preheating power is appropriately reduced. If the exhaust gas index is too low and the preheating temperature is insufficient, the preheating power and smoke liquid supply rate are moderately increased. Finally, the preliminary optimization parameters are adjusted adaptively according to the adaptability analysis results to obtain the adaptively adjusted control parameters, and the control parameters are transmitted to the execution unit for synchronously adjusting the valve opening of the smoke liquid supply pipeline and the heating power of the preheating chamber.

[0129] For example, smoke emission indicators and preheating temperature data are acquired through sensing devices. The preheating temperature data directly calls real-time data from the preheating chamber wall and outlet temperature sensors. At the 10th second, the preheating chamber wall temperature is collected at 98℃ and the preheating chamber outlet temperature is collected at 92℃. The smoke emission indicators are then derived by combining multiple types of acquired data. Specifically:

[0130] The real-time exhaust fan speed sensor collects data at 1600 r / min, corresponding to the equipment's inherent rated exhaust volume of 0.5 m³ / s. The concentration sensor in the control cabin collects initial smoke concentration of 0.8 mg / m³ and smoke concentration at 10 seconds of induction at 0.2 mg / m³. A timer records the exhaust duration as 10 seconds. The control cabin volume is a preset fixed parameter of 10 m³. Then, according to the formula... Calculate the smoke emission index, among which, For smoke emission indicators, The initial smoke concentration, The real-time residual concentration is given by V, the cockpit volume is given by Q, the rated exhaust volume of the exhaust fan is given by t, and the exhaust duration is given by substituting the values. .

[0131] Subsequently, based on the smoke emission index and the preheating temperature, an influence relationship model was constructed. The core logic of the model is that the smoke emission index is positively correlated with the preheating power and the preheating temperature is positively correlated with the smoke liquid supply rate. The model presets a reasonable range of 1.0-1.3 for the smoke emission index and a reasonable range of 95-100℃ for the preheating chamber wall temperature under stable and ready conditions. Currently, the smoke emission index of 1.2 is at the upper limit of the reasonable range and the preheating chamber wall temperature of 98℃ is close to the upper limit of the reasonable range.

[0132] Therefore, based on this model, the initial optimization parameters were adaptively adjusted, that is, the preheating power was appropriately reduced to avoid excessive temperature, and the smoke liquid supply rate was finely adjusted to match the heat change. The final adaptively adjusted control parameters were 815W preheating power and 0.61L / min smoke liquid supply rate. These parameters were transmitted to the execution unit to synchronously adjust the valve opening of the smoke liquid supply pipeline and the heating power of the preheating chamber to ensure that the smoke output concentration and duration meet the requirements of engine fault simulation training. The equipment operation data will be continuously monitored to provide basic data support for the optimization of control parameters in the next round of training.

[0133] In this embodiment of the invention, dynamic adaptation between control parameters and the preheating state of the smoke generator is achieved through multi-stage parameter adjustment and optimization, effectively eliminating mutual interference between parameters, and taking into account multiple operating objectives to ensure the stability and adaptability of smoke output, thereby meeting the precise requirements for smoke effect in different scenarios.

[0134] Figure 3 This is a schematic diagram illustrating a specific implementation of an intelligent control system for a smoke generator based on preheating state recognition, provided in this application. (Refer to...) Figure 3 The system may include:

[0135] The acquisition module 31 is used to acquire multi-dimensional data of the smoke generator and its surrounding environment.

[0136] Extraction module 32 is used to extract key feature parameters reflecting preheating efficiency from the multi-dimensional data.

[0137] The determination module 33 is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state based on the key feature parameters and using a decision tree classification algorithm integrated into the PLC, and obtain the state determination result.

[0138] The adjustment module 34 is used to adaptively adjust the control parameters of the smoke generator based on the state determination result and using a PID control algorithm to achieve smoke output control that matches the preheating state.

[0139] The intelligent control system for a smoke generator based on preheating state recognition in this application is used to implement the aforementioned intelligent control method for a smoke generator based on preheating state recognition. Therefore, the specific implementation of the intelligent control system for a smoke generator based on preheating state recognition can be found in the embodiment section of the intelligent control method for a smoke generator based on preheating state recognition described above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0140] like Figure 4As shown, this application also provides an electronic device, including: a memory 41 for storing a computer program; and a processor 42 for executing the computer program to implement the steps of the above-described intelligent control method for a smoke generator based on preheating state recognition.

[0141] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described intelligent control methods for a smoke generator based on preheating state recognition.

[0142] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0143] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the embodiments of the intelligent control method for a smoke generator based on preheating state recognition.

[0144] Those skilled in the art will further 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, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. 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.

[0145] The above provides a detailed description of the intelligent control method, system, electronic device, and storage medium for a smoke generator based on preheating state recognition provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A smart control method for a smoke generator based on preheating state recognition, characterized in that, include: Acquire multi-dimensional data of the smoke generator and its surrounding environment, including: instantaneous flow rate, cumulative consumption and supply pressure of the smoke liquid supply pipeline, real-time temperature of the preheating chamber wall, outlet temperature of the preheating chamber, smoke concentration in the cockpit, exhaust fan speed and exhaust duration, and temperature response delay time. Key feature parameters reflecting preheating efficiency are extracted from the multi-dimensional data. Based on the key feature parameters, a decision tree classification algorithm integrated into the PLC is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state, and the state determination result is obtained. Based on the state determination result, a PID control algorithm is used to adaptively adjust the control parameters of the smoke generator in order to achieve smoke output control that matches the preheating state. Based on the state determination result, the control parameters of the smoke generator are adaptively adjusted using a PID control algorithm, including: Based on the control parameter records in historical training rounds, establish control mapping relationships corresponding to different preheating states; Based on the control mapping relationship, determine the initial control parameters corresponding to the state determination result; Based on the initial control parameters, a composite adjustment mechanism including feedforward and feedback is adopted. Combining the heat accumulation index and the response coefficients of flow rate and temperature, the real-time adjustment amount of preheating power and the dynamic compensation amount of flue gas supply rate are determined. Through the multivariate coordination mechanism of PID control algorithm, the real-time adjustment amount and the dynamic compensation amount are converted into intermediate control parameters. Based on the preheating status evaluation data from historical training rounds, a target optimization function is constructed that includes the cumulative consumption of smoke liquid, the stability of preheating temperature, and the residual concentration of exhaust smoke. Based on the target optimization function, the intermediate control parameters are optimized to obtain preliminary optimized parameters. By combining the smoke emission index and preheating temperature data, the preliminary optimization parameters are adaptively adjusted to obtain the adaptively adjusted control parameters; Based on the key feature parameters, a decision tree classification algorithm integrated into the PLC is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state, and the state determination result is obtained, including: Based on the key feature parameters, an input feature vector for decision tree classification is constructed. The key feature parameters include preheating temperature difference, temperature rise rate, and the coordination coefficient between flow rate and temperature. The input feature vector is fed into a pre-generated decision tree classification model integrated into the PLC. The following classification logic is then executed sequentially through the decision tree classification model: At the root node, the preheating temperature difference is compared with a first threshold. If the first comparison result is that the preheating temperature difference is lower than the first threshold, the branch goes to the leaf node representing the not-ready state. If the first comparison result is that the preheating temperature difference is higher than or equal to the first threshold, the branch goes to the next internal node. In the next internal node, compare the temperature rise rate with the second threshold: if the second comparison result is that the temperature rise rate is lower than the second threshold, then branch to the leaf node representing the transition state; if the second comparison result is that the temperature rise rate is higher than or equal to the second threshold, then go to the final internal node. At the final internal node, compare the coordination coefficient between the flow rate and temperature with a third threshold: if the coordination coefficient between the flow rate and temperature is lower than the third threshold, then branch to the leaf node representing the transition state; if the coordination coefficient between the flow rate and temperature is higher than or equal to the third threshold, then branch to the leaf node representing the stable and ready state. The state corresponding to the reached leaf node is determined as the state determination result.

2. The intelligent control method for a smoke generator based on preheating state recognition according to claim 1, characterized in that, The step of determining the initial control parameters corresponding to the state determination result based on the control mapping relationship includes: Based on the state determination result, the corresponding basic control parameters are selected from the control mapping relationship; Based on the current environmental conditions, the basic control parameters are modified to adapt to environmental conditions, resulting in the parameters to be adjusted. Based on a preset domain knowledge base, the parameters to be adjusted are adjusted to obtain initial control parameters.

3. The intelligent control method for a smoke generator based on preheating state recognition according to claim 1, characterized in that, Based on the initial control parameters, a composite adjustment mechanism including feedforward and feedback is adopted. Combining the heat accumulation index and the response coefficients of flow rate and temperature, the real-time adjustment of preheating power and the dynamic compensation of the flue gas supply rate are determined. Through the multivariate coordination mechanism of the PID control algorithm, the real-time adjustment and the dynamic compensation are converted into intermediate control parameters, including: Based on the initial control parameters, a feedforward prediction mechanism is used to process the changing trend of the heat accumulation index to obtain the feedforward control quantity of the preheating power. Based on the response coefficients of flow rate and temperature, the deviation between the current preheating state and the target preheating state is processed through a feedback correction mechanism to obtain the feedback compensation amount of preheating power. The feedforward control quantity and the feedback compensation quantity are weighted and fused to obtain the real-time adjustment quantity of the preheating power; Based on the real-time adjustment amount, the dynamic compensation amount of the e-liquid supply rate is obtained by deriving the demand change of the e-liquid supply rate through the preset dynamic characteristic model of the supply system. A multivariate coordination mechanism using a PID control algorithm is employed to coordinate the real-time adjustment and dynamic compensation quantities to eliminate mutual interference between parameters. Based on the results of the coordinated processing, intermediate control parameters are generated.

4. The intelligent control method for a smoke generator based on preheating state recognition according to claim 1, characterized in that, The process involves combining smoke emission index and preheating temperature data to adaptively adjust the initial optimization parameters, resulting in adaptively adjusted control parameters, including: By cross-coupling analysis of smoke emission indicators and preheating temperature data, an influence relationship model was established; Based on the aforementioned influence relationship model, the initial optimization parameters are adaptively adjusted to obtain the adaptively adjusted control parameters.

5. The intelligent control method for a smoke generator based on preheating state recognition according to claim 1, characterized in that, The key feature parameters extracted from the multi-dimensional data to reflect preheating efficiency include: Calculate the difference between the outlet temperature of the preheating chamber and the real-time temperature to obtain the preheating temperature difference; The rate of temperature rise is obtained by performing a differential operation on the change of the real-time temperature in a continuous time series. The correspondence between instantaneous flow rate and real-time temperature in the multi-dimensional data at the same time is analyzed to obtain the coordination coefficient between flow rate and temperature; By integrating the preheating temperature difference, the temperature rise rate, and the synergy coefficient between the flow rate and temperature, key characteristic parameters are obtained.

6. An intelligent control system for a smoke generator based on preheating state recognition, characterized in that, A method for intelligent control of a smoke generator based on preheating state recognition as described in any one of claims 1 to 5 includes: The acquisition module is used to acquire multi-dimensional data of the smoke generator and its surrounding environment. The multi-dimensional data includes: instantaneous flow rate, cumulative consumption and supply pressure of the smoke liquid supply pipeline, real-time temperature of the preheating chamber wall, outlet temperature of the preheating chamber, smoke concentration in the cockpit, exhaust fan speed and exhaust time, and temperature response delay time. The extraction module is used to extract key feature parameters reflecting preheating efficiency from the multi-dimensional data; The determination module is used to determine whether the current preheating state of the smoke generator is in an unready state, a transitional state, or a stable ready state based on the key feature parameters and using a decision tree classification algorithm integrated into the PLC, and obtain the state determination result. The adjustment module is used to adaptively adjust the control parameters of the smoke generator based on the state determination result and using a PID control algorithm to achieve smoke output control that matches the preheating state.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the intelligent control method for a smoke generator based on preheating state recognition as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of the intelligent control method for a smoke generator based on preheating state recognition as described in any one of claims 1 to 5.

Images