Integrated fire-fighting emergency linkage control system

Abstract

The invention discloses an integrated fire-fighting emergency linkage control system which comprises a data acquisition module, an evaluation analysis module, a linkage strengthening module and a feedback training module. The data acquisition module acquires operation data of a fire control room, an automatic fire alarm system and a linkage subsystem in real time through a multi-type sensor and a protocol conversion engine, and ensures that the time synchronization error is less than or equal to 1 microsecond; a four-dimensional evaluation model is built in the evaluation analysis module, the timeliness, integrity, compliance and normalization of operation behaviors are quantitatively evaluated by combining an analytic hierarchy process, and operation conflicts are processed through a rule engine; the linkage strengthening module executes control such as fire alarm timeout automatic alarm and misoperation locking based on the evaluation result; the feedback training module pushes personalized training content by means of digital twinning replay operation, the system forms closed-loop logic, and normalization and safety of fire emergency operation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire control systems, more particularly, it relates to an integrated fire emergency linkage control system. BACKGROUND

[0002] The fire emergency linkage control system is a core technology system for ensuring building fire safety. By integrating fire automatic alarm, fire equipment control, emergency evacuation guidance and other functions, it realizes the integrated management of fire monitoring, alarm, equipment linkage and emergency disposal. Its core function is to quickly respond to alarm signals when a fire occurs, and to start the fire alarm, smoke control and automatic fire extinguishing systems, and to assist the fire control room personnel in completing the standard operation to minimize fire loss.

[0003] The standardization of fire emergency operation directly affects the efficiency and safety of fire disposal. However, in the prior art, operation evaluation relies on manual post-checking, and only qualitative description can be used to determine whether the operation is completed. At the same time, even if operation defects are found, there is a lack of targeted improvement mechanism based on evaluation results. This defect makes it difficult to accurately locate operation problems, and personnel capabilities cannot be continuously optimized, which may eventually delay fire disposal due to non-standard operation and increase safety risks.

[0004] Therefore, in order to solve the above technical problems, the present application provides an integrated fire emergency linkage control system. SUMMARY

[0005] In view of the deficiencies in the prior art, the purpose of the present application is to provide an integrated fire emergency linkage control system.

[0006] To achieve the above purpose, the present application provides the following technical scheme: an integrated fire emergency linkage control system, comprising:

[0007] A data acquisition module is configured to acquire operation data of fire control room operation equipment, a fire automatic alarm system and a linkage subsystem in real time.

[0008] An evaluation and analysis module is connected to the data acquisition module and has a four-dimensional evaluation model built-in to analyze the timeliness, correctness, integrity and standardization of operation behavior.

[0009] A linkage reinforcement module sends control instructions to fire equipment based on evaluation results.

[0010] A feedback training module generates an operation defect report and a targeted training strategy.

[0011] Preferably, the four-dimensional evaluation model comprises:

[0012] A timeliness analysis unit calculates the deviation coefficient of the actual response time and the standard threshold.

[0013] Integrity analysis unit, verifying operation missing rate by SOP step matcher;

[0014] Compliance analysis unit, verifying operation sequence logic based on safety interlock rules;

[0015] Normative analysis unit, detecting two-man operation compliance and operation record integrity.

[0016] Preferably, the data acquisition module comprises:

[0017] High-precision pressure sensor and acceleration sensor integrated in the manual alarm button to collect the pressing force variation curve and triggering acceleration;

[0018] Optical sensor and capacitive touch sensor deployed on the fire control console to record the operation trajectory and touch duration;

[0019] IEEE1588 protocol clock synchronization device to ensure that the cross-system event timestamp error is ≤1μs.

[0020] Preferably, the evaluation analysis module has a built-in rule engine that executes according to the following rules:

[0021] A: Concurrently running the timeliness analyzer, SOP matcher, safety interlock verifier, and two-man operation detector;

[0022] B: When detecting operation conflicts, starting the following three-level arbitration, preferentially using timestamped physical sensor data, secondarily using digitally signed control instructions, and finally fusing multi-channel video analysis results to generate optimal estimates.

[0023] Preferably, the rule engine supports: dynamic rule hot loading mechanism to automatically import new rule library when fire safety specifications are updated; fuzzy rule inference unit to distinguish between operation errors and equipment abnormalities based on equipment fault data.

[0024] Preferably, the data acquisition module parses OPCUA, ModbusTCP, and BACnet protocols through a special protocol conversion engine and fills in data missing due to network jitter based on an LSTM time series prediction model.

[0025] Preferably, when the fire alarm confirmation times out for 30 seconds, the linkage reinforcement module automatically triggers a 119 alarm call, and when detecting misoperation during the gas fire extinguishing system delay phase, the linkage reinforcement module immediately locks the emergency stop button and requires supervisor permission to unlock.

[0026] Preferably, the integrated fire emergency linkage control system further comprises:

[0027] Privacy desensitization engine to convert operator names to worker ID hash values based on RBAC dynamic desensitization strategies;

[0028] The data tamper-proofing unit generates operation chain data fingerprints through Merkle tree generation operation and stores them in a chain for evidence.

[0029] Preferably, the feedback training module replays operation events in the digital twin environment, verifies the consistency of device actions and instructions, and pushes personalized training content to the fire control room terminal according to the evaluation matrix output results.

[0030] Preferably, the linkage subsystem comprises:

[0031] Fire alarm system, sound and light alarm, emergency lighting and evacuation indication system, smoke control system, automatic sprinkler system and gas fire extinguishing system;

[0032] The operation data includes button pressing duration, device startup sequence timing, error attempt number and environmental feedback data consistency.

[0033] Compared with the prior art, the present application has the following beneficial effects:

[0034] 1. Through the four-dimensional evaluation model combined with the analytic hierarchy process, the operation behavior is multi-dimensionally quantified and evaluated, specific calculation formulas and rule engines are introduced, the whole process objective analysis from data acquisition to result output is realized, and subjective judgment deviation is avoided.

[0035] 2. The linkage reinforcement module executes automatic triggering of 119 alarm, locking of emergency stop button and other intervention measures for scenes such as fire alarm confirmation timeout and gas fire extinguishing system misoperation, reduces the risk of human delay and misoperation, and ensures the timeliness of emergency disposal.

[0036] 3. The feedback training module uses the digital twin environment to replay operation events, pushes personalized training content combined with evaluation results, forms a complete closed loop, and continuously optimizes the fire emergency operation capability. BRIEF DESCRIPTION OF DRAWINGS

[0037] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0038] Figure 1 The figure is a whole work flow diagram of the system of the present application;

[0039] Figure 2 The figure is a flow diagram of the evaluation and analysis module in the present application. DETAILED DESCRIPTION

[0040] As shown in the figure, the present application provides an integrated fire emergency linkage control system, which comprises:

[0041] A data acquisition module is configured to acquire operation data of the fire control room operation equipment, the fire automatic alarm system and the linkage subsystem in real time. The data acquisition module is disposed in the fire control room and each linkage equipment node, and data is acquired by means of combination of hardware sensors and software interfaces. For example, a pressure sensor and an acceleration sensor are installed at a manual alarm button to record the pressing time length and the force variation curve in real time.

[0042] The device operation log is acquired through the API interface of the fire host complying with the OPCUA protocol, including operation time, device ID, operation type and the like. The operation sequence is recorded through the console interface burying point technology, that is, the path from the selection of the region to the adjustment of the volume when starting the fire broadcast, and is matched with the video monitoring timestamp to associate the operator identity.

[0043] An evaluation analysis module is connected to the data acquisition module, and a built-in four-dimensional evaluation model is directly designed based on the user operation evaluation model in the scheme. For example, when a fire alarm signal is received, the model automatically calculates the timeliness, that is, the deviation of the actual confirmation time from the 30-second standard value, the integrity, that is, whether the whole step from confirming the fire alarm to starting the broadcast to starting the smoke exhaust is completed, the correctness, that is, whether the broadcast region matches the fire layer, and the standardization, that is, whether the double-person operation is performed.

[0044] A linkage reinforcement module triggers a control instruction based on the real-time result of the evaluation analysis module. For example, if the evaluation finds that the fire alarm confirmation is overdue by 30 seconds, the 119 automatic dialing is triggered immediately through the relay control module, and the alarm position is pushed to the fire command platform synchronously. If the gas fire extinguishing system is misoperated, the equipment is locked through the linkage control cabinet and an audible and visual alarm is issued.

[0045] A feedback training module generates an operation defect report and a targeted training strategy, counts high-frequency errors (such as the error rate of the smoke exhaust valve starting sequence is 30%), and designs to push the training content.

[0046] The four-dimensional evaluation model includes a timeliness analysis unit: a dynamic threshold-trend prediction composite model

[0047] Time series analysis and scene adaptive threshold are combined to realize multi-dimensional timeliness evaluation.

[0048] Dynamic threshold generation mechanism

[0049] A hierarchical response threshold library is constructed based on the fire development stage (initial stage / propagation stage / intense stage).

[0050] Initial fire (smoke sensing alarm ≤ 30s): the "basic threshold + environmental correction factor" is adopted, the basic threshold is set to 30s, the correction factor is related to the environmental temperature and humidity (such as the threshold is lowered by 10% in high-temperature environment), and the formula is:

[0051] T 动态= T 基础 x(1-αxΔT+βxΔH)

[0052] Wherein, α, β are correction coefficient, ΔT is environmental temperature difference, ΔH is temperature deviation.

[0053] Wherein, the spread / massive period: through the fire spread rate (based on the temperature gradient of infrared sensor) dynamic compression threshold, using Logistic growth model to predict threshold decay curve:

[0054]

[0055] Wherein, k is the fire spread coefficient, t0 is the fire determination time.

[0056] Through trend prediction and anomaly detection, sliding window-exponential smoothing method is introduced to process response time series:

[0057] Sliding window (window size = 5 operation periods) real-time calculation of response time mean and variance, identification of sudden delay;

[0058] Exponential smoothing (smoothing coefficient is 0.7) to predict the next period response time trend, the formula is:

[0059] S t = γx t +(1-γ)xs t-1

[0060] Wherein, S t is the smoothing value at t time, x t is the actual response time.

[0061] Here it needs to be explained that the "normal-delay-severe delay" three-state transition model is constructed combined with Markov chain, the probability distribution of continuous delay is calculated, and the systematic timeliness defect is warned in advance.

[0062] Integrity analysis unit: SOP topology matching-fuzzy association model, using "operation topology structure matching degree" evaluation, introducing the complex association between graph theory and fuzzy logic processing steps, and analyzing according to the following process

[0063] Step one: SOP topology modeling

[0064] The standard operation process (SOP) is disassembled into three levels of nodes:

[0065] Core node (such as "fire alarm confirmation" "start smoke exhaust"): indispensable, weight ratio 60%;

[0066] Auxiliary node (such as "notify the duty officer" "record alarm time"): optional but affects efficiency, weight ratio 30%;

[0067] Check node (e.g. "secondary confirmation device status"): fault-tolerant node, weight ratio 10%;

[0068] Directed acyclic graph (DAG) is constructed based on node dependency relationship, using adjacency matrix A n×n to represent the precedence constraints between steps (A ij = 1 means that step i must be executed before step j);

[0069] Step two: matching degree calculation

[0070] Floyd-Warshall algorithm is used to calculate the longest common path length between the actual operation sequence and the DAG, as the basis for structural matching:

[0071]

[0072] Introduce fuzzy Petri net to handle the uncertainty association between steps (e.g. "part of the auxiliary steps can be executed in parallel"):

[0073] The library represents the step state (not executed / execution / finished), and the transition represents the step trigger condition;

[0074] The membership function (such as triangular membership) is used to describe the "completeness degree" of step execution (such as "80% completion degree" is considered partially effective), and finally the comprehensive integrity score is output through the token flow rule;

[0075] Compliance analysis unit: temporal logic-fault tree composite verification model

[0076] Using "dynamic timing constraints + risk quantification" evaluation, combining formal verification and reliability analysis;

[0077] Temporal logic modeling:

[0078] Based on CTL* (Computational Tree Logic Extension) to describe the space-time constraints of operation sequence, for example:

[0079] Safety constraints: (In any case, the smoke exhaust valve must be opened before starting the sprinkler);

[0080] Activity constraints: (After the fire alarm is confirmed, the next operation must be to start the broadcast);

[0081] Using model checking tool SPIN to convert the actual operation sequence into a finite state machine, automatically verifying whether it meets the CTL* formula, and outputting the violation path (such as "starting the sprinkler directly without opening the smoke exhaust valve").

[0082] Risk quantification association

[0083] Combining Fault Tree Analysis (FTA) to associate compliance defects with potential risks:

[0084] Top event set as "fire loss expansion", intermediate event as "operation violation" (such as sequence error, insufficient authority), and bottom event as specific violation behavior;

[0085] Calculate the risk contribution of each violation operation using the minimum cut set method:

[0086] R i = P i × S i × C i

[0087] P i is the probability of violation, S i is the severity of consequences, and C i is the detection difficulty.

[0088] Final compliance score = basic compliance score × (1 - Σ risk contribution), realizing the linkage evaluation of "compliance-risk".

[0089] Normative analysis unit: behavior characteristic spectrum-formal verification model

[0090] Using "operation behavior characteristic spectrum matching + record formal verification", introducing behavior dynamics and formal methods.

[0091] Operation behavior characteristic spectrum analysis

[0092] Based on multi-sensor data to construct operator behavior baseline:

[0093] Pressure sensor: slope of pressure change curve

[0094] (K p = ΔF / Δt)

[0095] Optical sensor: curvature of operation trajectory (C = 1 / R, R is the trajectory radius), node dwell time (t stop );

[0096] Touch sensor: standard deviation of touch duration (σ t ).

[0097] Using Dynamic Time Warping (DTW) algorithm to calculate the similarity between actual operation characteristics and baseline:

[0098]

[0099] (X is the actual feature sequence, Y is the baseline sequence), similarity < 70% is determined as non-standard.

[0100] Record integrity formal verification

[0101] Based on the formal specification of the record defined in Z language (such as "the operation record must contain a timestamp, operator ID, device state code"), for example:

[0102] Record structure: Record::=(t:Time, id:WorkerID, s:DeviceState)

[0103] Constraints: (the record must be generated within 10 minutes after the event ends).

[0104] Combined with the time sequence evidence of the blockchain (Merkle tree hash chain), the continuity of the record is ensured by verifying the hash association H(r i+1 )=H(H(r i )+r i+1 ), and the final output integrity verification report is ensured.

[0105] It should be noted that the four-dimensional evaluation model uses the analytic hierarchy process (AHP) to construct a hierarchical structure, including the target layer: comprehensive evaluation of user operation correctness; the criterion layer: timeliness, integrity, compliance and standardization; the index layer: subdivided sub-dimension indicators, using the 1-9 scale method to compare two indicators at the same level, and determining the relative weights between the indicators at each level by expert scoring and pairwise comparison, constructing a judgment matrix, and then processing the constructed judgment matrix through the following steps:

[0106] Step one: calculate the product M of each row element of the matrix;

[0107] Step two: calculate the Xth root of the product M, X is determined according to the number of indicators at each level, for example, if there are four indicators at the criterion layer, then X=4, and then get the vector W;

[0108] Step three: normalize the vector W to get the weight vector W1;

[0109] In the subsequent score calculation, first, based on the original score of each sub-index of the index layer, the score is obtained by comparing the actual operation data with the preset standard threshold, such as the fire alarm confirmation delay score calculated according to the deviation proportion of the actual time and the standard time of 30 seconds, and the operation sequence order score determined according to the deduction value corresponding to the violation point number, combined with the index layer weight obtained by the analytic hierarchy process, such as the fire alarm confirmation delay accounting for 60% of the timeliness criterion weight, and the smoke exhaust start delay accounting for 40%, the score of each criterion layer is calculated, for example, timeliness score = fire alarm confirmation delay score x 60% + smoke exhaust start delay score x 40%, and the integrity, compliance and standardization scores are calculated according to the same logic; then, the scores of the four criterion layers are multiplied by their corresponding criterion layer weights, such as compliance weight 45%, timeliness weight 25%, integrity weight 15%, and standardization weight 15%, and the product results are summed to obtain the comprehensive score of the target layer, that is, the final score = timeliness score x 25% + integrity score x 15% + compliance score x 45% + standardization score x 15%, which directly reflects the correctness of the user's operation behavior, for example, when the comprehensive score is 82.425, it is determined as good level, and the monthly intensive training mechanism is triggered; when the comprehensive score is 62.175, it is determined as risk level, and the off-duty training and system review process is triggered.

[0110] The data acquisition module includes: high-precision pressure sensor and acceleration sensor integrated in the manual alarm button to collect the pressure change curve and trigger acceleration; optical sensor and capacitive touch sensor deployed on the fire control console to record the operation trajectory and touch duration; IEEE1588 protocol clock synchronization device to ensure that the cross-system event timestamp error is less than or equal to 1 μs; the piezoresistive type pressure sensor outputs the pressure change curve in real time, for example, the time from 0N to 5N should be less than or equal to 0.3 seconds, otherwise it is determined as operation hesitation;

[0111] It should be noted that in the fire emergency, operation hesitation is a key and unique phenomenon, which can be defined as: in the fire emergency action, due to the complexity of the scene environment, the uncertainty of the information, the urgency of the task and the psychological pressure of the individual, etc. Various factors are intertwined, resulting in hesitation, decision delay or repeated weighing behavior state of the fire personnel in the execution of key operations (such as selection of fire fighting tactics, rescue path, start of fire fighting equipment, etc.).

[0112] In order to quantify the operation hesitation, the operation hesitation coefficient H is introduced, and its calculation formula is:

[0113]

[0114] Where, T ab±a ; is the hesitation time of the fire personnel in a certain operation decision process;

[0115] is the total length from obtaining decision information to making the final decision;

[0116] n is the number of factors affecting operation hesitation, such as on-site smoke concentration F1, noise interference degree F2, fire spread speed F3, etc.

[0117] w i : is the weight of each factor, which is determined according to the research of actual fire fighting scene and expert experience. For example, in a general building fire scene, the on-site smoke concentration has a greater impact on operation hesitation, and its weight w1 may be set to 0.4, the noise interference degree weight w2 is set to 0.2, and the fire spread speed weight w3 is set to 0.3, etc. (The specific weight needs to be continuously optimized according to a large number of actual cases and simulation experiments). The greater the value of the operation hesitation coefficient H, the higher the degree of hesitation of the fire fighter in the operation decision, which may have a more adverse impact on the fire emergency effect.

[0118] In the fire training scene, a training link specifically for operation hesitation is set. For example, a training facility simulating a complex fire scene is constructed, information interference sources (such as false alarms, incorrect fire information markers, etc.) are randomly set in the scene, and the decision time and hesitation behavior of the fire fighters when facing different tasks are recorded through monitoring equipment. According to the calculation results of the operation hesitation coefficient H, the operation hesitation characteristics of each fire fighter under different scene factors are analyzed to provide a basis for individualized training. For personnel with a higher operation hesitation coefficient, targeted psychological stress resistance training, rapid decision-making ability training, and complex scene information screening and processing ability training are strengthened.

[0119] The acceleration sensor adopts MEMS type, and the trigger acceleration threshold is set to 2g. If it is lower than the value, it is determined as a false touch; the optical sensor records the operation trajectory through infrared imaging, such as the path of clicking the smoke valve icon first and then clicking the start button; the capacitive touch sensor has a response time ≤1ms, and records the touch duration, such as pressing the reset button for ≥2 seconds to determine as valid operation; the clock synchronization device adopts IEEE1588PTP protocol, and realizes time synchronization of the fire host, sensor and video monitoring through an independent clock server, ensures that the timestamp error of the detector alarm time and the control console confirmation time is ≤1μs, and avoids evaluation errors caused by time deviation;

[0120] The evaluation and analysis module has a built-in rule engine, which is executed according to the following rules:

[0121] A: Parallel running of time effectiveness analyzer, SOP matcher, safety interlock verifier and double-person operation detector;

[0122] B: When operation conflict is detected, start the following three-level arbitration, prefer to use the physical sensor data with timestamp, secondly the digital signature control instruction, finally fuse the multi-channel video analysis results to generate the optimal estimate;

[0123] Time-effectiveness analyzer: Real-time calculation of the delay of each operation node, such as the time consumption from fire alarm confirmation to broadcast start;

[0124] SOP matcher: Compare the consistency of operation steps;

[0125] Safety interlock verifier: Verify the logic that the smoke control system needs to close the air conditioner before starting the smoke exhaust, etc;

[0126] Two-person operation detector: Confirm the operator's identity through face recognition by camera, and match the two persons' ID numbers;

[0127] When data conflict occurs, for example, the fire host log shows that the sprinkler has been started, but the sensor feedback shows that the valve is not open, the following priority is determined:

[0128] Prefer to use the physical sensor data with timestamp, such as the reading of the valve pressure sensor;

[0129] Secondly, the digital signature control instruction generated by the fire host needs to include the operator ID and the encrypted timestamp;

[0130] Finally, fuse more than three video analysis results, such as the picture of the operator not clicking the sprinkler start button, to generate the optimal estimate;

[0131] Rule engine support: Dynamic rule hot loading mechanism, automatically import new rule library when the fire code is updated; Fuzzy rule inference unit, associate device fault data to distinguish between operation errors and device abnormalities, the system has built-in GB50116, NFPA72, etc. Rule base, when the specification is updated, for example, the fire alarm confirmation time is shortened from 30 seconds to 20 seconds, new rules can be automatically imported through a USB flash disk or a remote server, ensuring that the evaluation standard is consistent with the latest specification, and associating with the device fault database, such as the record of the smoke exhaust fan jamming in the past 1 year, when the smoke exhaust start fails, through the inference model, if the fan current > 30% of the rated value, it is determined as a device fault rather than an operation error;

[0132] The data acquisition module parses OPCUA, ModbusTCP and BACnet protocols through a special protocol conversion engine, fills in the missing data caused by network jitter based on an LSTM time series prediction model, supports parsing of 12 kinds of industrial protocols, wherein: the OPCUA protocol is used for real-time data interaction between the fire host and the evaluation system; the ModbusTCP protocol is used for reading the valve state of the smoke control system; the BACnet protocol is used for collecting voltage and current data of the emergency lighting system; when network jitter causes data packet loss, an LSTM model is used, the model architecture is 3 layers of hidden layers and 64 neurons, the missing values are predicted based on the operation sequence of the previous 50 ms, and the data integrity after filling in;

[0133] When the fire alarm confirmation timeout is 30 seconds, the linkage reinforcement module automatically triggers 119 alarm call, and when the linkage reinforcement module detects misoperation in the delay stage of the gas fire extinguishing system, the linkage reinforcement module immediately locks the emergency stop button and needs the authority of the supervisor to unlock, if the evaluation analysis module determines that the fire alarm confirmation time is greater than 30 seconds, the linkage reinforcement module automatically dials 119 through the GSM module, and sends a message containing the fire location and the device state to the fire command center;

[0134] In the 30-second delay stage, if the pressure sensor of the emergency stop button detects non-standard pressing, for example, the force is greater than 5N and lasts less than 0.5 seconds, the linkage reinforcement module immediately locks the button through the electromagnetic lock, and triggers the instruction that needs the supervisor IC card to unlock, to avoid gas misfire;

[0135] The integrated fire emergency linkage control system further includes: a privacy desensitization engine that converts operator names to worker number hash values based on a RBAC dynamic desensitization strategy, an RBAC strategy is used, the operator name is converted to a worker number hash value through a SHA-256 algorithm, only the administrator can decrypt it through a key, a data tamper-proofing unit that generates an operation chain data fingerprint through a Merkle tree and chains it for storage, generates an operation data block every 10 minutes, and calculates a hash value through a Merkle tree;

[0136] A feedback training module replays operation events in a digital twin environment, verifies the consistency of device actions and instructions, and pushes personalized training content to the fire control room terminal based on the evaluation matrix output, builds a fire control room digital twin scene based on a Unity engine, imports operation data that needs to include timestamps and device states, replays the operation process, including but not limited to details such as the operator mistakenly touching the gas fire extinguishing button, and verifies consistency by comparing the instruction issuance time and the device action time;

[0137] The linkage subsystem includes: a fire alarm system, an audible and visual alarm, an emergency lighting and evacuation indication system, a smoke control system, an automatic sprinkler fire extinguishing system, and a gas fire extinguishing system;

[0138] The operation data includes button pressing duration, device startup sequence timing, error attempt number, and consistency of environmental feedback data;

[0139] The operation data of each subsystem is strictly collected according to the requirements of the scheme:

[0140] Fire alarm system: record the timing of selecting the area, cutting off the background sound, and playing the evacuation instructions;

[0141] Emergency lighting system: collect the power switching response time (≤3 seconds for qualified) and the turning angle of the indicator;

[0142] Smoke control system: record the fan startup delay time (≤20 seconds) and valve state feedback, such as on / off signal;

[0143] Define the comprehensive effect coefficient E of fire emergency training, which is used to comprehensively measure the overall performance of the "evaluation-control-feedback" closed loop in fire training, and its formula is:

[0144]

[0145] Where, α, β, γ are the weight coefficients of evaluation effect, control accuracy, and feedback timeliness, and α+β+γ=1. In the actual application scenario of fire emergency training, according to the differences of training targets and focuses, the three weight coefficients can be dynamically adjusted. For example, for training with the main goal of improving the execution effect of fire fighting tactics, the α weight can be set to 0.5, emphasizing the importance of the evaluation link to the training effect; for training focusing on the accuracy of fire equipment operation control, the β weight can be set to 0.4; and for training scenarios requiring rapid response and cooperative combat, the γ weight can be adjusted to 0.3, highlighting the impact of feedback timeliness on the overall training effect.

[0146] It is the value obtained by quantitatively evaluating the actual results after fire training (such as shortening of fire extinguishing time, improvement of rescue success rate, reduction of personnel casualties, etc.);

[0147] It is the target value of the training results expected to be achieved before training. For example, in a fire drill, the fire extinguishing time is expected to be controlled within 15 minutes, and the actual fire extinguishing time is 12 minutes, so the score on this item is calculated as 12 / 15;

[0148] Reflects the actual accuracy of various control factors (such as operation parameter control of fire equipment, personnel action route control, fire extinguishing agent usage control, etc.) in the process of fire training;

[0149] is the pre-set ideal control standard range. Taking the control of the spray angle of a fire water cannon as an example, the ideal control range is ± 5°

[0150] the average deviation in actual training is ± 3°, the score of this item is calculated as

[0151] (quantify the accuracy by the proportion of deviation from the ideal control range).

[0152] refers to the actual time from collecting training data to forming effective feedback information and conveying it to the training personnel and relevant departments after the fire training ends;

[0153] is the standard upper limit of feedback time set according to the timeliness requirement of fire training. For example, the standard feedback time requires that the feedback be completed within 2 hours after the training ends, while the actual feedback time is 1.5 hours, then the score of this item is calculated as

[0154] Through this comprehensive effect coefficient E of fire emergency training, the running effect of the "evaluation-control-feedback" closed loop in fire training can be intuitively reflected, and the innovation and practicality of the first application of the closed loop in the field of fire emergency are highlighted. According to the size of E value, each link of fire training can be optimized and improved. When the E value is low, it is analyzed whether there is a problem of inaccurate evaluation of training effect in the evaluation link, or the accuracy of the control link is insufficient, or the timeliness of the feedback link needs to be improved, so as to take targeted measures such as perfecting the evaluation index system, strengthening the control technology training, optimizing the feedback process, etc., so as to continuously improve the quality and level of fire training, and ensure that the best effect can be played in the actual fire emergency scene.

[0155] In summary, the data acquisition module acquires the operation data of the fire control room operation equipment, the automatic fire alarm system and the linkage subsystem in real time, specifically covering the trigger time, position, button pressing duration and force change curve of the automatic fire alarm, the sequence branch, sub-step duration, error attempt number and type of the fire control room operation, the trigger condition, response time, equipment action delay and sequence of the linkage system, and the acquisition mode is achieved through the fire host API / OPCUA protocol, alarm controller event bus listening, linkage control cabinet log integration, console interface burying and video monitoring matching, while the high-precision sensor is used to ensure the accuracy of the data, and the AES-256 encryption combined with the SSL / TLS protocol transmission is used, and the LSTM model is used to fill in the data missing. Then, the evaluation and analysis module is based on these data, uses the analytic hierarchy process to build a hierarchical structure including the target layer, the criterion layer and the index layer, evaluates through the four-dimensional evaluation model, the built-in rule engine runs the analysis units in parallel, starts the three-level arbitration when the operation conflicts, supports dynamic rule hot loading and fuzzy reasoning to distinguish operation errors and equipment abnormalities, then, the linkage reinforcement module executes the linkage control according to the evaluation results, such as automatically dialing 119 when the fire confirmation is timed out for 30 seconds, locking the button and requiring the supervisor to unlock when the gas fire extinguishing delay stage is misoperated and the emergency stop is pressed, finally, the feedback training module generates an operation defect report, replays the operation event through the digital twin environment to verify the consistency of the equipment action and the instruction, and pushes personalized training content according to the evaluation results, forming a complete closed loop of data acquisition, evaluation and analysis, linkage control and feedback training, driving the continuous optimization of operation ability and system management.

[0156] The foregoing merely describes some exemplary embodiments of the present application by way of illustration, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of the claims of the present application.

Claims

1. An integrated fire emergency linkage control system, characterized by, The system comprises: a data acquisition module for real-time acquisition of operation data of fire control room operation equipment, automatic fire alarm system and linkage subsystem; an evaluation and analysis module connected to the data acquisition module, which has a four-dimensional evaluation model for time effectiveness, correctness, integrity and standardization analysis of operation behavior; a linkage reinforcement module that sends control instructions to fire equipment based on evaluation results; a feedback training module that generates operation defect reports and targeted training strategies.

2. The integrated fire emergency linkage control system according to claim 1, wherein: The four-dimensional evaluation model includes: a time effectiveness analysis unit that calculates the deviation coefficient of actual response time from the standard threshold; an integrity analysis unit that verifies the operation omission rate through a SOP step matcher; a compliance analysis unit that verifies the operation sequence logic based on safety interlocking rules; a standardization analysis unit that detects the compliance of two-person operation and the integrity of operation records.

3. The integrated fire emergency linkage control system of claim 2, wherein: The data acquisition module includes: high-precision pressure sensors and acceleration sensors integrated in manual alarm buttons to collect pressure change curves and trigger acceleration; optical sensors and capacitive touch sensors deployed on the fire control console to record operation trajectories and touch duration; IEEE1588 protocol clock synchronization device to ensure that the cross-system event timestamp error is less than or equal to 1μs.

4. The integrated fire emergency linkage control system of claim 3, wherein: The evaluation and analysis module has a rule engine that executes according to the following rules: A: Run the time effectiveness analyzer, SOP matcher, safety interlocking verifier and two-person operation detector in parallel; B: When operation conflicts are detected, the following three-level arbitration is started, with physical sensor data with timestamps being the first choice, digital signature control instructions being the second choice, and the final result being the optimal estimate generated by fusing multiple video analysis results.

5. The integrated fire emergency linkage control system of claim 4, wherein: The rule engine supports: dynamic rule hot loading mechanism that automatically imports new rule library when fire safety standards are updated; fuzzy rule inference unit that distinguishes between operation errors and equipment abnormalities based on device fault data.

6. The integrated fire emergency linkage control system of claim 5, wherein: The data acquisition module analyzes OPCUA, ModbusTCP and BACnet protocols through a special protocol conversion engine, and fills in the missing data caused by network jitter based on the LSTM time series prediction model.

7. The integrated fire emergency linkage control system of claim 6, wherein: When the fire alarm confirmation times out for 30 seconds, the linkage reinforcement module automatically triggers a 119 alarm call, and when a misoperation is detected during the delay stage of the gas fire extinguishing system, the linkage reinforcement module immediately locks the emergency stop button and requires supervisor permission to unlock.

8. The integrated fire emergency linkage control system of claim 7, wherein: The integrated fire emergency linkage control system further comprises: a privacy desensitization engine that converts operator names to worker ID hash values based on RBAC dynamic desensitization strategies; a data tamper-proofing unit that generates operation chain data fingerprints using a Merkle tree and stores them on the chain for evidence.

9. The integrated fire emergency linkage control system of claim 8, wherein: The feedback training module replays operation events in a digital twin environment, verifies the consistency of device actions and instructions, and pushes personalized training content to fire control room terminals based on the evaluation matrix output.

10. The integrated fire emergency linkage control system according to any one of claims 1-9, characterized in that: The linkage subsystem includes: fire alarm system, audible and visual alarm, emergency lighting and evacuation indication system, smoke control system, automatic sprinkler system and gas fire extinguishing system; The operation data includes button press duration, device start sequence timing, error attempt count and environmental feedback data consistency.

Images