Special equipment potential safety hazard management and control big data monitoring system and monitoring method

Abstract

The invention discloses a special equipment potential safety hazard management and control big data monitoring system and method, and relates to the technical field of special equipment monitoring, and the method achieves the comprehensive monitoring of the elevator shaft environment, the operation state and the internal passenger behavior through the steps 1-5. Through combination of shaft environment monitoring, running state early warning and elevator internal behavior collection, a comprehensive monitoring view angle is provided, and potential safety hazards can be found in time. An environment degradation degree coefficient HJlh, an elevator state variation index Pwd, an elevator internal environment state factor BhvPyz, a behavior deviation factor DevRunz and an abnormity early warning index AlarmIdx are constructed. And comprehensively evaluating to obtain a risk evaluation level, and taking corresponding control measures. The system can identify and prevent various possible potential safety hazards in advance. According to the comprehensive safety guarantee system, the running reliability of the elevator is improved, the fault occurrence probability is reduced, and the safety management level of special equipment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of special equipment monitoring, and in particular to a special equipment safety hazard management and control big data monitoring system and a monitoring method. Background Art

[0002] Special equipment plays a vital role in modern society. They cover a variety of equipment including elevators, boilers, pressure vessels, lifting machinery, etc. These equipment are widely used in industrial production, commercial buildings, public facilities and other fields. The safety of special equipment not only directly affects the normal operation of the equipment, but also involves the life safety and property safety of personnel. For example, as a high-frequency vertical transportation equipment, the operation stability and safety of the elevator are of vital importance. If the elevator fails or abnormalities during operation, it may cause passengers to be trapped in the elevator, and even cause serious safety accidents. Similar situations may also occur in other special equipment, such as boiler explosions or pressure vessel leaks, which will pose a great threat to production safety.

[0003] Traditional elevator monitoring systems mainly focus on the detection of operating status, such as the acceleration, vibration, displacement and other parameters of the elevator. These systems usually collect data through basic sensors, but the monitoring of the elevator shaft environment is relatively limited. This monitoring blind spot may lead to the failure to detect environmental anomalies in time, which in turn affects the normal operation of the elevator. For example, excessive humidity in the shaft may cause corrosion of the elevator equipment, while abnormal temperature may cause overheating or failure of the elevator system. In addition, the monitoring of the elevator operation status also has certain limitations. Existing systems can usually only perform basic measurements of parameters such as acceleration and vibration of the elevator, but the monitoring and evaluation of subtle changes in operation, such as elevator jitter, landing error and control delay, are not sufficient. For example, improper operation of passengers in the elevator, such as blocking the elevator door from closing, may cause the system to fail to issue an alarm in time in the early stage of potential failure of the elevator, delaying the maintenance time. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a special equipment safety hazard management and control big data monitoring system and monitoring method to solve the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a special equipment safety hazard management and control big data monitoring system, including an environment monitoring module, an operation monitoring module, a second monitoring module, an analysis module and an evaluation control module;

[0006] The environment monitoring module is used to use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator shaft in real time to obtain an environmental status data set, and to construct an environmental degradation degree coefficient HJlh based on the environmental status data set, and to pre-set an environmental degradation threshold E. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current shaft environment is abnormal, and a first warning instruction is generated;

[0007] The operation monitoring module further monitors the elevator operation status data after receiving the first warning instruction, establishes an operation data set, pre-processes the relevant data in the operation data set, extracts the elevator operation jitter characteristics, the stop floor error characteristics and the control delay characteristics, obtains the elevator state variation index Pwd, and issues a second operation warning instruction if the elevator state variation index Pwd exceeds the abnormal operation threshold K;

[0008] The second monitoring module is used to use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information about improper operation of passengers, and obtain lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set;

[0009] The analysis module is used to establish a risk prediction model. The monitoring data set is used as an input item and input into the risk prediction model for analysis and calculation to obtain: the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz; and the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz are summarized and analyzed to generate the abnormal warning index AlarmIdx;

[0010] The evaluation control module is used to pre-set the early warning evaluation threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning evaluation threshold AlarmK, and take corresponding control plans according to the risk assessment level.

[0011] Preferably, the environment monitoring module includes a sensor acquisition unit and an environment calculation unit;

[0012] The sensor acquisition unit is used to determine the key positions in the elevator shaft according to the internal structure diagram of the elevator shaft, and deploy several groups of environmental monitoring sensors and data acquisition equipment in the elevator shaft to monitor the temperature, humidity, air quality and vibration conditions in the shaft in real time to build an environmental status data set;

[0013] The environmental status data set includes: the surface temperature T of the shaft equipment, the humidity S at the bottom of the shaft, the PM2.5 concentration P in the shaft, and the vibration intensity value Z of the shaft wall.

[0014] Preferably, after preprocessing the environmental state data set, the environmental computing unit extracts the surface temperature T of the hoistway equipment, the humidity S at the bottom of the hoistway, the PM2.5 concentration P in the hoistway, and the vibration intensity value Z of the hoistway wall surface. After performing linear normalization processing, the corresponding numerical features are mapped to the interval [0, 1], and then the environmental degradation degree coefficient HJlh is obtained according to the following calculation method:

[0015]

[0016] where i = 1, 2,..., n, and n is the number of environmental state data acquisition time nodes; T i is the surface temperature of the hoistway equipment at the i-th acquisition time node, is the ideal temperature threshold of the hoistway equipment surface, S i represents the humidity at the bottom of the hoistway at the i-th acquisition time node, is the ideal humidity threshold of the hoistway bottom, P i represents the PM2.5 concentration in the hoistway at the i-th acquisition time node, is the ideal PM2.5 concentration threshold in the hoistway, Z i represents the vibration intensity value of the hoistway wall surface at the i-th acquisition time node, is the bearing vibration intensity threshold of the hoistway wall surface, and F1, F2, F3, and F4 are weight coefficients: 0 < F1 < 1, 0 < F2 < 1, 0 < F3 < 1, 0 < F4 < 1, and F1 + F2 + F3 + F4 = 1;

[0017] Meanwhile, an environmental degradation threshold E is preset in advance. When the environmental degradation degree coefficient HJlh > the environmental degradation threshold E, it indicates that the fluctuation of the current hoistway environment is abnormal, and a first warning instruction is generated, including starting the environmental control equipment for real-time control and activating the shock absorption device;

[0018] When the environmental degradation degree coefficient HJlh ≤ the environmental degradation threshold E, it indicates that the fluctuation of the current hoistway environment is normal, and continuous monitoring is carried out.

[0019] Preferably, the operation monitoring module includes an operation monitoring unit, a feature extraction unit, and an operation calculation unit.

[0020] The operation monitoring unit is used to further monitor the elevator operation state data. By installing an acceleration sensor, a vibration sensor, and a displacement sensor at the bottom of the elevator car, the elevator operation acceleration data, elevator displacement data, and vibration data during operation are collected to establish an operation data set.

[0021] The feature extraction unit is used to preprocess the relevant data in the operation data set, which includes data cleaning, data standardization and time alignment, and extract the elevator operation jitter characteristics, floor error characteristics and control delay characteristics to obtain: elevator instantaneous acceleration Ss, elevator maximum acceleration Sdmax, elevator average acceleration Savg, elevator vibration frequency Zdpl, elevator vibration intensity Zdqd, vertical displacement difference DCy between the elevator car and the predetermined stop position when it stops at the floor, elevator start delay Qdyc and elevator deceleration delay Jsyc.

[0022] Preferably, the operation calculation unit is used to extract the elevator instantaneous acceleration Ss, the elevator maximum acceleration Sdmax, the elevator average acceleration Savg, the elevator vibration frequency Zdpl, the elevator vibration intensity Zdqd, the vertical displacement difference DCy between the elevator car and the predetermined parking position when the elevator car stops at the floor, the elevator start delay Qdyc and the elevator deceleration delay Jsyc; after dimensionless processing, the elevator state variation index Pwd is calculated by the following formula:

[0023]

[0024] Wherein: bz1 represents the standard threshold of instantaneous acceleration of the elevator, bz2 represents the maximum acceleration bearing threshold of the elevator, bz3 represents the standard threshold of average acceleration of the elevator, bz4 represents the preset tolerance threshold of the vibration frequency of the elevator, bz5 represents the preset tolerance threshold of the vibration intensity of the elevator, bz6 represents the preset allowable threshold of the vertical displacement difference between the elevator car and the predetermined parking position when the elevator car stops at the floor, bz7 represents the preset allowable threshold of the elevator start delay, bz8 represents the preset allowable threshold of the elevator deceleration delay, w1, w2, w3, w4, w5, w6, w7 and w8 represent weight values, and the sum of the weights is 1;

[0025] At the same time, an abnormal operation threshold K is pre-set. When the elevator status variation index Pwd>abnormal operation threshold K, it indicates that the fluctuation of the current elevator operation is estimated to be abnormal, and a second early warning instruction is generated, including activating the elevator internal alarm system to remind passengers of the operation risk and inform them that the elevator is temporarily out of service. Then, technicians are arranged to conduct on-site inspections of the elevator, check the status of each component, monitor the equipment parts related to acceleration, vibration, displacement and delay, and resume the elevator operation after repair and adjustment. In the initial stage of resuming operation, the monitoring frequency is strengthened to 3-5 times per hour to prevent the problem from happening again.

[0026] When the elevator state variation index Pwd ≤ abnormal operation threshold K, it means that the fluctuation of the current elevator operation is estimated to be normal and continuous monitoring is performed.

[0027] Preferably, the second monitoring module includes an image recognition unit and a second sensor acquisition unit;

[0028] The image recognition unit is used to collect and identify the improper operation information of passengers in the elevator through an image acquisition device, including;

[0029] Identify the risk of items carried by passengers through the image acquisition device, identify and calculate the size and shape of the items, and identify and obtain the area Mj of the carried items;

[0030] Identify the long-press time of the floor button pressed by the passenger through the image acquisition device. If it exceeds five seconds, it is calculated as an improper operation to obtain the number of improper operations bdcz1, and collect the behavior information of the passenger forcibly blocking the elevator door from closing, including the behavior of using hands, feet or items to block the elevator door, to obtain the extra delay time Ycsj that interferes with the normal closing of the elevator door;

[0031] The second sensing and acquisition unit is used to install a smoke sensor in the elevator to monitor the smoke concentration in the air in real time and obtain the smoke concentration value Ywnd; and install a lighting sensor in the elevator to monitor the lighting brightness in the elevator in real time and obtain the light brightness Gzld;

[0032] The monitoring data set includes: the area Mj of the carried items, the number of improper operations bdcz1, the extra delay time Ycsj that interferes with the normal closing of the elevator door, the smoke concentration value Ywnd, and the light brightness Gzld.

[0033] Preferably, the analysis module includes a model establishment unit, an environmental state factor calculation unit, and an improper behavior calculation unit;

[0034] The model establishment unit is used to establish a risk prediction model through random forest and support vector machine technologies. After data cleaning and normalization of the monitoring data set, it is divided into a training set and a test set for training, and then the performance of the model is evaluated. Calculate and analyze the accuracy, precision, recall rate, and F1 score indicators of the model until the evaluation is qualified;

[0035] The environmental state factor calculation unit is used to extract the smoke concentration value Ywnd and the light brightness Gzld in the monitoring data set. After dimensionless processing, the environmental state factor BhvPyz in the elevator is calculated through the following formula:

[0036]

[0037] where i = 1, 2,..., p, p is the number of acquisition time nodes in the monitoring data set; d1 and d2 are weight coefficients: 0 < d1 < 1, 0 < d2 < 1, and d1 + d2 = 1; Gzld i is the light brightness at the i-th acquisition time node, is the light brightness threshold; Ywnd i represents the smoke concentration value at the i-th acquisition time node, is the preset threshold for smoke concentration;

[0038] The improper behavior calculation unit is used to extract the carried item area Mj, the number of improper operations bdcz1, and the extra delay time Ycsj that interferes with the normal closing of the elevator door from the monitoring dataset. After dimensionless processing, the behavior deviation factor DevRunz is calculated through the following formula:

[0039]

[0040] In the formula, Mj k represents the carried item area of the k-th passenger, Mj max represents the preset threshold for the maximum carried item area allowed for passengers, bdcz1 k represents the number of improper operations of the k-th passenger, bdcz1 max represents the preset threshold for the maximum number of improper operations allowed for passengers, Ycsj k represents the extra delay time that the behavior of the k-th passenger causes to interfere with the normal closing of the elevator door, Ycsj max represents the preset threshold for the maximum extra delay time that passengers are allowed to interfere with the normal closing of the elevator door; d3, d4, and d5 are weight coefficients: 0 < d3 < 1, 0 < d4 < 1, 0 < d5 < 1, and d3 + d4 + d5 = 1.

[0041] Preferably, the analysis module further includes an association unit, and the association unit is used to associate the elevator interior environment state factor BhvPyz and the behavior deviation factor DevRunz after dimensionless processing through the following formula to obtain the abnormal warning index AlarmIdx:

[0042]

[0043] In the formula, α and δ respectively represent the weight values of the environment state factor BhvPyz and the behavior deviation factor DevRunz.

[0044] Preferably, the evaluation and control module includes an evaluation unit and a control unit:

[0045] The evaluation unit is used to preset the warning evaluation threshold AlarmK, and compare the warning evaluation threshold AlarmK with the abnormal warning index AlarmIdx to obtain the risk assessment level:

[0046] When the abnormal warning index AlarmIdx ≤ the warning evaluation threshold AlarmK, it is determined as low risk, and the elevator continues to work normally and continues to monitor;

[0047] When the warning assessment threshold AlarmK < abnormal warning index AlarmIdx ≤ warning assessment threshold AlarmK*130%, it indicates a medium risk, and the control unit triggers the internal risk warning and prompt of the elevator, immediately stops the elevator operation, reminds relevant personnel to check, and restarts the elevator after the improper behavior is eliminated;

[0048] When the abnormal warning index AlarmIdx>the warning assessment threshold AlarmK*130%, it indicates a high-level risk, and the elevator internal risk warning is triggered through the control unit, and the elevator operation is immediately stopped, and the elevator door is set to the open state, and an emergency notification is automatically sent to the emergency service fire department and rescue team, triggering the automatic ventilation system or backup lighting in the elevator. After confirming that the passengers have been safely evacuated, the power of the elevator is remotely shut down to prevent any accidental startup or further damage.

[0049] A big data monitoring method for safety hazard management and control of special equipment includes the following steps:

[0050] Step 1: Hoistway environment monitoring and early warning: Use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator hoistway in real time to obtain an environmental status data set, and construct an environmental degradation degree coefficient HJlh based on the environmental status data set. At the same time, the environmental degradation threshold E is pre-set. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current hoistway environment is abnormal, and a first early warning instruction is generated;

[0051] Step 2: Operation status warning: After receiving the first warning instruction, further monitor the elevator operation status data, establish an operation data set, pre-process the relevant data in the operation data set, extract the elevator operation jitter characteristics, floor error characteristics and control delay characteristics, and obtain the elevator state variation index Pwd. If the elevator state variation index Pwd exceeds the abnormal operation threshold K, issue a second operation warning instruction;

[0052] Step 3: Elevator internal behavior and environment collection: Use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information on improper operation by passengers, and obtain lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set;

[0053] Step 4: Risk prediction: Establish a risk prediction model, use the monitoring data set as input, input it into the risk prediction model for analysis and calculation to obtain: elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz; summarize and analyze the elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz to generate the abnormal warning index AlarmIdx;

[0054] Step 5. Evaluate and obtain the risk assessment level and then respond and control: Pre-set the early warning assessment threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning assessment threshold AlarmK, and take corresponding control measures according to the risk assessment level.

[0055] The present invention provides a special equipment safety hazard control big data monitoring system and monitoring method. It has the following beneficial effects:

[0056] (1) This method achieves comprehensive monitoring of the elevator shaft environment, operating status, and internal passenger behavior through steps 1 to 3. The combination of shaft environment monitoring, operating status warning, and internal behavior collection provides a comprehensive monitoring perspective, which helps to timely discover potential safety hazards.

[0057] (2) Through the risk prediction model in step 4, the monitoring data (environmental status, behavioral deviation, etc.) from different data sources are comprehensively analyzed to generate the abnormal warning index AlarmIdx. This data fusion improves the accuracy of identifying potential risks and avoids important information that may be missed by a single data source. Through comprehensive monitoring and analysis of the elevator environment, operating status, and passenger behavior, the system can identify and prevent various possible safety hazards in advance. This comprehensive safety assurance system improves the reliability of elevator operation, reduces the probability of failure, and thus enhances passengers' sense of security and trust.

[0058] (3) The environmental monitoring module uses a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator shaft in real time, and constructs the environmental degradation degree coefficient based on the environmental status data set, and generates an early warning instruction when it exceeds the preset threshold. After receiving the early warning instruction, the operation monitoring module further monitors the operation status data of the elevator, extracts the jitter characteristics, floor error characteristics and control delay characteristics of the elevator operation, calculates the elevator state variation index, and issues a second early warning instruction when it exceeds the abnormal operation threshold. The second monitoring module monitors the behavior of passengers in the elevator through image acquisition equipment, captures improper operation information, and collects lighting information and smoke concentration information in the elevator. The analysis module establishes a risk prediction model based on the monitoring data set, calculates the environmental status factor and behavior deviation factor in the elevator, and generates an abnormal early warning index. The evaluation and control module compares the abnormal early warning index according to the early warning evaluation threshold, obtains the risk assessment level, and takes corresponding control measures. This comprehensive monitoring system helps to timely discover and deal with potential safety hazards and improve the safety management level of special equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1This is a flowchart diagram of a big data monitoring system for safety hazard management and control of special equipment according to the present invention;

[0060] Figure 2 This is a schematic diagram of the steps of a big data monitoring method for safety hazard management and control of special equipment according to the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Example 1, please refer to Figure 1 , the present invention provides a special equipment safety hazard management and control big data monitoring system, including an environment monitoring module, an operation monitoring module, a second monitoring module, an analysis module and an evaluation control module;

[0063] The environment monitoring module is used to use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator shaft in real time to obtain an environmental status data set, and to construct an environmental degradation degree coefficient HJlh based on the environmental status data set, and to pre-set an environmental degradation threshold E. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current shaft environment is abnormal, and a first warning instruction is generated;

[0064] The operation monitoring module further monitors the elevator operation status data after receiving the first warning instruction, establishes an operation data set, pre-processes the relevant data in the operation data set, extracts the elevator operation jitter characteristics, the stop floor error characteristics and the control delay characteristics, obtains the elevator state variation index Pwd, and issues a second operation warning instruction if the elevator state variation index Pwd exceeds the abnormal operation threshold K;

[0065] The second monitoring module is used to use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information about improper operation of passengers, and obtain lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set;

[0066] The analysis module is used to establish a risk prediction model. The monitoring data set is used as an input item and input into the risk prediction model for analysis and calculation to obtain: the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz; and the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz are summarized and analyzed to generate the abnormal warning index AlarmIdx;

[0067] The evaluation control module is used to pre-set the early warning evaluation threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning evaluation threshold AlarmK, and take corresponding control plans according to the risk assessment level.

[0068] In this embodiment, the environmental monitoring module uses a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator shaft in real time, and constructs the environmental degradation degree coefficient based on the environmental status data set, and generates an early warning instruction when it exceeds the preset threshold. After receiving the early warning instruction, the operation monitoring module further monitors the operation status data of the elevator, extracts the jitter characteristics, stop layer error characteristics and control delay characteristics in the elevator operation, calculates the elevator state variation index, and issues a second early warning instruction when it exceeds the abnormal operation threshold. The second monitoring module monitors the behavior of passengers in the elevator through an image acquisition device, captures improper operation information, and collects lighting information and smoke concentration information in the elevator. The analysis module establishes a risk prediction model based on the monitoring data set, calculates the environmental status factor and behavior deviation factor in the elevator, and generates an abnormal early warning index. The evaluation control module compares the abnormal early warning index according to the early warning evaluation threshold, obtains the risk assessment level, and takes corresponding control measures. This comprehensive monitoring system helps to timely discover and deal with potential safety hazards and improve the safety management level of special equipment.

[0069] Example 2: This example is an explanation of Example 1. Figure 1 ,Specifically, the environment monitoring module includes a sensor acquisition unit and an ,environmental computing unit;

[0070] The sensor acquisition unit is used to determine the key positions in the elevator shaft according to the internal structure diagram of the elevator shaft, and deploy several groups of environmental monitoring sensors and data acquisition equipment in the elevator shaft to monitor the temperature, humidity, air quality and vibration conditions in the shaft in real time to build an environmental status data set;

[0071] The environmental status data set includes: the surface temperature T of the shaft equipment, the humidity S at the bottom of the shaft, the PM2.5 concentration P in the shaft, and the vibration intensity value Z of the shaft wall.

[0072] The environmental calculation unit is used to pre-process the environmental status data set, extract the surface temperature T of the well equipment, the humidity S at the bottom of the well, the PM2.5 concentration P in the well, and the vibration intensity value Z of the well wall, perform linear normalization, and map the corresponding numerical features to the interval [0,1], and then obtain the environmental degradation degree coefficient HJlh according to the following calculation method:

[0073]

[0074] where \(i = 1, 2, \ldots, n\), and \(n\) is the number of environmental status data collection time nodes; \(T\) i is the surface temperature of the hoistway equipment at the \(i\)-th collection time node, is the ideal temperature threshold of the hoistway equipment surface, \(S\) i represents the humidity at the bottom of the hoistway at the \(i\)-th collection time node, is the ideal humidity threshold of the hoistway bottom, \(P\) i represents the PM2.5 concentration in the hoistway at the \(i\)-th collection time node, is the ideal concentration threshold of PM2.5 in the hoistway, \(Z\) i represents the vibration intensity value of the hoistway wall at the \(i\)-th collection time node, is the vibration intensity threshold borne by the hoistway wall, and \(F1, F2, F3\) and \(F4\) are weight coefficients: \(0 \lt F1 \lt 1\), \(0 \lt F2 \lt 1\), \(0 \lt F3 \lt 1\), \(0 \lt F4 \lt 1\), and \(F1 + F2 + F3 + F4 = 1\);

[0075] At the same time, an environmental degradation threshold \(E\) is preset. When the environmental degradation degree coefficient \(HJlh \gt\) the environmental degradation threshold \(E\), it indicates that the fluctuation of the current hoistway environment is abnormal, and a first warning instruction is generated, including starting the environmental control equipment for real-time control and activating the shock absorption device;

[0076] When the environmental degradation degree coefficient \(HJlh \leq\) the environmental degradation threshold \(E\), it indicates that the fluctuation of the current hoistway environment is normal, and continuous monitoring is carried out.

[0077] In this embodiment, by deploying multiple sets of environmental monitoring sensors in the elevator shaft, the temperature, humidity, air quality and vibration conditions in the shaft can be fully monitored. This comprehensive environmental monitoring can cover all key positions in the shaft, effectively make up for the lack of coverage of the blind spots of the shaft environment by the traditional monitoring system, and improve the perception of the shaft environment status. The environmental monitoring module collects and analyzes environmental data in real time, calculates the environmental degradation degree coefficient HJlh and compares it with the preset environmental degradation threshold E. When an abnormal environmental state is detected, the system can generate a first early warning instruction in time. This real-time early warning mechanism can respond quickly to environmental changes and ensure intervention before potential equipment failures occur, thereby reducing the probability of accidents. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, the system will automatically start the environmental control device to adjust the environmental parameters in the shaft, such as temperature and humidity. In addition, the system will also activate the shock absorbing device to reduce the impact of vibration in the shaft on the elevator equipment. This active control measure can prevent further damage to the equipment due to environmental anomalies and improve the durability and stability of the elevator equipment. By carefully monitoring and controlling the hoistway environment in real time, the negative impact of the environment on the elevator equipment can be effectively reduced, thereby improving the safety and operational reliability of the equipment. Preventive maintenance and timely intervention measures can reduce the frequency of equipment failures and ensure the stability of elevator operation and the safety of passengers.

[0078] Example 3, this example is explained in Example 1, please refer to Figure 1 Specifically, the operation monitoring module includes an operation monitoring unit, a feature extraction unit and an operation calculation unit.

[0079] The operation monitoring unit is used to further monitor the elevator operation status data. By installing an acceleration sensor, a vibration sensor and a displacement sensor at the bottom of the elevator car, the elevator operation acceleration data, the elevator displacement data and the vibration data during operation are collected and acquired to establish an operation data set.

[0080] The feature extraction unit is used to preprocess the relevant data in the operation data set, which includes data cleaning, data standardization and time alignment, and extract the elevator operation jitter characteristics, floor error characteristics and control delay characteristics to obtain: elevator instantaneous acceleration Ss, elevator maximum acceleration Sdmax, elevator average acceleration Savg, elevator vibration frequency Zdpl, elevator vibration intensity Zdqd, vertical displacement difference DCy between the elevator car and the predetermined stop position when it stops at the floor, elevator start delay Qdyc and elevator deceleration delay Jsyc.

[0081] The operation calculation unit is used to extract the elevator instantaneous acceleration Ss, the elevator maximum acceleration Sdmax, the elevator average acceleration Savg, the elevator vibration frequency Zdpl, the elevator vibration intensity Zdqd, the vertical displacement difference DCy between the elevator car and the predetermined stop position when the elevator car stops at the floor, the elevator start delay Qdyc and the elevator deceleration delay Jsyc; after dimensionless processing, the elevator state variation index Pwd is calculated by the following formula:

[0082]

[0083] Wherein: bz1 represents the standard threshold of instantaneous acceleration of the elevator, bz2 represents the maximum acceleration bearing threshold of the elevator, bz3 represents the standard threshold of average acceleration of the elevator, bz4 represents the preset tolerance threshold of the vibration frequency of the elevator, bz5 represents the preset tolerance threshold of the vibration intensity of the elevator, bz6 represents the preset allowable threshold of the vertical displacement difference between the elevator car and the predetermined parking position when the elevator car stops at the floor, bz7 represents the preset allowable threshold of the elevator start delay, bz8 represents the preset allowable threshold of the elevator deceleration delay, w1, w2, w3, w4, w5, w6, w7 and w8 represent weight values, and the sum of the weights is 1;

[0084] At the same time, an abnormal operation threshold K is pre-set. When the elevator status variation index Pwd>abnormal operation threshold K, it indicates that the fluctuation of the current elevator operation is estimated to be abnormal, and a second early warning instruction is generated, including activating the elevator internal alarm system to remind passengers of the operation risk and inform them that the elevator is temporarily out of service. Then, technicians are arranged to conduct on-site inspections of the elevator, check the status of each component, monitor the equipment parts related to acceleration, vibration, displacement and delay, and resume the elevator operation after repair and adjustment. In the initial stage of resuming operation, the monitoring frequency is strengthened to 3-5 times per hour to prevent the problem from happening again.

[0085] When the elevator state variation index Pwd ≤ abnormal operation threshold K, it means that the fluctuation of the current elevator operation is estimated to be normal and continuous monitoring is performed.

[0086] In this embodiment, the operation monitoring module can comprehensively monitor the acceleration, vibration and displacement data of the elevator by installing an acceleration sensor, a vibration sensor and a displacement sensor at the bottom of the elevator car. This meticulous monitoring helps to grasp the operation status of the elevator in real time, provide detailed data on various performance indicators of the elevator, and ensure that the equipment is within the normal operating range. The feature extraction unit extracts the characteristics of the elevator's operation jitter, landing error and control delay by cleaning, standardizing and time aligning the collected operation data. In this way, the key features in the elevator operation, such as instantaneous acceleration, maximum acceleration, average acceleration, vibration frequency, vibration intensity, etc., can be accurately identified, providing a comprehensive operation status evaluation. The operation calculation unit can accurately evaluate the operation fluctuation of the elevator by dimensionlessly processing various operation feature data and calculating the elevator state variation index Pwd. When Pwd exceeds the preset abnormal operation threshold K, the system can issue a second early warning instruction in time. This abnormal detection mechanism can provide early warning when there is a potential fault in the elevator operation, thereby avoiding more serious faults. When the system detects an abnormal operation, it will automatically activate the elevator's internal alarm system to remind passengers that there is an operation risk and temporarily stop the elevator. Technicians will be sent to the site to inspect and repair various components, monitoring parts of the equipment related to acceleration, vibration, displacement and delay. This rapid response and troubleshooting mechanism can effectively prevent accidents and ensure that the elevator can operate safely after repair. In the early stage of the elevator's resumption of operation, the system will strengthen the monitoring frequency of the elevator, monitoring 3-5 times per hour to prevent the problem from happening again. This intensive monitoring helps to detect potential problems after repair and ensure the long-term reliability and stability of the elevator equipment.

[0087] Example 4: This example is an explanation of Example 1. Figure 1 ,Specifically, the second monitoring module includes an image recognition unit and a second sensor acquisition unit;

[0088] The image recognition unit is used to collect and recognize improper operation information of passengers in the elevator through an image acquisition device, including:

[0089] The image acquisition device is used to identify the risk of passengers carrying items, identify and calculate the size and shape of the items, and identify and obtain the area Mj of the items carried; oversized items may hinder the closing of the elevator door or cause the door to be unable to close completely, thus affecting the normal operation of the elevator. In this case, the elevator door may become stuck or unable to close completely, increasing the safety risk to passengers. Excessive size or weight of items may cause an imbalance in the center of gravity of the elevator car. The elevator may become unstable during operation, such as tilting or swinging, which affects the comfort and safety of passengers. Large items moving or falling in the elevator may cause harm to passengers and equipment. Items may slide or tilt in the elevator, increasing the risk of injury to passengers. At the same time, if the items are unstable, they may cause damage to components such as sensors and control systems inside the elevator. For example, furniture and battery vehicles will have potential risks;

[0090] The image acquisition device is used to identify the long time that the passenger presses the floor button. If it exceeds five seconds, it is counted as an improper operation to obtain the number of improper operations bdcz1. The passenger's behavior information of forcibly blocking the elevator door from closing is also collected, including the behavior of using hands, feet or objects to block the elevator door, to obtain the excess delay time Ycsj that interferes with the normal closing of the elevator door;

[0091] The second sensing acquisition unit is used to install a smoke sensor in the elevator to monitor the smoke concentration in the air in real time and obtain the smoke concentration value Ywnd; and install a lighting sensor in the elevator to monitor the lighting brightness in the elevator in real time and obtain the light brightness Gzld;

[0092] The monitoring data set includes: the area of ​​the items carried Mj, the number of improper operations bdcz1, the excess delay time that interferes with the normal closing of the elevator door Ycsj, the smoke concentration value Ywnd and the light brightness Gzld.

[0093] Specifically, the analysis module includes a model building unit, an environmental state factor calculation unit, and an improper behavior calculation unit;

[0094] The model building unit is used to build a risk prediction model through random forest and support vector machine technology, and after data cleaning and normalization of the monitoring data set, it is divided into a training set and a test set for training and then the performance of the model is evaluated. The accuracy, precision, recall and F1 score indicators of the model are calculated and analyzed until the evaluation is qualified;

[0095] The environmental state factor calculation unit is used to extract the smoke concentration value Ywnd and light brightness Gzld in the monitoring data set. After dimensionless processing, the elevator internal environmental state factor BhvPyz is calculated using the following formula:

[0096]

[0097] where \(i = 1, 2, \ldots, p\), and \(p\) is the number of acquisition time nodes of the monitoring dataset; \(d_1\) and \(d_2\) are weight coefficients: \(0 \lt d_1 \lt 1\), \(0 \lt d_2 \lt 1\), and \(d_1 + d_2 = 1\); \(Gzld\) i is the illumination brightness at the \(i\)-th acquisition time node, is the illumination brightness threshold; \(Ywnd\) i represents the smoke concentration value at the \(i\)-th acquisition time node, is the preset threshold of smoke concentration;

[0098] The improper behavior calculation unit is used to extract the carried item area \(Mj\), the number of improper operations \(bdcz1\), and the extra delay time \(Ycsj\) that interferes with the normal closing of the elevator door from the monitoring dataset. After dimensionless processing, the behavior deviation factor \(DevRunz\) is calculated through the following formula:

[0099]

[0100] In the formula, \(Mj\) k represents the carried item area of the \(k\)-th passenger, \(Mj\) max represents the preset threshold of the maximum carried item area allowed for passengers, \(bdcz1\) k represents the number of improper operations of the \(k\)-th passenger, \(bdcz1\) max represents the preset threshold of the maximum number of improper operations allowed for passengers, \(Ycsj\) k represents the extra delay time that the \(k\)-th passenger's behavior causes to interfere with the normal closing of the elevator door, \(Ycsj\) max represents the preset threshold of the maximum extra delay time allowed for passengers to interfere with the normal closing of the elevator door; \(d_3\), \(d_4\), and \(d_5\) are weight coefficients: \(0 \lt d_3 \lt 1\), \(0 \lt d_4 \lt 1\), \(0 \lt d_5 \lt 1\), and \(d_3 + d_4 + d_5 = 1\).

[0101] In this embodiment, the environmental state factor \(BhvPyz\) can comprehensively monitor the environmental conditions in the elevator by analyzing the smoke concentration and illumination brightness in the elevator. The smoke concentration and illumination brightness are monitored in real time to ensure the safety and comfort of the environment in the elevator. For example, a high smoke concentration can timely warn of possible fire hazards, while the monitoring of illumination brightness ensures that the light conditions in the elevator meet safety standards.

[0102] By analyzing the environmental monitoring data through the random forest and support vector machine technologies of the model building unit, a high-precision risk prediction model can be established. This model can accurately predict the environmental risks in the elevator, timely identify potential problems, and provide a reliable basis for subsequent risk control.

[0103] DevRunz, a behavioral deviation factor, can effectively identify improper passenger behavior by monitoring the size of the items carried, the number of improper operations, and the extra delay time that interferes with the normal closing of the elevator door. This identification can help managers promptly detect and deal with improper passenger behavior, such as overly large items that may hinder the normal operation of the elevator, and long-term button pressing that may affect the normal operation of the elevator.

[0104] Example 5: This example is an explanation of Example 4. Please refer to Figure 1 Specifically, the analysis module also includes a correlation unit, which is used to associate the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz dimensionlessly through the following formula to obtain the abnormal warning index AlarmIdx:

[0105]

[0106] Where α and δ represent the weight values ​​of the environmental state factor BhvPyz and the behavioral deviation factor DevRunz, respectively.

[0107] By combining the environmental status factor BhvPyz and the behavioral deviation factor DevRunz to calculate the abnormal warning index AlarmIdx, the real-time risk status of the elevator can be accurately assessed. Based on these data, the system can promptly detect potential safety hazards and evaluate them, thereby improving the overall safety of the elevator.

[0108] The evaluation control module consists of an evaluation unit and a control unit:

[0109] The evaluation unit is used to pre-set the warning evaluation threshold AlarmK, and compare the warning evaluation threshold AlarmK with the abnormal warning index AlarmIdx to obtain the risk assessment level:

[0110] When the abnormal warning index AlarmIdx ≤ the warning assessment threshold AlarmK, it is judged as low risk, the elevator continues to work normally, and continues to monitor;

[0111] When the warning assessment threshold AlarmK < abnormal warning index AlarmIdx ≤ warning assessment threshold AlarmK*130%, it indicates a medium risk, and the control unit triggers the internal risk warning and prompt of the elevator, immediately stops the elevator operation, reminds relevant personnel to check, and restarts the elevator after the improper behavior is eliminated;

[0112] When the abnormal warning index AlarmIdx>the warning assessment threshold AlarmK*130%, it indicates a high-level risk, and the elevator internal risk warning is triggered through the control unit, and the elevator operation is immediately stopped, and the elevator door is set to the open state, and an emergency notification is automatically sent to the emergency service fire department and rescue team, triggering the automatic ventilation system or backup lighting in the elevator. After confirming that the passengers have been safely evacuated, the power of the elevator is remotely shut down to prevent any accidental startup or further damage.

[0113] In this embodiment, according to the comparison between the abnormal warning index and the warning assessment threshold, the system can classify the risk into three levels: low, medium and high. This hierarchical management makes the response measures for different risk levels more specific and appropriate, which helps to take corresponding treatment measures, thereby effectively reducing the potential threats to passengers and equipment. When a medium or high risk is detected, the system can immediately trigger an early warning and take corresponding measures, such as stopping the elevator operation, reminding relevant personnel to check, and starting the emergency notification procedure. This real-time early warning and response mechanism can quickly handle abnormal situations and reduce potential safety accidents. In the case of high-level risks, the system will not only stop the elevator operation, but also set the elevator door to the open state to ensure that passengers can evacuate safely. In addition, the system will automatically notify the emergency service and rescue team, and enable the automatic ventilation system or backup lighting. These measures can effectively ensure the safety of passengers and provide necessary information and support for rescue. By monitoring and analyzing the abnormal warning index, the system can perform preventive maintenance before the problem becomes serious. The elevator will be stopped in time when an abnormal situation is found, avoiding equipment damage caused by long-term potential failures, thereby reducing maintenance costs and equipment downtime.

[0114] Example 6, please refer to Figure 2 A big data monitoring method for safety hazard management and control of special equipment includes the following steps:

[0115] Step 1: Hoistway environment monitoring and early warning: Use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator hoistway in real time to obtain an environmental status data set, and construct an environmental degradation degree coefficient HJlh based on the environmental status data set. At the same time, the environmental degradation threshold E is pre-set. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current hoistway environment is abnormal, and a first early warning instruction is generated;

[0116] Step 2: Operation status warning: After receiving the first warning instruction, further monitor the elevator operation status data, establish an operation data set, pre-process the relevant data in the operation data set, extract the elevator operation jitter characteristics, floor error characteristics and control delay characteristics, and obtain the elevator state variation index Pwd. If the elevator state variation index Pwd exceeds the abnormal operation threshold K, issue a second operation warning instruction;

[0117] Step 3: Elevator internal behavior and environment collection: Use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information on improper operation by passengers, and obtain lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set;

[0118] Step 4: Risk prediction: Establish a risk prediction model, use the monitoring data set as input, input it into the risk prediction model for analysis and calculation to obtain: elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz; summarize and analyze the elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz to generate the abnormal warning index AlarmIdx;

[0119] Step 5. Evaluate and obtain the risk assessment level and then respond and control: Pre-set the early warning assessment threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning assessment threshold AlarmK, and take corresponding control measures according to the risk assessment level.

[0120] In this embodiment, the method realizes comprehensive monitoring of the elevator shaft environment, operating status and internal passenger behavior through steps one to three. The combination of shaft environment monitoring, operating status warning and internal behavior collection provides a comprehensive monitoring perspective, which helps to timely discover potential safety hazards. Through the risk prediction model in step four, the monitoring data (environmental status, behavioral deviation, etc.) from different data sources are comprehensively analyzed to generate an abnormal warning index AlarmIdx. This data fusion improves the accuracy of identifying potential risks and avoids important information that may be missed by a single data source. In step five, the abnormal warning index AlarmIdx is compared by the preset warning evaluation threshold AlarmK, and the risk can be divided into different levels (low, medium, and high). This hierarchical risk handling mechanism enables the system to take appropriate control measures according to the actual risk situation, such as stopping the elevator operation, issuing an alarm, enabling emergency response, etc., thereby improving the flexibility and effectiveness of safety management. When the risk level is high, the system can quickly generate warning instructions and trigger corresponding emergency response measures, such as stopping the elevator, activating the alarm system, notifying relevant personnel, etc. This rapid response mechanism can effectively prevent the problem from further deteriorating, protect passenger safety, and reduce potential losses.

[0121] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0122] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula that is close to the actual value. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited to this. Any technical personnel familiar with the technical field within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, make equivalent replacement or change, which should be covered within the protection scope of the present invention.

Claims

1. A big data monitoring system for safety hazard management and control of special equipment, characterized by: It includes an environment monitoring module, an operation monitoring module, a second monitoring module, an analysis module and an evaluation control module; The environmental monitoring module is used to use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator shaft in real time to obtain an environmental status data set, and to construct an environmental degradation degree coefficient HJlh based on the environmental status data set, and to pre-set an environmental degradation threshold E. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current shaft environment is abnormal, and a first warning instruction is generated; The operation monitoring module further monitors the elevator operation status data after receiving the first warning instruction, establishes an operation data set, pre-processes the relevant data in the operation data set, extracts the elevator operation jitter characteristics, the stop floor error characteristics and the control delay characteristics, obtains the elevator state variation index Pwd, and issues a second operation warning instruction if the elevator state variation index Pwd exceeds the abnormal operation threshold K; The second monitoring module is used to use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information about improper operation of passengers, and acquire lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set; The analysis module is used to establish a risk prediction model, and input the monitoring data set as an input item into the risk prediction model for analysis and calculation to obtain: the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz; and summarize and analyze the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz to generate an abnormal warning index AlarmIdx; The evaluation control module is used to pre-set the early warning evaluation threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning evaluation threshold AlarmK, and adopt a corresponding control plan according to the risk assessment level.

2. A special equipment safety hazard management and control big data monitoring system according to claim 1, characterized in that: The environment monitoring module includes a sensor acquisition unit and an environment calculation unit; The sensor acquisition unit is used to determine the key positions in the elevator shaft according to the internal structure diagram of the elevator shaft, and deploy several groups of environmental monitoring sensors and data acquisition equipment in the elevator shaft to monitor the temperature, humidity, air quality and vibration conditions in the shaft in real time to construct an environmental status data set; The environmental status data set includes: the surface temperature T of the shaft equipment, the humidity S at the bottom of the shaft, the PM2.5 concentration P in the shaft, and the vibration intensity value Z of the shaft wall.

3. A special equipment safety hazard management and control big data monitoring system according to claim 2, characterized in that: The environmental calculation unit is used to pre-process the environmental status data set, extract the surface temperature T of the well equipment, the humidity S at the bottom of the well, the PM2.5 concentration P in the well, and the vibration intensity value Z of the well wall, perform linear normalization, and map the corresponding numerical features to the interval [0,1], and then obtain the environmental degradation degree coefficient HJlh according to the following calculation method: where \(i = 1, 2, \ldots, n\), and \(n\) is the number of environmental status data collection time nodes; \(T\) i is the surface temperature of the hoistway equipment at the \(i\)-th collection time node, is the ideal temperature threshold of the hoistway equipment surface, \(S\) i represents the humidity at the bottom of the hoistway at the \(i\)-th collection time node, is the ideal humidity threshold at the bottom of the hoistway, \(P\) i represents the PM2.5 concentration in the hoistway at the \(i\)-th collection time node, is the ideal PM2.5 concentration threshold in the hoistway, \(Z\) i represents the vibration intensity value of the hoistway wall at the \(i\)-th collection time node, is the bearing vibration intensity threshold of the hoistway wall, and \(F1, F2, F3\) and \(F4\) are weight coefficients: \(0 < F1 < 1\), \(0 < F2 < 1\), \(0 < F3 < 1\), \(0 < F4 < 1\), and \(F1 + F2 + F3 + F4 = 1\); At the same time, an environmental degradation threshold E is preset. When the environmental degradation degree coefficient HJlh> the environmental degradation threshold E, it is estimated that the fluctuation of the current well environment is abnormal, and a first early warning instruction is generated, including starting the environmental control equipment for real-time control and activating the shock absorbing device; When the environmental degradation degree coefficient HJlh≤the environmental degradation threshold E, it indicates that the fluctuation of the current well environment is estimated to be normal and continuous monitoring is performed.

4. A special equipment safety hazard management and control big data monitoring system according to claim 1, characterized in that: The operation monitoring module includes an operation monitoring unit, a feature extraction unit and an operation calculation unit. The operation monitoring unit is used to further monitor the elevator operation status data, by installing an acceleration sensor, a vibration sensor and a displacement sensor at the bottom of the elevator car, to collect and obtain the elevator operation acceleration data, the elevator displacement data and the vibration data during the operation, and establish an operation data set; The feature extraction unit is used to preprocess the relevant data in the operation data set, and the preprocessing includes data cleaning, data standardization and time alignment, and extract the elevator operation jitter characteristics, floor error characteristics and control delay characteristics to obtain: elevator instantaneous acceleration Ss, elevator maximum acceleration Sdmax, elevator average acceleration Savg, elevator vibration frequency Zdpl, elevator vibration intensity Zdqd, vertical displacement difference DCy between the elevator car and the predetermined stop position when it stops at the floor, elevator start delay Qdyc and elevator deceleration delay Jsyc.

5. A special equipment safety hazard management and control big data monitoring system according to claim 4, characterized in that: The operation calculation unit is used to extract the elevator instantaneous acceleration Ss, the elevator maximum acceleration Sdmax, the elevator average acceleration Savg, the elevator vibration frequency Zdpl, the elevator vibration intensity Zdqd, the vertical displacement difference DCy between the elevator car and the predetermined parking position when the elevator car stops at the floor, the elevator start delay Qdyc and the elevator deceleration delay Jsyc; after dimensionless processing, the elevator state variation index Pwd is calculated by the following formula: Wherein: bz1 represents the standard threshold of instantaneous acceleration of the elevator, bz2 represents the maximum acceleration bearing threshold of the elevator, bz3 represents the standard threshold of average acceleration of the elevator, bz4 represents the preset tolerance threshold of the vibration frequency of the elevator, bz5 represents the preset tolerance threshold of the vibration intensity of the elevator, bz6 represents the preset allowable threshold of the vertical displacement difference between the elevator car and the predetermined parking position when the elevator car stops at the floor, bz7 represents the preset allowable threshold of the elevator start delay, bz8 represents the preset allowable threshold of the elevator deceleration delay, w1, w2, w3, w4, w5, w6, w7 and w8 represent weight values, and the sum of the weights is 1; At the same time, an abnormal operation threshold K is pre-set. When the elevator state variation index Pwd> the abnormal operation threshold K, it indicates that the fluctuation of the current elevator operation is estimated to be abnormal, and a second early warning instruction is generated, including activating the elevator internal alarm system to remind passengers of the operation risk and inform them that the elevator is temporarily out of service, and then arranging technicians to conduct on-site inspections of the elevator, check the status of each component, monitor the equipment parts related to acceleration, vibration, displacement and delay, and resume the elevator operation after maintenance and adjustment. In the initial stage of resuming operation, the monitoring frequency is strengthened to 3-5 times per hour to prevent the problem from happening again; When the elevator state variation index Pwd≤the abnormal operation threshold K, it is estimated that the fluctuation of the current elevator operation is normal and the monitoring is continued.

6. A special equipment safety hazard management and control big data monitoring system according to claim 1, characterized in that: The second monitoring module includes an image recognition unit and a second sensor acquisition unit; The image recognition unit is used to collect and recognize improper operation information of passengers in the elevator through an image acquisition device, including: Identify the risk of items carried by passengers through image acquisition equipment, identify and calculate the size and shape of items, and identify and obtain the area Mj of the items carried; The image acquisition device is used to identify the long time that the passenger presses the floor button. If it exceeds five seconds, it is counted as an improper operation to obtain the number of improper operations bdcz1. The passenger's behavior information of forcibly blocking the elevator door from closing is also collected, including the behavior of using hands, feet or objects to block the elevator door, to obtain the excess delay time Ycsj that interferes with the normal closing of the elevator door; The second sensing acquisition unit is used to monitor the smoke concentration in the air in real time by installing a smoke sensor in the elevator to obtain a smoke concentration value Ywnd; and to monitor the lighting brightness in the elevator in real time by installing a lighting sensor in the elevator to obtain a light brightness Gzld; The monitoring data set includes: the area of ​​the carried items Mj, the number of improper operations bdcz1, the excess delay time Ycsj that interferes with the normal closing of the elevator door, the smoke concentration value Ywnd and the light brightness Gzld.

7. A special equipment safety hazard management and control big data monitoring system according to claim 6, characterized in that: The analysis module includes a model building unit, an environmental state factor calculation unit and an improper behavior calculation unit; The model building unit is used to build a risk prediction model through random forest and support vector machine technology, and after data cleaning and normalization of the monitoring data set, divide it into a training set and a test set for training, and then evaluate the performance of the model, calculate and analyze the accuracy, precision, recall rate and F1 score indicators of the model until the evaluation is qualified; The environmental state factor calculation unit is used to extract the smoke concentration value Ywnd and the light brightness Gzld in the monitoring data set, and after dimensionless processing, calculate the elevator internal environmental state factor BhvPyz through the following formula: where \(i = 1, 2, \ldots, p\), \(p\) is the number of acquisition time nodes of the monitoring data set; \(d_1\) and \(d_2\) are weight coefficients: \(0 < d_1 < 1\), \(0 < d_2 < 1\), and \(d_1 + d_2 = 1\); \(G_{zld}\) i is the light intensity at the \(i\)-th acquisition time node, is the light intensity threshold; \(Y_{wnd}\) i represents the smoke concentration value at the \(i\)-th acquisition time node, is the preset threshold of smoke concentration; The improper behavior calculation unit is used to extract the area of ​​the carried items Mj, the number of improper operations bdcz1, and the redundant delay time Ycsj that interferes with the normal closing of the elevator door in the monitoring data set, and after dimensionless processing, the behavior deviation factor DevRunz is calculated by the following formula: where Mj k represents the area of the carried items of the k-th passenger, and Mj max represents the threshold value of the maximum allowable area of the carried items of the passenger, bdcz1 k represents the number of improper operations of the k-th passenger, and bdcz1 max represents the threshold value of the maximum allowable number of improper operations of the passenger, Ycsj k represents the extra delay time caused by the behavior of the k-th passenger that interferes with the normal closing of the elevator door, and Ycsj max represents the threshold value of the maximum allowable extra delay time for the passenger to interfere with the normal closing of the elevator door; d3, d4, and d5 are weighting coefficients: 0 < d3 < 1, 0 < d4 < 1, 0 < d5 < 1, and d3 + d4 + d5 = 1.

8. A special equipment safety hazard management and control big data monitoring system according to claim 7, characterized in that: The analysis module further includes a correlation unit, which is used to associate the elevator internal environment state factor BhvPyz and the behavior deviation factor DevRunz dimensionlessly through the following formula to obtain the abnormal warning index AlarmIdx: Where α and δ represent the weight values ​​of the environmental state factor BhvPyz and the behavioral deviation factor DevRunz, respectively.

9. A special equipment safety hazard management and control big data monitoring system according to claim 1, characterized in that: The evaluation control module includes an evaluation unit and a control unit: The evaluation unit is used to pre-set the warning evaluation threshold AlarmK, and compare the warning evaluation threshold AlarmK with the abnormal warning index AlarmIdx to obtain the risk evaluation level: When the abnormal warning index AlarmIdx ≤ the warning assessment threshold AlarmK, it is judged as low risk, the elevator continues to work normally, and continues to monitor; When the warning assessment threshold AlarmK < abnormal warning index AlarmIdx ≤ warning assessment threshold AlarmK*130%, it indicates a medium risk, and the control unit triggers the internal risk warning and prompt of the elevator, immediately stops the elevator operation, reminds relevant personnel to check, and restarts the elevator after the improper behavior is eliminated; When the abnormal warning index AlarmIdx>the warning assessment threshold AlarmK*130%, it indicates a high-level risk, and the elevator internal risk warning is triggered through the control unit, and the elevator operation is immediately stopped, and the elevator door is set to the open state, and an emergency notification is automatically sent to the emergency service fire department and rescue team, triggering the automatic ventilation system or backup lighting in the elevator. After confirming that the passengers have been safely evacuated, the power of the elevator is remotely shut down to prevent any accidental startup or further damage.

10. A special equipment safety hazard management and control big data monitoring method, applied to a special equipment safety hazard management and control big data monitoring system according to any one of claims 1 to 9, characterized in that: The following steps are included: Step 1: Hoistway environment monitoring and early warning: Use a variety of sensor devices to monitor the temperature, humidity, air quality and vibration conditions in the elevator hoistway in real time to obtain an environmental status data set, and construct an environmental degradation degree coefficient HJlh based on the environmental status data set. At the same time, an environmental degradation threshold E is pre-set. When the environmental degradation degree coefficient HJlh exceeds the environmental degradation threshold E, it indicates that the fluctuation of the current hoistway environment is estimated to be abnormal, and a first early warning instruction is generated; Step 2, operation status warning: after receiving the first warning instruction, further monitor the elevator operation status data, establish an operation data set, pre-process the relevant data in the operation data set, extract the elevator operation jitter characteristics, stop floor error characteristics and control delay characteristics, obtain the elevator state variation index Pwd, if the elevator state variation index Pwd exceeds the abnormal operation threshold K, issue a second operation warning instruction; Step 3: Elevator internal behavior and environment collection: Use image acquisition equipment to capture the behavior of passengers in the elevator from multiple angles, collect information on improper operation by passengers, and obtain lighting information and smoke concentration information in the elevator through sensors to establish a monitoring data set; Step 4: Risk prediction: Establish a risk prediction model, use the monitoring data set as input, input it into the risk prediction model for analysis and calculation to obtain: elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz; summarize and analyze the elevator internal environment state factor BhvPyz and behavior deviation factor DevRunz to generate the abnormal warning index AlarmIdx; Step 5. Evaluate and obtain the risk assessment level and then respond and control: Pre-set the early warning assessment threshold AlarmK, obtain the risk assessment level by comparing the abnormal early warning index AlarmIdx with the early warning assessment threshold AlarmK, and take corresponding control measures according to the risk assessment level.

Images