Fire-fighting maintenance system based on fire-fighting control room graphic display device and implementation method

Abstract

The invention discloses a fire-fighting maintenance system based on a fire-fighting control room graphic display device and an implementation method, and the system comprises a data collection module which is used for collecting historical maintenance records of a plurality of pieces of equipment, and constructing a maintenance method recommendation table; the state analysis module is used for obtaining an inspection period of each device according to the historical maintenance records, performing state judgment on the devices according to the inspection periods to obtain normal devices and abnormal devices, and obtaining maintenance suggestions of the abnormal devices according to a maintenance method recommendation table; a trend prediction module which is used for carrying out fire hazard early warning and equipment state early warning according to the monitoring data sequence of the normal equipment, and updating the inspection period of the equipment according to the equipment state early warning and historical maintenance records; and the chart display module is used for performing chart display on output results of the state analysis module and the trend prediction module. The invention relates to the technical field of fire protection and maintenance, and solves the technical problems of low maintenance method efficiency and high maintenance cost of an existing maintenance system.

Description

Technical Field

[0001] The present invention belongs to the field of fire equipment maintenance and protection, involves data processing and bus communication technologies, and specifically relates to a fire maintenance and protection system and implementation method for a graphical display device in a fire control room. Background Art

[0002] In the field of fire safety, fire maintenance work is crucial for ensuring the reliable operation of fire protection systems. With the acceleration of urbanization and the expansion of building scale, the complexity of fire protection systems is increasing continuously, and traditional fire maintenance methods are facing many challenges.

[0003] Currently, the popular fire remote maintenance and guarantee systems on the market generally follow the Technical Specification for Urban Fire Remote Monitoring Systems GB 50440-2007. The original data is transmitted from the fire alarm system and the fire linkage control system through the user information transmission device to the remote server. Functions such as status warning, hidden danger analysis, fault statistics, and maintenance suggestions are completed on the remote maintenance and guarantee platform and then pushed to the responsible person's terminal. However, this mode requires additional equipment and leased lines, resulting in high construction costs and self-maintenance expenses, bringing a large economic burden to many users and restricting its wide application.

[0004] At the same time, traditional fire maintenance relies on manual experience to process maintenance records. When facing a large amount of complex fire equipment data, it is difficult to quickly and accurately analyze the cause of faults and provide effective maintenance suggestions. Moreover, manual judgment of equipment status is prone to misjudgment, and potential fault hidden dangers cannot be discovered in time, resulting in untimely equipment maintenance, affecting the normal operation of the fire protection system, and reducing the maintenance efficiency. In addition, when formulating maintenance plans, existing fire maintenance systems often adopt fixed inspection cycles and cannot be dynamically adjusted according to changes in equipment status, which may lead to over-maintenance or under-maintenance. Over-maintenance causes waste of resources, while under-maintenance increases the risk of equipment failure and makes it difficult to ensure that the fire protection system can play its role at critical moments.

[0005] In addition, data transmission in the remote maintenance system relies on the network, there are network transmission delay problems, which affect the real-time monitoring and warning of the status of fire equipment. Data also faces the risk of external attacks during remote transmission, and data security is difficult to guarantee. Once the data is leaked or tampered with, it will pose a serious threat to fire safety. Summary of the Invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present invention proposes a fire maintenance and protection system and implementation method based on a graphical display device in a fire control room, which is used to solve the technical problems of low efficiency of traditional maintenance methods, inaccurate fire warning, high maintenance costs, and network transmission delay and data security existing in the remote deployment system. The present invention solves the above problems.

[0007] To achieve the above object, a first aspect of the present invention provides a fire protection maintenance system based on a graphic display device in a fire control room, including:

[0008] A data collection module: used to collect historical maintenance records of a number of devices and construct a maintenance method recommendation form;

[0009] A status analysis module: used to obtain the inspection period of each device according to the historical maintenance records, judge the status of the device according to the inspection period, obtain normal devices and abnormal devices, and obtain maintenance opinions for abnormal devices according to the maintenance method recommendation form;

[0010] A trend prediction module: used to perform fire hazard early warning and device status early warning according to the monitoring data sequence of normal devices, and update the inspection period of the device according to the device status early warning and historical maintenance records;

[0011] A chart display module: used to display the output results of the status analysis module and the trend prediction module in the form of charts.

[0012] Further, the construction method of the maintenance method recommendation form includes:

[0013] A1. Using natural language processing technology to perform word segmentation on historical maintenance records to obtain a word segmentation sequence text;

[0014] A2. Training a sequence annotation model based on a machine learning algorithm and inputting the word segmentation sequence text into the sequence annotation model to obtain an annotation result; wherein, the annotation result includes device model, abnormal data, abnormal time, and maintenance operation;

[0015] A3. Storing the annotation result as structured data according to the device model, and using a clustering algorithm to cluster the abnormal data to obtain abnormal types;

[0016] A4. Associating the abnormal data, abnormal types, abnormal times, and maintenance operations under the same device model to obtain a maintenance method recommendation form.

[0017] By using natural language processing technology to perform word segmentation on historical maintenance records, and then using a sequence annotation model based on a machine learning algorithm for annotation, the unstructured historical maintenance records are converted into structured data, realizing the automation of the annotation process, improving the accuracy and consistency of data processing; and using a clustering algorithm to cluster the abnormal data can identify the common fault modes of similar devices, providing a standardized basis for the matching of subsequent maintenance suggestions, avoiding one-by-one comparison of massive data; finally, associating the abnormal data, abnormal types, abnormal times, and maintenance operations under the same device model to form a mapping relationship of "fault phenomenon - solution", making the maintenance suggestions more targeted.

[0018] Further, obtaining the inspection period of each device according to the historical maintenance records includes:

[0019] B11, dividing the historical maintenance records into several time segments according to a preset time interval, and calculating the failure rate of the device in each time segment according to the device model to obtain several segment failure rates;

[0020] B12, assigning weight values to the several segment failure rates according to the time sequence, and performing weighted summation on the segment failure rates according to the weight values to obtain the failure rate of each device;

[0021] B13, performing a reciprocal operation on the failure rate of each device to obtain the inspection period of each device.

[0022] Further, assigning weight values to the several segment failure rates according to the time sequence includes:

[0023] B12-1, arranging the several segment failure rates of each device in reverse order according to the time and numbering them to obtain a first sequence;

[0024] B12-2, judging whether the number of data in the first sequence is greater than a preset quantity threshold; if yes, deleting the data with numbers greater than the preset quantity threshold in the first sequence to obtain a second sequence; if no, marking the first sequence as the second sequence;

[0025] B12-3, according to the formula calculate to obtain the weight value w i ; where, η represents the attenuation rate, and η ∈ (0, 1], n represents the number of segment failure rates, j represents the summation loop variable, and i represents the time segment index.

[0026] Dividing the historical maintenance data according to a preset time interval and calculating the failure rates in each period avoids the averaging error of long-term data and can capture the seasonal or periodic laws of equipment failures; assigning higher weights to the recent failure rates through the attenuation rate conforms to the law of equipment aging, and the exponential attenuation weights strengthen the timeliness, ensuring that the inspection period is dynamically adjusted according to the health status of the equipment.

[0027] Further, judging the state of the device according to the inspection period includes:

[0028] B21, collecting the monitoring data sequences of several devices according to the inspection period, and converting the device model into a unique numerical code to obtain a device code;

[0029] B22, after standardizing the monitoring data sequence, splicing it with the device code to obtain device data;

[0030] Input the device data into the status analysis model to obtain the status type of the device; among them, the status type includes normal devices and abnormal devices; among them, the status analysis model is constructed based on deep learning algorithms.

[0031] Further, obtaining the maintenance opinions for abnormal devices according to the maintenance method recommendation table includes:

[0032] B31, screen the abnormal data, abnormal types, and maintenance operations in the maintenance method recommendation table according to the device model of the abnormal device to obtain the candidate set C = {R1, R2,..., R k}; where R i ∈C represents a candidate maintenance record, including the abnormal data vector V i , abnormal type T i , abnormal time t i and maintenance operation M i ;

[0033] B32, calculate the similarity S(V current , V i ) between the monitoring data sequence V current of the abnormal device and the abnormal data in the candidate set C according to the formula S(V i ) = 1 / (1 + |V current - V i | / σ) × e^[-0.1×(t

[0034] current - t current i current i ); where t current represents the acquisition time of the monitoring data sequence, and σ represents the standard deviation of historical normal AD data;

[0035] B33, screen out the candidate maintenance record R i with the highest similarity, and extract the maintenance operation and abnormal type to obtain the maintenance opinion for the abnormal device.

[0036] By using the formula for calculating similarity, balance the impact of data differences and time decay, give priority to recommending recent effective solutions, accurately find the historical maintenance case that best matches the current abnormal device, and provide targeted and effective maintenance suggestions for abnormal devices.

[0037] Further, the fire hazard warning and device status warning based on the monitoring data sequence of normal devices include:

[0038] C1, input the monitoring data sequence of normal devices into the fire warning model to obtain the fire warning probability P; where the fire warning model is constructed based on deep learning algorithms;

[0039] C2. Calculate T according to the formula alert = T base ×(1 + α × cos(2πt / 365)) × β building Obtain the fire dynamic warning threshold T alert ; where T base represents the preset warning basic threshold, α represents the seasonal influence coefficient, β building represents the building type coefficient, and t represents the time difference between the current time and January 1st of the current year;

[0040] C3. Judge whether the fire warning probability P is greater than the dynamic warning threshold T alert ; if yes, send a fire hazard warning signal and conduct equipment status warning according to the current status of several normal devices; if no, send a normal signal.

[0041] In the trend warning module, the fire hazard warning is used to judge the probability of future fires according to the time change trend of equipment monitoring values, and compare it with the dynamic threshold in real time to judge whether to issue a fire warning; while the equipment status warning is used to judge whether each current fire protection device is prepared to prevent fires, ensuring that the fire protection system can operate normally at critical moments and minimizing the losses caused by fires.

[0042] Furthermore, the equipment status warning according to the current status of several normal devices includes:

[0043] C4-1. For the required value of the water level of the sprinkler system, calculate the required water level height H according to the formula H required = H min +(H max - H min ) × tanh(γ × P) required ; where H min and H max respectively represent the legal minimum water level and the maximum capacity of the system, and γ represents the demand response steepness coefficient, which is obtained through equipment pressure testing;

[0044] C4-2. For the required value of the water pressure of the sprinkler system, calculate the required water pressure P according to the formula water ; where P low 、P mid 、P high respectively represent the basic water pressure grades of different classes, which are determined according to the pipeline pressure-bearing design, and η1, η2, and η3 respectively represent the adjustment coefficients of each stage, and η3 > η2 > η1;

[0045] C4-3. For the required value of the pressure of the fire extinguisher, calculate the required fire extinguisher pressure FE according to the formula required ​​; where FE min represents the legal minimum pressure, κ represents the pressure growth rate, which is determined through fire extinguisher performance tests, and δ represents the non - linear correction term, which is obtained through regression of historical fire data;

[0046] C4 - 4, determine whether the actual value in several normal devices is greater than or equal to the corresponding required value; if so, send a device normal signal; if not, send a device status warning signal; where the several devices include sprinkler system devices, fire extinguisher devices, and other devices, and the required values of other devices are set according to historical experience.

[0047] In the formula for calculating the required water level height, the fire probability is mapped to the device required parameter through a non - linear function, which satisfies the characteristics that the demand grows smoothly when the fire probability is low and rapidly approaches the maximum water level after the probability exceeds the critical value, so as to ensure that the water level is always fully stocked; in the formula for calculating the required water pressure, the water pressure is calculated in a staged manner to match the water pressure requirements under different risk levels; in the formula for calculating the required pressure of the fire extinguisher, the actual demand of the fire extinguisher pressure under different risk probabilities is simulated through an exponential growth function and a non - linear correction term. The lower the predicted probability, the smaller the possibility of a fire occurring or the smaller the scope of the fire initiation, so the required fire extinguisher pressure is also smaller, avoiding the problems of increased maintenance costs and waste of resources caused by a unified demand threshold under different probabilities.

[0048] Furthermore, updating the inspection period of the device according to the device status warning and historical maintenance records includes:

[0049] According to the formula T new = T current ×(1 - λ1×S - λ2×R recent / T current - λ3×P / T alert ) to calculate the updated inspection period T new ; where T current represents the current inspection period, S represents the device status signal, and S = 0 represents the device normal signal, S = 1 represents the device status warning signal, R recent represents the difference between the time of the most recent maintenance and the current time, and λ1, λ2, λ3 respectively represent preset weight coefficients, which are determined through historical experience.

[0050] By comprehensively considering multiple factors such as device status, maintenance history, and fire risk to update the period, it can shorten the inspection period accordingly when there is a warning of device status, a long maintenance interval, or a high fire risk, making the maintenance plan more scientific and reasonable.

[0051] The second aspect of the present invention provides a method for implementing a fire protection maintenance system based on a fire control room graphic display device, where

[0052] The fire protection maintenance system includes: a fire linkage subsystem, a fire warning subsystem, and an expansion subsystem. Several subsystems are connected to the fire control room graphic display device through RS232 bus, CAN bus, or ARCNET bus. Among them, the expansion subsystem includes, but is not limited to, a fire water supply monitoring subsystem, a gas fire extinguishing subsystem, a water spray fire extinguishing subsystem, a foam fire extinguishing subsystem, a dry powder fire extinguishing subsystem, a smoke prevention and exhaust monitoring subsystem, a fire door and rolling shutter subsystem, an elevator control subsystem, a fire telephone subsystem, a fire emergency broadcast subsystem, a combustible gas alarm subsystem, an electrical fire monitoring subsystem, a fire emergency lighting and evacuation indication subsystem, a fire equipment power status monitoring subsystem, and a fire power switching device;

[0053] The fire protection maintenance system also includes: a communication module, a data acquisition module, a status analysis module, a trend prediction module, and a chart display module. Each module performs data transmission and bus protocol adaptation with several subsystems through the communication module;

[0054] The fire control room graphic display device serves as the central node of the local fire protection maintenance system, communicates with several subsystems in real time through the bus protocol, integrates each module, and marks the processing results of several modules on the points of the electronic plane map according to the location information of several subsystems to generate charts and reports.

[0055] Compared with the prior art, the beneficial effects of the present invention are:

[0056] The present invention deeply integrates historical maintenance records and real-time device data through intelligent technology, significantly improving the accuracy and efficiency of fire protection maintenance. The system uses natural language processing and machine learning technologies to convert unstructured maintenance logs into structured maintenance recommendation forms, automatically associating device models, fault types, and maintenance operations, solving the problem of low efficiency caused by traditional reliance on manual experience. At the same time, the status analysis model built based on deep learning algorithms can automatically identify abnormal device states and dynamically match historical similar cases to generate maintenance suggestions, greatly reducing the misjudgment rate. In terms of fire warning, the system dynamically adjusts the warning threshold in combination with environmental factors such as seasonal changes and building types, and calculates device requirements (such as water level, pressure) in real time through a non-linear response mechanism to ensure the rapid response ability of fire protection equipment in high-risk scenarios. This data-driven decision-making mode enables maintenance work to shift from passive response to active prevention;

[0057] Different from traditional fire protection maintenance systems that rely on remote servers, the present invention is directly deployed on the fire control room graphic display device and communicates with fire subsystems in real time through bus protocols (such as RS232 / CAN), avoiding network transmission delays and risks of external attacks. Localized processing enables efficient storage and analysis of data, and sensitive information does not need to be transmitted externally, significantly enhancing security. In addition, the system makes full use of the existing hardware resources in the fire control room, and only realizes function upgrades through software expansion, without the need to deploy dedicated devices or servers additionally, greatly reducing the implementation and operation and maintenance costs. This lightweight architecture is particularly suitable for small and medium-sized venues, taking into account both efficiency and economy;

[0058] The present invention introduces a dynamic inspection cycle adjustment mechanism, which comprehensively considers the real-time status of equipment, historical failure frequencies, and fire risk levels to automatically optimize the maintenance plan. When the equipment status does not meet the standards or the fire probability increases, the system automatically shortens the inspection cycle to prioritize the protection of high-risk equipment; otherwise, it extends the cycle to reduce redundant inspections. This adaptive maintenance strategy not only improves the resource utilization efficiency but also reduces the probability of sudden failures through preventive maintenance. At the same time, the chart-based reports generated by the system (such as planar icon annotations and trend analysis charts) can intuitively display the fault points and warning areas, supporting maintenance personnel to quickly locate problems and reducing the dependence on professional outsourcing services. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0060] Figure 1 It is a schematic diagram of the technical process of the fire protection maintenance system provided by the present invention;

[0061] Figure 2 It is a schematic diagram of the framework of the fire protection maintenance system provided by the present invention;

[0062] Figure 3 It is a schematic diagram of the working process of the trend prediction module in the fire protection maintenance system provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0063] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0064] Please refer to Figures 1-3 , an embodiment of the first aspect of the present invention provides a fire protection maintenance system based on a graphical display device of a fire control room, including:

[0065] A data acquisition module: used to collect historical maintenance records of a number of devices and construct a maintenance method recommendation form;

[0066] A status analysis module: used to obtain the inspection cycle of each device according to the historical maintenance records, judge the status of the device according to the inspection cycle, obtain normal devices and abnormal devices, and obtain maintenance opinions for abnormal devices according to the maintenance method recommendation form;

[0067] A trend prediction module: used to perform fire hazard early warning and equipment status early warning according to the monitoring data sequence of normal devices, and update the inspection cycle of the device according to the equipment status early warning and historical maintenance records;

[0068] A chart display module: used to display the output results of the status analysis module and the trend prediction module in the form of charts.

[0069] It should be noted that the data acquisition module of the fire protection maintenance system in the present invention can also collect the monitoring data of a number of devices in real time, and transmit the monitoring data to the status analysis module, the trend prediction module and the chart display module. Moreover, the data acquisition module, the status analysis module, the trend prediction module and the chart display module of the present invention are in communication connection.

[0070] Based on the above technical modules, the present invention realizes the intelligentization, precision and high efficiency of the maintenance work of fire protection equipment. Among them, the maintenance method recommendation form constructed by the data acquisition module provides rich experience reference for the maintenance of abnormal devices; the status analysis module can timely discover potential fault hazards through the judgment of the device status by the inspection cycle, realize preventive maintenance, reduce the risk of sudden device failures, and ensure the stable operation of fire protection equipment; the trend prediction module combines fire hazard early warning and equipment status early warning, and dynamically updates the inspection cycle in combination with historical maintenance records, making the maintenance plan more in line with the actual situation of the equipment, avoiding waste of resources caused by over-maintenance, and preventing safety problems caused by insufficient maintenance; the chart display module presents complex data results in intuitive and easy-to-understand charts, facilitating staff to quickly master the device status, fault information and trend changes, reducing the difficulty of data interpretation, and improving decision-making efficiency.

[0071] In order to realize the intelligentization, precision and high efficiency of the maintenance work of fire protection equipment, in the data acquisition module, by using natural language processing technology to perform word segmentation on historical maintenance records, then obtaining annotations through a sequence annotation model trained by machine learning algorithms, and clustering abnormal data by using clustering algorithms, a systematic and targeted maintenance method recommendation form is constructed. The specific process is as follows:

[0072] The following is a detailed description of the specific operation process, design function of the data acquisition module, and corresponding implementation examples:

[0073] First, obtain the historical maintenance records of several devices, and then use natural language processing techniques (such as using natural language processing libraries like NLTK, spaCy, etc.) to perform word segmentation on the historical maintenance records to obtain the segmented sequence text. For example, for the maintenance record "The XX model fire pump had an abnormal current on May 10, 2023, and the maintenance personnel inspected and repaired the pump body circuit", after word segmentation, the segmented sequence text might be "The XX model fire pump had an abnormal current on May 10, 2023, and the maintenance personnel inspected and repaired the pump body circuit";

[0074] Train a sequence annotation model based on machine learning algorithms (such as conditional random field CRF, long short-term memory network LSTM, etc.). During the training process, a large amount of pre-labeled historical maintenance record data needs to be used as training samples to enable the model to learn the semantic roles represented by different words or phrases in the maintenance records (such as device model, abnormal data, abnormal time, maintenance operations, etc.);

[0075] Next, input the segmented sequence text into the trained sequence annotation model, and the model will annotate each word or phrase and output the annotation result. For example, for the above segmented sequence text, the annotation result is "[XX model: device model], [fire pump: device model], [on: none], [May 10, 2023: abnormal time], [had: none], [current: abnormal data], [abnormal: none], [maintenance personnel: none], [to: none], [pump body: device part], [circuit: device part], [performed: none], [the: none], [inspection: maintenance operation], [and: none], [repair: maintenance operation]";

[0076] According to the device model in the annotation result, store all relevant information (such as abnormal data, abnormal time, maintenance operations, etc.) as structured data;

[0077] Next, use clustering algorithms (such as K-Means clustering algorithm, DBSCAN clustering algorithm, etc.) to perform clustering analysis on the abnormal data in the annotation result. By calculating the similarity or distance between abnormal data, similar abnormal data are grouped into one category to obtain different abnormal types. For example, for the current abnormal, voltage abnormal data of multiple fire pumps, after clustering, it can be classified into the type of "electrical current abnormal", and those related to voltage are classified into the type of "electrical voltage abnormal";

[0078] For the same device model, associate its corresponding abnormal data, abnormal type, abnormal time, and maintenance operations. For example, for the "XX model fire pump", associate the "current abnormality" (belonging to the type of "electrical current abnormality") that occurred on "May 10, 2023" and the corresponding maintenance operation of "inspecting and repairing the pump body circuit";

[0079] Finally, organize all the associated data into a table form to obtain a recommended maintenance method table. Each row in this table may represent a maintenance record, and the columns are information such as device model, abnormal data, abnormal type, abnormal time, and maintenance operations respectively.

[0080] The following is an example table of the "recommended maintenance method table", showing the relevant information of different fire-fighting equipment under various abnormal conditions and the corresponding maintenance operations:

[0081]

[0082]

[0083] The recommended maintenance method table provides rich experience references for the subsequent maintenance work of abnormal equipment. When new equipment has similar faults, maintenance personnel can use the recommended table to quickly locate similar cases and draw on their maintenance methods and experiences, which can greatly improve the maintenance efficiency. At the same time, storing equipment maintenance knowledge in a structured way effectively realizes the accumulation and inheritance of knowledge. In addition, the abnormal types obtained through cluster analysis of historical maintenance records can help technicians summarize the problems that are likely to occur in equipment and provide a basis for preventive maintenance and improvement of equipment.

[0084] In the fire protection maintenance system, it is crucial to timely check and repair the operating conditions and potential hidden dangers of equipment. In order to quickly detect equipment abnormalities and accurately locate faults, the present invention obtains an inspection cycle that conforms to the actual situation of the equipment through operations such as carefully classifying historical maintenance records, calculating the failure rate by time segments, and assigning weights. On this basis, through a state analysis model, intelligently analyze the equipment that reaches the inspection cycle to judge the real-time operating state of the equipment. When it is found that the equipment is in an abnormal state, the system can immediately issue an alarm and obtain maintenance opinions according to the recommended maintenance record table, providing detailed and targeted maintenance guidance for maintenance personnel and shortening the time for troubleshooting and solving equipment faults.

[0085] In one implementation manner, obtaining the inspection cycle of each device according to historical maintenance records in the state analysis module may include the following operation steps:

[0086] First, obtain the historical maintenance records of the equipment and divide the historical maintenance records into several time segments according to a preset time interval, such as one week;

[0087] For each time segment, count the number of failures of the equipment of this model in the time period according to the equipment model, and calculate the failure rate of the equipment in each time segment based on the equipment operation time in the time period. For example, collect the maintenance records of model A fire pump in the past two years, divide the two years of maintenance records into one-month time intervals, and get 24 time segments. Count the number of maintenance times of equipment A in each time segment, divide by the time (in days), and get the failure rate of several segments of the equipment;

[0088] Next, the segment failure rates of each device are arranged in reverse order according to time, that is, the latest segment failure rate is placed in front, and they are numbered in sequence to form the first sequence. Then, it is determined whether the number of data in the first sequence is greater than the preset number threshold; assuming that the preset number threshold is 12, it means that the data within the most recent year is uniformly calculated;

[0089] If the number of data is greater than the preset number threshold, the data with numbers greater than the preset number in the first sequence are removed to obtain the second sequence; otherwise, the first sequence is directly marked as the second sequence;

[0090] Then according to the formula Calculate the weight value w i ; Where η represents the decay rate, and η∈(0,1], which determines the degree of decay of the early data weight over time. n represents the number of segment failure rates in the second sequence, j represents the summation loop variable, and i represents the time segment index;

[0091] The weighted segment failure rates are weighted summed, that is, each segment failure rate is multiplied by its corresponding weight value and then added to obtain the failure rate f of each device;

[0092] Finally, the failure rate of each device is inversely calculated, that is, the inspection period T = 1 / F.

[0093] In one implementation, the status analysis module determines the status of the device according to the inspection cycle, which may include the following steps:

[0094] B22, according to the inspection cycle calculated above, collect monitoring data sequences of several devices at corresponding time nodes. These monitoring data may come from various sensors connected to the devices, such as temperature sensors, pressure sensors, etc. After analog-to-digital conversion, the monitoring data is converted into discrete digital signal data (AD data);

[0095] At the same time, the device model is converted into a unique numerical code so that the device can be accurately identified in subsequent processing;

[0096] B22. Standardize the collected monitoring data sequence and splice it with the device code to form a complete device data vector. For example, if the device code is "001" and the standardized monitoring data sequence is [0.2, -0.1, 0.3], the device data obtained after splicing is [001, 0.2, -0.1, 0.3].

[0097] B23. Input the device data into the state analysis model constructed based on the deep learning algorithm. During the training phase, this model uses a large number of labeled device data (including normal device data and abnormal device data) for learning, so as to be able to identify the characteristic patterns in the device data and judge the state type of the device. The output result of the model is the state type of the device, that is, a normal device or an abnormal device.

[0098] Then for abnormal devices, the system will send a warning signal and obtain the maintenance opinions for the abnormal devices according to the maintenance method recommendation table to guide the accurate maintenance work of the devices. Specifically, it may include:

[0099] B31. Screen the abnormal data, abnormal types, and maintenance operations in the maintenance method recommendation table according to the device model of the abnormal device to obtain the candidate set C = {R1, R2,..., R k}; where R i ∈ C represents a candidate maintenance record, which includes the abnormal data vector V i , abnormal type T i , abnormal time t i and maintenance operation M i .

[0100] B32. Calculate the similarity S(V current , V i ) between the monitoring data sequence V current of the abnormal device and the abnormal data in the candidate set C according to the formula S(V i ) = 1 / (1 + |V current - V i | / σ) × e^[-0.1×(t current - t current )]

[0101] ; where t i represents the acquisition time of the monitoring data sequence, and σ represents the standard deviation of historical normal AD data; in this formula, |V current - V current | represents the difference between the current monitoring data sequence V i and the abnormal data V current in the candidate set, and V iThe absolute difference reflects the degree of numerical difference between the two. And σ is the standard deviation of historical normal AD data, which is used to measure the dispersion degree of historical normal data. Dividing the difference by the standard deviation is to normalize the data difference, eliminate the influence of different data magnitudes and units, and enable the comparison of data differences of different devices and different types on the same scale; e^[-0.1×(t current -t i ) is used to reflect the influence of time factor on similarity. As the time interval increases, the value of this exponential function will gradually decrease, indicating that the reference value of historical abnormal data will decrease over time;

[0102] B33, screen out the candidate maintenance record R with the highest similarity i , and extract the maintenance operations and abnormal types to obtain the maintenance opinions for the abnormal devices.

[0103] By analyzing historical maintenance records, comprehensively considering the failure rates in different time segments and the influence of time factors on failure rates, accurately calculate the reasonable inspection cycles for each device, effectively avoiding the waste of resources caused by over-inspection. At the same time, using the state analysis model constructed based on deep learning algorithms, accurately identify the operating states of devices, timely detect potential device failures, and improve the timeliness and accuracy of fire-fighting equipment maintenance work. In addition, based on the maintenance method recommendation table, the system can quickly screen similar maintenance cases for abnormal devices, lock the most matching maintenance records by calculating similarity, provide targeted maintenance opinions for maintenance personnel, shorten the exploration time of maintenance personnel, and achieve the efficient utilization and inheritance of maintenance knowledge.

[0104] For normal devices, in the trend prediction module, based on their monitoring data sequences, conduct fire hazard early warnings and device status early warnings, predict the probability of future fires, and determine whether the devices are prepared for fire prevention to ensure that fire-fighting equipment can function properly at critical moments and reduce the possible losses caused by fires; finally, update the inspection cycles of devices according to the device status early warnings and historical maintenance records to adapt to the actual operating conditions of the devices and achieve the reasonable allocation of device maintenance resources.

[0105] Specifically, in the trend prediction module, the following steps can be used to conduct fire hazard early warnings and device status early warnings:

[0106] (1) Fire hazard early warning:

[0107] First, collect the monitoring data sequences of normal devices; then input these monitoring data sequences into the fire early warning model constructed based on deep learning algorithms; this model uses a large number of monitoring data sequences containing different fire scenarios and normal scenarios during the training stage to learn the characteristic patterns and rules in the data; after the operation and processing of the model, output the fire early warning probability P;

[0108] Next, determine the values of several parameters in the threshold calculation formula. For example, preset the warning base threshold T base Set according to fire protection codes and historical experience, serving as a benchmark for measuring fire risks; the seasonal influence coefficient α is determined based on historical data and seasonal characteristics; the building type coefficient β building is determined according to the different fire risks of different building types (such as residential buildings, commercial buildings, industrial buildings, etc.), considering factors such as the fire protection facilities equipped, personnel density, and fire load of different types of buildings;

[0109] Substitute each parameter into the formula T alert = T base ×(1 + α×cos(2πt / 365))×β building Calculate the fire dynamic warning threshold T alert ; where t represents the time difference between the current date and January 1st of the current year;

[0110] Compare the fire warning probability P with the calculated fire dynamic warning threshold T alert If P is greater than T alert , the system sends a fire hazard warning signal and conducts equipment status warning based on the status of several current normal devices; if P is not greater than Talert, a normal signal is sent;

[0111] (2) Equipment status warning:

[0112] C4-1. For the water level requirement of the sprinkler system, calculate the required water level height H according to the formula H required = H min +(H max - H min )×tanh(γ×P); where H required and H min and H max represent the legal minimum water level and the maximum capacity of the system respectively, and γ represents the demand response steepness coefficient, which is obtained through equipment pressure testing;

[0113] C4-2. For the water pressure requirement of the sprinkler system, calculate the required water pressure Pwater according to the formula ; where Plow, Pmid, and Phigh represent the basic water pressure levels of different classes, determined according to the pipeline pressure-bearing design, and η1, η2, and η3 represent the adjustment coefficients of each stage, and η3 > η2 > η1;

[0114] C4-3. For the pressure requirement of the fire extinguisher, calculate the required fire extinguisher pressure FE according to the formula required ; where FE minIt represents the legal minimum pressure, κ represents the pressure growth rate, which is determined through fire extinguisher performance tests, and δ represents the non-linear correction term, which is obtained through regression of historical fire data;

[0115] C4-4, determine whether the actual value in several normal devices is greater than or equal to the required value; if so, send a device normal signal; if not, send a device status warning signal; among them, several devices include sprinkler system devices, fire extinguisher devices, and other devices, and the required value of other devices is set according to historical experience.

[0116] In one implementation, the formula for updating the device inspection cycle in the trend detection module is as follows: According to the formula T new = T current ×(1 - λ1×S - λ2×R recent / T current - λ3×P / T alert ) Calculate the updated inspection cycle T new ; where, T current represents the current inspection cycle, S represents the device status signal, and S = 0 represents the device normal signal, S = 1 represents the device status warning signal, R recent represents the difference between the time of the most recent maintenance and the current time.

[0117] Suppose in a fire protection system of a large commercial complex:

[0118] Collect the monitoring data sequences of various fire protection devices in the commercial complex in the past year, including data such as smoke concentration and temperature, and input them into the fire warning model to obtain a fire warning probability P of 0.4;

[0119] Determine the relevant parameters, T base = 0.2, α = 0.15 (the fire risk is relatively high in summer), β building = 1.3 (the fire risk of commercial buildings is relatively high), the current date is August 15th, and t = 227. Substitute into the formula to calculate the fire dynamic warning threshold T alert = 0.2×(1 + 0.15×cos(2π×227 / 365))×1.3 ≈ 0.24;

[0120] Since P (0.4) is greater than T alert (0.24), the system sends a fire hazard warning signal and starts device status warning.

[0121] Taking the sprinkler system as an example, it is known that H min = 2m, H max = 6m, γ = 0.6 (obtained through pressure tests), substitute into the formula H required = H min +(H max - Hmin ) × tanh(γ × P), and H is calculated required = 2 + (6 - 2) × tanh(0.6 × 0.4) ≈ 2.9 m. The actual water level of the sprinkler system is 2.5 m, which is less than the required water level height, and an equipment status warning signal is sent;

[0122] Assume the inspection period T of the current sprinkler system current = 2 months. Since an equipment status warning signal has been sent, S = 1, and the difference Rrecent between the last maintenance time and the current time is 0.5 months. Substitute into the formula T new = T current × (1 - λ1 × S - λ2 × R recent / T current - λ3 × P / T alert ). According to historical experience, set λ1 = 0.3, λ2 = 0.2, and λ3 = 0.1. Calculate Tnew = 2 × (1 - 0.3 × 1 - 0.2 × (0.5 / 2) - 0.1 × (0.4 / 0.24)) ≈ 1.17 months. Therefore, shorten the inspection period of the sprinkler system to 1.17 months to strengthen its maintenance and monitoring.

[0123] Through the above technical methods, in the trend prediction module, by using a deep learning model and dynamic threshold calculation, combined with equipment monitoring data sequences and environmental factors such as season and building type, early warning of fire hazards is realized, leaving sufficient time for the fire department and relevant personnel to take effective preventive measures, significantly reducing the probability of fire occurrence, and fully protecting the lives and property of people. At the same time, for the equipment status under fire hazard warning, by analyzing the reasonable demand values of different equipment in the fire risk and comparing them with the actual values, ensure that the equipment always meets the potential fire response requirements, enabling the fire-fighting equipment to operate normally and efficiently during a fire and giving full play to the fire extinguishing and rescue efficiency. In addition, according to the equipment status warning situation and historical maintenance records, the equipment inspection period is intelligently and dynamically adjusted. When the equipment warning occurs frequently or the maintenance is frequent, shorten the inspection period to strengthen monitoring and maintenance. When the equipment status is good, reasonably extend the inspection period to avoid waste of resources, thereby comprehensively improving the equipment maintenance efficiency and scientificity, and being able to effectively guarantee the long-term stable operation of the equipment.

[0124] Finally, in the chart display module, using data visualization technology, the normal or abnormal status of the equipment determined by the status analysis module, as well as the results such as the fire hazard warning probability and equipment status warning situation output by the trend prediction module, are presented in an intuitive and easy-to-understand chart form. For the output of the status analysis module, different colors, shapes or identifiers can be used to distinguish normal equipment from abnormal equipment. For example, a green circle represents normal equipment, and a red triangle represents abnormal equipment, and the current status information of the equipment is briefly described in the form of a text annotation beside the chart;

[0125] For the output of the trend prediction module, a line chart can be used to show the changing trend of the fire hazard warning probability over time, enabling users to clearly observe the dynamic trend of the fire risk; a bar chart can be used to compare the required values and actual values of different devices under fire risk, highlighting the device status warning situation;

[0126] In terms of the floor plan display, by drawing the building layout and accurately deploying device icons such as controllers and components, the fire protection maintenance situation can be clearly presented. When the user hovers the mouse over the device icon, through the front-end interaction design, the detailed status information of the current device, such as device name, model, operating parameters, etc., is quickly displayed in the form of a pop-up window; the future trend is elaborated in simple and clear language, such as the increasing or decreasing trend of fire hazards and the prediction of changes in device performance within a certain period in the future; for faulty devices, the maintenance methods matched from the maintenance method recommendation table are listed in detail in the pop-up window, including specific operation steps, required tools, etc., and at the same time, maintenance suggestions are given, such as the urgency of maintenance and the qualification requirements for maintenance personnel, providing users with comprehensive and practical device status and maintenance-related information to facilitate the efficient development of fire protection equipment management and maintenance work.

[0127] The second aspect of the embodiments of the present invention provides a method for implementing a fire protection maintenance system based on a fire control room graphic display device, where,

[0128] The fire protection maintenance system includes: a fire linkage subsystem, a fire warning subsystem, and an expansion subsystem, and several subsystems are connected to the fire control room graphic display device through an RS232 bus or a CAN bus or an ARCNET bus; where the expansion subsystem includes but is not limited to a fire water supply monitoring subsystem, a gas fire extinguishing subsystem, a water spray fire extinguishing subsystem, a foam fire extinguishing subsystem, a dry powder fire extinguishing subsystem, a smoke prevention and exhaust monitoring subsystem, a fire door and rolling shutter subsystem, an elevator control subsystem, a fire telephone subsystem, a fire emergency broadcast subsystem, a combustible gas alarm subsystem, an electrical fire monitoring subsystem, a fire emergency lighting and evacuation indication subsystem, a fire equipment power status monitoring subsystem, and a fire power supply switching device;

[0129] The fire protection maintenance system further includes: a communication module, a data acquisition module, a status analysis module, a trend prediction module, and a chart display module, and each module performs data transmission and bus protocol adaptation with several subsystems through the communication module;

[0130] The fire control room graphic display device serves as the central node of the local fire protection maintenance system, communicates with several subsystems in real time through the bus protocol, integrates each module, and marks the processing results of several modules on the points of the electronic floor map according to the location information of several subsystems to generate charts and reports.

[0131] Specifically, in the locally deployed fire protection maintenance system:

[0132] Communication module: As the hub of data transmission, the communication module uses the RS232 bus, CAN bus or ARCNET bus to establish a physical connection with the fire linkage subsystem, fire warning subsystem and expansion subsystem. According to GB16806 "Fire Linkage Control System" and related specifications, it can not only receive conventional alarm information such as fire alarms, supervision, activation, feedback, faults, and shielding, but also detect the real-time communication status with each controller. During the alarm interval of the controller, the communication module actively sends status query instructions to obtain data such as the detection values and average values of smoke fire alarm detectors and heat fire alarm detectors. To ensure reliable data transmission, a retransmission mechanism is adopted. If the slave response times out or the master does not receive the slave response, retransmission operations will be performed. At the same time, the CRC-16 check method is used to check the transmitted data, and 2 bytes of overhead are added to ensure data accuracy. The communication module converts the various received data into a format that meets the processing requirements of subsequent modules, laying the foundation for the data processing of the entire system;

[0133] Data acquisition module: Closely cooperating with the communication module, it obtains the preliminarily processed data from the communication module. This module focuses on collecting the operation data of various devices in the fire protection system, including but not limited to the working status of fire pumps, the pressure data of gas fire extinguishing systems, and the residual current values of electrical fire monitoring systems. The data acquisition module sorts and classifies these data, converts them into a format that can be deeply analyzed by subsequent modules, ensures the consistency and usability of the data, and provides rich and accurate data support for the intelligent analysis of the system; and constructs a maintenance method recommendation form through the historical stored maintenance record data;

[0134] Status analysis module: With the data provided by the data acquisition module, it comprehensively evaluates the operation status of fire protection equipment by combining intelligent algorithms. Through multi-dimensional analysis of the type, frequency, AD value, etc. of alarm data, it accurately identifies whether there are faults in the equipment, and gives the fault type and maintenance opinions of the faulty equipment according to the maintenance method recommendation form;

[0135] Trend prediction module: Based on long-term monitoring data, it predicts the fire trend and judges whether the equipment resources meet the fire extinguishing requirements, providing a favorable decision-making basis for fire protection work.

[0136] Chart display module: Utilize the programming layout information of fire alarm and linkage components and fire-fighting equipment on the graphical display device in the fire control room to present the processing results of the fault status statistical analysis module and the trend prediction module in an intuitive graphical manner. On the electronic floor plan, mark the fault and trend information at the corresponding points of each subsystem device. For example, use different colors to distinguish normal and faulty devices, and use flashing icons to indicate devices with early warning situations. At the same time, according to the user's needs, this module can generate statistical tables, statistical pie charts, and trend line charts by month and year, facilitating maintenance personnel to trace the cause of problems, understand the operation history of the fire protection system, and provide data support for formulating long-term maintenance plans;

[0137] Graphical display device in the fire control room: As the core of the entire local fire protection maintenance system, it maintains real-time communication with each subsystem through the bus protocol. It integrates a communication module, a data acquisition module, a status analysis module, a trend prediction module, and a chart display module, and summarizes the processing results of each module. According to the location information of each subsystem, accurately mark data such as the operating status, fault information, and trend prediction of the equipment at the corresponding points on the electronic floor plan, and generate intuitive and easy-to-understand charts and reports. These charts and reports can be accessed by fire protection maintenance personnel, management personnel, and supervision departments, realizing the efficient sharing and visual management of fire protection maintenance information, and improving the efficiency and accuracy of fire protection maintenance work.

[0138] Some of the data in the above formula are calculated by removing the dimension and taking its numerical value. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the actual situation; the preset parameters and preset thresholds in the formula are set by technicians in the field according to the actual situation or obtained through simulation of a large amount of data.

[0139] Working principle of the present invention:

[0140] The present invention collects the historical maintenance records of equipment through the data acquisition module, uses natural language processing and machine learning technologies for structured processing, and constructs a maintenance method recommendation form;

[0141] The status analysis module calculates the equipment failure rate based on historical maintenance data and determines the dynamic inspection cycle. Combining real-time monitoring data and deep learning models to judge the equipment status, and matching similar maintenance cases for abnormal equipment to generate maintenance suggestions;

[0142] The trend prediction module predicts the fire risk probability through time series analysis, dynamically adjusts the early warning threshold in combination with factors such as seasons and building types, synchronously evaluates the equipment resource requirements and triggers status warnings, and finally updates the inspection cycle according to the warning results and maintenance history;

[0143] Finally, the system presents the device status, warning information, etc. in a visual manner through the chart display module, supports electronic map marking and report generation, and realizes the intelligence, precision, and efficiency of fire protection maintenance.

[0144] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A fire protection maintenance system based on a fire control room graphic display device, characterized in that, Including: Data acquisition module: used to collect historical maintenance records of several devices and construct a maintenance method recommendation form; Status analysis module: used to obtain the inspection cycle of each device according to the historical maintenance records, judge the status of the device according to the inspection cycle to obtain normal devices and abnormal devices, and obtain maintenance opinions for abnormal devices according to the maintenance method recommendation form; Trend prediction module: used to conduct fire hazard early warning and device status early warning according to the monitoring data sequence of normal devices, and update the inspection cycle of the device according to the device status early warning and historical maintenance records; Chart display module: used to display the output results of the status analysis module and the trend prediction module in the form of charts.

2. The fire protection maintenance system based on the graphical display device of the fire control room according to claim 1, wherein The construction method of the maintenance method recommendation form includes: A1. Using natural language processing technology to perform word segmentation on historical maintenance records to obtain a word segmentation sequence text; A2. Training a sequence annotation model based on a machine learning algorithm and inputting the word segmentation sequence text into the sequence annotation model to obtain an annotation result; among them, the annotation result includes device model, abnormal data, abnormal time, and maintenance operation; A3. Storing the annotation result as structured data according to the device model, and using a clustering algorithm to cluster the abnormal data to obtain abnormal types; A4. Associating the abnormal data, abnormal types, abnormal times, and maintenance operations under the same device model to obtain a maintenance method recommendation form.

3. The fire protection maintenance system based on the fire control room graphic display device according to claim 1, characterized in that, The obtaining of the inspection cycle of each device according to the historical maintenance records includes: B11. Dividing the historical maintenance records into several time segments at a preset time interval, and calculating the failure rate of the device in each time segment according to the device model to obtain several segment failure rates; B12. Assigning weight values to several segment failure rates according to the time sequence, and performing weighted summation on the segment failure rates according to the weight values to obtain the failure rate of each device; B13. Performing a reciprocal operation on the failure rate of each device to obtain the inspection cycle of each device.

4. The fire protection maintenance system based on the graphical display device of the fire control room according to claim 3, characterized in that, The assigning of weight values to several segment failure rates according to the time sequence includes: B12-1. Sorting and numbering the several segment failure rates of each device in reverse order according to the time to obtain a first sequence; B12-2. Judging whether the number of data in the first sequence is greater than a preset number threshold; if so, deleting the data with numbers greater than the preset number threshold in the first sequence to obtain a second sequence; if not, marking the first sequence as the second sequence; B12-3, calculated according to the formula to obtain the weight value w i ; where η represents the decay rate and η ∈ (0, 1], n represents the number of segmented failure rates, j represents the summation loop variable, and i represents the time segmentation index.

5. The fire protection maintenance system based on the graphical display device of the fire control room according to claim 1, characterized in that, The judging of the device status according to the inspection cycle includes: B21. Collecting the monitoring data sequences of several devices according to the inspection cycle, and converting the device model into a unique value numerical code to obtain a device code; B22. After standardizing the monitoring data sequence, splicing it with the device code to obtain device data; B23. Inputting the device data into a status analysis model to obtain the status type of the device; among them, the status type includes normal devices and abnormal devices; among them, the status analysis model is constructed based on a deep learning algorithm.

6. The fire protection maintenance system based on the fire control room graphic display device according to claim 5, characterized in that, The obtaining of the maintenance opinion for the abnormal device according to the maintenance method recommendation form includes: B31, screen the abnormal data, abnormal types, and repair operations in the repair method recommendation table according to the device model of the abnormal device, and obtain the candidate set C = {R1, R2, …, R k}; where, R i ∈ C represents a candidate repair record, including the abnormal data vector V i , the abnormal type T i , the abnormal time t i , and the repair operation M i ; B32, according to the formula S(V current , V i ) = 1 / (1 + |V current - V i | / σ) × e^[-0.1×(t current - t i )] Calculate the monitoring data sequence V of the abnormal device current The similarity S(V current , V i ) with the abnormal data in the candidate set C; where t current represents the acquisition time of the monitoring data sequence, and σ represents the standard deviation of the historical normal AD data; B33. Screen out the candidate repair record R with the highest similarity i , and extract the repair operations and abnormal types to obtain the repair opinions for the abnormal devices.

7. The fire protection maintenance system based on the fire control room graphic display device according to claim 5, characterized in that, The conducting of fire hazard early warning and device status early warning according to the monitoring data sequence of normal devices includes: C1. Input the monitoring data sequence of normal devices into the fire warning model to obtain the fire warning probability P; wherein, the fire warning model is constructed based on the deep learning algorithm; C2. Calculate T according to the formula alert = T base × (1 + α × cos(2πt / 365)) × β building to obtain the dynamic fire warning threshold T alert ; where T base represents the preset warning basic threshold, α represents the seasonal influence coefficient, and β building represents the building type coefficient, and t represents the time difference between the current time and January 1 of the current year; C3, determine whether the fire warning probability P is greater than the dynamic warning threshold T alert ; if yes, send a fire hazard warning signal and conduct equipment status warning based on the current status of several normal devices; if no, send a normal signal.

8. The fire protection maintenance system based on the fire control room graphic display device according to claim 7, characterized in that, The device status warning according to the current status of several normal devices includes: C4-1. For the required value of the water level of the sprinkler system, according to the formula H required = H min +(H max - H min ) × tanh(γ × P), the required water level height H required is calculated; where H min and H max represent the legal minimum water level and the maximum capacity of the system respectively, and γ represents the demand response steepness coefficient; C4-2, the required value of the water pressure for the sprinkler system, is calculated according to the formula to obtain the required water pressure P water ; where P low , P mid , and P high represent the basic water pressure levels of the first stage, the second stage, and the third stage respectively, and η1, η2, and η3 represent the adjustment coefficients of each stage, and η3 > η2 > η1; C4-3, the required value of the fire extinguisher pressure, according to the formula Calculate the required pressure FE of the fire extinguisher required ; where FE min represents the legal minimum pressure, κ represents the pressure growth rate, and δ represents the non-linear correction term; C4-4. Determine whether the actual value in several normal devices is greater than or equal to the corresponding required value; if so, send a device normal signal; if not, send a device status warning signal; wherein, several devices include sprinkler system devices, fire extinguisher devices and other devices, and the required value of other devices is set according to historical experience.

9. The fire protection maintenance system based on the fire control room graphic display device according to claim 7, characterized in that, The update of the device inspection cycle according to the device status warning and historical maintenance records includes: According to the formula T new = T current ×(1 - λ1×S - λ2×R recent / T current - λ3×P / T alert ), the updated inspection period T new is calculated; where T current represents the current inspection period, S represents the device status signal, and S = 0 represents the normal device signal, S = 1 represents the device status warning signal, R recent represents the difference between the last maintenance time and the current time, and λ1, λ2, and λ3 respectively represent the preset weight coefficients.

10. An implementation method of a fire protection maintenance system based on a fire control room graphic display device, applied to the fire protection maintenance system based on a fire control room graphic display device according to any one of claims 1-9, characterized in that The fire protection maintenance system includes: a fire linkage subsystem, a fire warning subsystem and an expansion subsystem, and several subsystems are connected to the fire control room graphic display device through an RS232 bus or a CAN bus or an ARCNET bus; wherein the expansion subsystem includes, but is not limited to, a fire water supply monitoring subsystem, a gas fire extinguishing subsystem, a water spray fire extinguishing subsystem, a foam fire extinguishing subsystem, a dry powder fire extinguishing subsystem, a smoke prevention and exhaust monitoring subsystem, a fire door and rolling shutter subsystem, an elevator control subsystem, a fire telephone subsystem, a fire emergency broadcast subsystem, a combustible gas alarm subsystem, an electrical fire monitoring subsystem, a fire emergency lighting and evacuation indication subsystem, a fire equipment power status monitoring subsystem and a fire power switching device; The fire protection maintenance system further includes: a communication module, a data acquisition module, a status analysis module, a trend prediction module and a chart display module, and each module performs data transmission and bus protocol adaptation with several subsystems through the communication module; The fire control room graphic display device serves as the central node of the local fire protection maintenance system, communicates with several subsystems in real time through the bus protocol, integrates each module, marks the processing results of several modules on the points of the electronic plane map according to the position information of several subsystems, and generates charts and reports according to the transmission results of the icon display module.

Images