Environment safety assessment early warning system and method applied to emergency rescue drill site

Abstract

The invention relates to the field of gas monitoring, surrounding rock monitoring and evaluation early warning, in particular to an environment safety evaluation early warning system and method applied to an emergency rescue drill site, and the system comprises a gas monitoring module, a surrounding rock deformation monitoring module and an evaluation early warning software module. The gas monitoring module comprises an electrochemical gas sensor, a 433 wireless transmission module, a GPRS wireless transmission module and a gas concentration data acquisition module; the surrounding rock deformation monitoring module comprises a target lamp, a power supply module, a monitoring camera and a distribution box; and the evaluation and early warning software module is used for realizing functions of data receiving and recording, environmental data input, index grade determination and overall safety evaluation and prediction of the emergency rescue drill site environment. By monitoring the concentration of carbon dioxide, methane, hydrogen sulfide and oxygen in the environment of the emergency rescue drill site, the deformation rate and the deformation amount of surrounding rock and combining the environmental index grade of the field influencing the safety of the emergency rescue drill site, the environment of the emergency rescue drill site is evaluated and early warned by utilizing an evaluation and data prediction method.

Description

Technical Field

[0001] The present invention relates to the fields of gas monitoring, surrounding rock monitoring and evaluation and early warning, and particularly relates to an environmental safety evaluation and early warning system and method applied to an emergency rescue drill site. Background Art

[0002] With the large-scale and complexity of modern engineering construction projects, especially in the fields of mine exploitation, underground engineering construction and oil drilling, etc., how to ensure the safety of the rescue environment and rescue personnel and prevent the occurrence of secondary disasters after a sudden accident has occurred has become a key technical problem.

[0003] Secondly, for the monitoring of surrounding rock deformation, it is all applied to the construction aspect. Considering from the rescue perspective, the deformation monitoring of the rescue drill site is crucial. Therefore, there is an urgent need to study an environmental safety evaluation and early warning method applied to an emergency rescue drill site. Summary of the Invention

[0004] In order to improve the emergency rescue efficiency, the present invention provides an environmental safety evaluation and early warning method applied to an emergency rescue drill site, a safety evaluation and early warning system integrating functions such as gas monitoring and surrounding rock monitoring, provides an innovative solution for engineering safety management, can collect, process and display multi-dimensional monitoring data, and provide intelligent early warning based on an evaluation model, thereby optimizing emergency response and decision support.

[0005] An environmental safety evaluation and early warning system applied to an emergency rescue drill site includes: a gas monitoring module, a surrounding rock deformation monitoring module, and an evaluation and early warning software module;

[0006] The gas monitoring module includes an electrochemical gas sensor, a gas concentration data acquisition module, a 433 wireless transmission module, and a GPRS wireless transmission module; the gas concentration data acquisition module is used to receive data and supply power to the electrochemical gas sensor, the 433 wireless transmission module, and the GPRS wireless transmission module; the 433 wireless transmission module includes receiving and sending antennas and is used to wirelessly transmit data to a local host computer; the GPRS wireless transmission module includes a sending antenna and an in-built wireless network card and is used to wirelessly transmit data to the cloud;

[0007] The surrounding rock deformation monitoring module includes a target lamp, a power supply module, a monitoring camera, and a distribution box. The target lamp is used to emit laser; the power supply module is used to supply power to the target lamp or charge a battery; the monitoring camera is used to receive the laser of the target lamp; the distribution box includes a data acquisition module and a wireless transmission module and supplies power to the monitoring camera;

[0008] An evaluation and early warning software module, which is used to receive the data monitored by the gas monitoring module and the surrounding rock deformation monitoring module, and realizes the functions of data reception and recording, environmental data input, index level determination, and overall safety assessment and prediction of the emergency rescue drill site environment.

[0009] An environmental safety evaluation and early warning method applied to an emergency rescue drill site, which is realized based on the above-mentioned environmental safety evaluation and early warning system. The method includes the following steps:

[0010] S101, collect and integrate gas monitoring and surrounding rock deformation monitoring data;

[0011] S102, establish an evaluation index system for the emergency rescue drill site;

[0012] The index system includes two primary indexes: risk-pregnant environment and risk-causing factors. The primary indexes are further subdivided into the following secondary indexes and index systems:

[0013] Secondary indexes of the risk-pregnant environment: surrounding rock level, degree of rock mass fragmentation, hardness of the rock mass, underground water volume, original support measures for the surrounding rock, drill site reinforcement, fault fracture zone;

[0014] Secondary indexes of the risk-causing factors: ventilation volume, vibration speed of the drill site, amount of surrounding rock deformation, rate of surrounding rock deformation, oxygen concentration, hydrogen sulfide concentration, methane concentration, carbon monoxide concentration, degree of standardization of on-site personnel operations;

[0015] After establishing the risk evaluation index system, use the AHP hierarchical analysis method to discretize the indexes; and combine the questionnaire form of expert questionnaire scoring to fuzzify the evaluation indexes, and calculate the weights of each index through the combined weighting method.

[0016] S103, evaluate the safety situation of the emergency rescue drill site according to the risk events, index values, and optimal values;

[0017] S104, according to the data change situation, use the ARIMAX multi-factor time series prediction model to predict the monitored surrounding rock deformation and gas concentration data. According to the continuously predicted monitored data, combined with the evaluation indexes, predict the safety situation of the emergency rescue drill site, and give an early warning when it exceeds the safety setting requirements.

[0018] A computer device includes a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps of the above method.

[0019] A computer-readable storage medium stores a computer program, and when the program is executed by a processor, it implements the steps of the above method.

[0020] A computer program product includes a computer program or instruction, and when the program or instruction is executed by a processor, it implements the steps of the above method.

[0021] The beneficial effects brought by the technical solution provided by the present invention are as follows: The present invention can collect and integrate four harmful gases, namely methane, hydrogen sulfide, oxygen, and carbon monoxide, which are common risk-causing factors in the emergency rescue drill site environment, as well as the amount and rate of surrounding rock deformation. By adjusting the risk-bearing environment indicators actually existing at the emergency rescue site, the evaluation and early warning system can be adjusted to adapt to the complex and changeable safety environment of the emergency rescue drill site; through various data transmission modes, the present invention can transmit and summarize data more safely and effectively; through a more efficient dynamic algorithm, the situation of risk prediction and early warning can be predicted; it is beneficial to realize the evaluation of the safety status of the emergency rescue drill site, predict the future safety situation of the emergency rescue drill site, give certain guiding opinions, and play an effective reference and guiding role for on-site decision-making to prevent secondary disasters. Moreover, by using lithium batteries, power can be supplied to some remote components, enabling short-term off-grid use and providing time for subsequent line laying. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0023] Figure 1 is the structural diagram of the emergency rescue drill site and safety monitoring hardware system in the embodiment of the present invention;

[0024] Figure 2 is the flowchart of the environmental safety evaluation and early warning method applied to the emergency rescue drill site in the embodiment of the present invention;

[0025] Figure 3 is the schematic diagram of the gas monitoring data and prediction data curve in the embodiment of the present invention;

[0026] Figure 4 is the schematic diagram of the surrounding rock monitoring data curve in the embodiment of the present invention;

[0027] Figure 5 is the calculation flowchart of the local software in the embodiment of the present invention;

[0028] Figure 6 is the structural schematic diagram of the preferred gas monitoring module in the embodiment of the present invention;

[0029] Figure 7 is the structural schematic diagram of the preferred surrounding rock deformation monitoring module in the embodiment of the present invention.

[0030] In the drawings: 1 is a target lamp, 2 is an electrochemical gas sensor with four gas probes, 3 is a DS1004 acquisition box, 4 is an evaluation and early warning software module installed on the drilling rig computer, 5 is a surrounding rock deformation monitoring sensor, and 6 is a distribution box. DETAILED DESCRIPTION OF THE INVENTION

[0031] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0032] Example 1

[0033] At present, there are very few studies on the environmental safety of emergency rescue drilling sites. After an accident, the safety issues of rescue personnel and the losses of rescue drilling rigs caused by secondary disasters are even more difficult to estimate. In order to prevent and reduce the occurrence of secondary disasters, ensure the safety of rescue personnel and the progress of rescue, this paper takes the gas concentration data and surrounding rock deformation data measured in a tunnel and on-site in Yunnan as an example to explore the types and weights of hazardous environments and disaster-causing factors, and constructs a safety evaluation index system suitable for emergency rescue drilling sites. AHP is used to construct the index system, Topsis is used for evaluation and calculation, ARMA and exponential smoothing methods are used to predict data with different trends, and the safety situation of emergency rescue drilling sites is evaluated, predicted and warned. The self-developed evaluation and warning software and cloud platform are integrated to evaluate, warn and display the safety situation of emergency rescue drilling sites.

[0034] The present invention relates to an environmental safety assessment and early warning method for emergency rescue drilling sites. According to the hazardous environment and disaster-causing factors at the site, the indicator weights calculated based on expert scoring are preset (the user can modify the weight value, delete or add indicators at will), and the algorithm is programmed and integrated into the software framework to achieve the assessment and early warning of the safety situation of the emergency rescue drilling site, and verified in combination with the on-site situation. The system structure diagram is shown in the figure. Figure 1 As shown, it includes: a gas monitoring module, a surrounding rock deformation monitoring module 5, and an evaluation and early warning software module 4. The data monitored by the gas monitoring module and the surrounding rock deformation monitoring module are input into the evaluation and early warning software module, and the early warning information of the emergency rescue drilling site is output. In this embodiment, the surrounding rock deformation monitoring module is a surrounding rock deformation monitoring sensor 5.

[0035] The gas monitoring module includes an electrochemical gas sensor 2 with four gas probes, a gas concentration data acquisition module, a 433 wireless transmission module, and a GPRS wireless transmission module; the gas concentration data acquisition module has a built-in lithium battery for receiving data and powering the electrochemical gas sensor, the 433 wireless transmission module, and the GPRS wireless transmission module; the 433 wireless transmission module includes receiving and transmitting antennas for wirelessly transmitting data to a local host computer; the GPRS wireless transmission module includes a transmitting antenna and a built-in wireless network card for wirelessly transmitting data to the cloud. Figure 6 As shown, the gas concentration data acquisition module receives the gas concentrations monitored by the oxygen, carbon monoxide, hydrogen sulfide, and methane sensors, and transmits them to the host computer through the 433 wireless transmission module or the 485 wireless transmission module, or can be uploaded to the cloud through the GPRS wireless transmission module.

[0036] Preferably, the gas monitoring module has a high-density aluminum alloy housing, which can protect the gas sensor and provide explosion-proof function, and the explosion-proof rating ensures the stable operation of the function in the emergency rescue drill site environment.

[0037] Preferably, the electrochemical gas sensor 2 with four gas probes, model Pilitong PLT239; Equipment operating environment conditions: Temperature: -20~50℃ Humidity: 10~95%RH no condensation; Dimensions: 280*180*100(mm); Weight: about 1.8kg; Power supply: DC24V or AC220V (external power adapter); Explosion-proof rating: Exd IC T6 Gb; Protection rating: IP65 dust-proof, splash-proof, explosion-proof; Response time: T<30S; Alarm methods: on-site sound and light alarm, external alarm, remote controller alarm, data acquisition alarm, etc.; It can integrate four electrochemical gas sensor probes with different gas types and different concentration ranges, has a 485 interface, and can transmit the collected data.

[0038] Preferably, the methane gas electrochemical gas sensor probe, range 3%-100%LEL, accuracy 0.1LEL, alarm concentration.

[0039] Preferably, the carbon monoxide electrochemical gas sensor probe, range 0-2000ppm, accuracy 1ppm.

[0040] Preferably, the hydrogen sulfide electrochemical gas sensor probe, range 0-200ppm, accuracy 0.1ppm.

[0041] Preferably, the oxygen electrochemical gas sensor probe, range 0-25%VOL, accuracy 0.1%VOL.

[0042] Preferably, the gas concentration data acquisition module, selects the DS1004 acquisition box 3, can receive four-way sensor signals and power the sensors, including 2*12V+2*24V four-way, can be compatible with various types of sensors; equipped with 485 interface, 433 antenna interface, GPRS module interface, can realize multiple data transmission methods such as local limited, local wireless, and cloud upload; the built-in module can realize multiple upload intervals of 5s-1d; configure a USB interface to connect a USB flash drive, and can locally record data without a host computer; built-in lithium battery, can power the connected devices, and can work for 3h at an upload frequency of 5s / time.

[0043] Preferably, the GPRS wireless transmission module, built-in Internet of Things card, can wirelessly transmit data to the cloud.

[0044] Preferably, the 433 wireless transmission module, can wirelessly transmit data to the local host computer.

[0045] The surrounding rock deformation monitoring module 5 includes a target light 1, a power supply module, a monitoring camera, and a distribution box (i.e., the substation box calculation module in Figure 7 ). The target light is used to emit laser light; the power supply module is used to supply power to the target light or charge the battery; the monitoring camera is used to receive the laser light of the target light; the distribution box includes a data acquisition module and a wireless transmission module, and supplies power to the monitoring camera. As shown in Figure 7 , the target light emits laser light. According to the position information obtained by the camera for surrounding rock deformation monitoring, deformation data is obtained through the distribution box, and the obtained surrounding rock deformation data is transmitted to the upper computer, and the surrounding rock deformation data is uploaded to the cloud through the GPRS wireless transmission module. Figure 7 (the substation box calculation module in Figure 7 ), the target light is used to emit laser light; the power supply module, which is used to supply power to the target light or charge the battery; the monitoring camera, which is used to receive the laser light of the target light; the distribution box contains a data acquisition module, a wireless transmission module, and supplies power to the monitoring camera. As shown in Figure 7 , the target light emits laser light. According to the position information obtained by the camera for surrounding rock deformation monitoring, deformation data is obtained through the distribution box, and the obtained surrounding rock deformation data is transmitted to the upper computer, and the surrounding rock deformation data is uploaded to the cloud through the GPRS wireless transmission module. Figure 7 As shown in Figure 7 , the target light emits laser light. According to the position information obtained by the camera for surrounding rock deformation monitoring, deformation data is obtained through the distribution box, and the obtained surrounding rock deformation data is transmitted to the upper computer, and the surrounding rock deformation data is uploaded to the cloud through the GPRS wireless transmission module.

[0046] Preferably, the target light 1 can be set to emit frequency / constant on, has a built-in battery, and can supply power for 1 month at 1 time per day.

[0047] Preferably, the monitoring camera can receive the position data of the target light. When staying still, the distance accuracy is 0.1 mm at a distance of 20 - 50 m, and it can receive the position data of <50 target lights simultaneously to obtain the deformation data of the surrounding rock.

[0048] The distribution box 6 uses a metal shell to supply power to the displacement monitoring camera; the distribution box has a built-in operation module, which can organize and store the original data collected by the sensor, calculate the relative displacement of the monitoring point through a graphic algorithm, and then realize the surrounding rock deformation monitoring through a geometric algorithm; it has a built-in 4G module and a built-in wireless network card, and can set the start time, frequency, or continuous working time to upload the relative displacement data and surrounding rock deformation data of each measuring point.

[0049] The evaluation and early warning software module 4 is used to realize the functions of data reception and recording, environmental data input, index level determination, and overall safety assessment and prediction of the emergency rescue drill site environment. The four functions are specifically as follows:

[0050] Data integration and display: Integrate the data of different monitoring modules, display the gas concentrations, displacements of each measuring point, and the positions of selected measuring points on the cross-section, and then calculate the deformation amount and deformation rate of the surrounding rock.

[0051] Construction of the risk assessment index system for the emergency rescue drill site: Select and input different environmental conditions (including risk-bearing environments and risk-causing factors), and establish an evaluation index system suitable for the emergency rescue drill site in combination with the monitoring data in 1.

[0052] Risk assessment of the emergency rescue drill site: Grade the values of each index according to the on-site conditions; evaluate the risk levels of each risk event on-site, and comprehensively determine the risk level of the emergency rescue drill site.

[0053] Risk prediction and early warning for emergency rescue drill sites: Combining the data volume, data form, and data trends of monitoring data, using corresponding algorithms in combination with historical data to obtain an expression for data fitting, thereby predicting subsequent data, and verifying the predicted data with subsequent monitoring data, continuously and dynamically updating the calculation method and expression; substituting the predicted data into the evaluation system for fuzzy calculation to obtain the safety status of the emergency rescue drill site in the next stage, achieving the prediction of possible risks and giving early warnings in case of poor conditions.

[0054] The data integration platform collects, integrates, and uploads data to the cloud. It has all the functions of the software and backs up data in the cloud. The storage time of data backup is determined according to the ordered server resources; it can remotely view the data of each project in a list.

[0055] The self-developed and self-programmed data integration platform and evaluation software can monitor, evaluate, and predict the safety situation of the emergency rescue drill site locally in real time, update in real time, and give specific and relatively basic disposal measures.

[0056] It can also adjust the indicators, indicator levels, ideal values of indicators, and relative importance of indicators (thereby calculating the indicator weights) by itself.

[0057] The data integration platform collects, integrates, and uploads data to the cloud. It has all the functions of the software and backs up data in the cloud. The storage time of data backup is determined according to the ordered server resources; it can remotely view the data of each project in a list.

[0058] Specifically, it is applied to the environmental safety assessment and early warning system for emergency rescue drill sites, such as Figure 1 shown, and the implementation steps are as follows:

[0059] 1) Arrange the electrochemical gas sensor 2 with four gas probes and the surrounding rock deformation monitoring target lamp 1.

[0060] 2) Arrange the DS1004 acquisition box 3.

[0061] The specific method is: Connect the gas sensor 2 to the DS1004 acquisition box 3, and the DS1004 acquisition box 3 is connected to the 485 / 433 and GPRS wireless transmission modules to transmit the data to the local host computer and the cloud; when necessary, connect the acquisition module to the power supply to ensure long-term operation.

[0062] 3) Arrange the surrounding rock deformation monitoring sensor 5 and the distribution box 6, and the distribution box 6 is connected to the 220 power supply.

[0063] 4) Calibrate the surrounding rock deformation monitoring camera and the target lamp.

[0064] The specific method is: Adjust the surrounding rock deformation monitoring camera to directly face the target lamp, which means that in the monitoring image, it is near when the brightness of the target lamp is the largest; when necessary, connect the target lamp to the power supply to ensure long-term operation.

[0065] 5) Turn on the computing module of the computer in the distribution box 6, the GPRS and wifi wireless transmission modules. The computing module processes and integrates data, cleans, stores, and visualizes the data, and uploads it to the cloud and the host computer. The schematic diagram of the gas monitoring data and prediction data curve is as Figure 3 shown.

[0066] 6) Turn on the early warning software module 4 of the software platform in the host computers for surrounding rock deformation monitoring and gas monitoring to perform program settings. After determining the monitoring data upload time interval and the position of the relatively stable points for surrounding rock deformation monitoring, the monitoring work can be carried out. The schematic diagram of the surrounding rock monitoring data curve obtained is as Figure 4 shown.

[0067] 7) The local host computer turns on the emergency rescue drill site safety assessment and early warning software of the early warning software module 4. The monitoring data is automatically imported, and risk events are set. In the assessment module, determine the evaluation indicators and their weights (manually or automatically using the default indicators and weights) to evaluate the safety of the emergency rescue drill site; in the early warning module, the prediction data and the predicted next-stage emergency rescue drill site safety situation can be viewed. The software calculation process is as Figure 5 shown. The entire system is divided into a risk assessment system and a risk prediction system. In the risk assessment system, indicators can be added / screened / edited after starting, and then the original data table of the indicators is imported to calculate the ideal solutions of each indicator, fill in the importance of the indicators to calculate the weights, and finally obtain the assessment level. In the risk prediction system, historical data is imported and processed through the ARMA model and the exponential smoothing model to obtain the risk prediction curve graph.

[0068] 8) The monitoring data is uploaded to the cloud through the DS1004 acquisition box 3 and the GPRS module in the distribution box 6. The cloud integrates the core algorithm functions of the emergency rescue drill site safety assessment and early warning software, can back up the monitoring data according to the situation, and remotely view the emergency rescue drill site safety assessment and early warning situation.

[0069] The risk assessment module performs hierarchical analysis on the data in combination with the indicator system and outputs the risk assessment results to the intelligent early warning module. The intelligent early warning module generates an early warning based on the assessment results and the prediction model and notifies it to the system interaction and control module. The assessment decision support module provides optimization suggestions based on the assessment and early warning results and realizes on-site guidance through the system interaction and control module.

[0070] Embodiment 2

[0071] An environmental safety assessment and early warning method applied to an emergency rescue drill site, implemented based on the described environmental safety assessment and early warning system. The overall process is as Figure 2 shown. The method includes the following steps:

[0072] S101, Collect and integrate gas monitoring and surrounding rock deformation monitoring data.

[0073] S102, Establish an evaluation index system for emergency rescue drill sites.

[0074] The said index system includes two primary indexes: risk-bearing environment and risk-causing factors, and the primary indexes are further broken down into the following secondary indexes and index system:

[0075] Secondary indexes of risk-bearing environment: surrounding rock grade, degree of rock mass fragmentation, hardness of rock mass, underground water volume, original support measures for surrounding rock, drill site reinforcement, fault fracture zone;

[0076] Secondary indexes of risk-causing factors: ventilation volume, vibration speed of drill site, deformation amount of surrounding rock, deformation rate of surrounding rock, oxygen concentration, hydrogen sulfide concentration, methane concentration, carbon monoxide concentration, degree of standardization of on-site personnel operation;

[0077] The determined index system and grading are as follows in the table:

[0078] Table 1 Evaluation Index System for On-site Environmental Safety of Underground Space Collapse Emergency Rescue

[0079]

[0080] After establishing the risk evaluation index system, use the AHP (Analytic Hierarchy Process) method to discretize the indexes; and combine with the form of expert questionnaire scoring to fuzzify the evaluation indexes, calculate the weights of each index through the method of combined weighting, adopt the OOB (Out-of-Bag) method, use the increment of standardized mean square error or the decline rate of classification accuracy as the measure, quantify the attenuation degree of model prediction performance caused by specific feature perturbations, and thus construct a feature importance score matrix, and obtain the weight matrix of features after normalizing the scores.

[0081] The said method of combined weighting includes subjective weighting method and objective weighting method. The specific steps of the subjective weighting method are as follows:

[0082] (1.1) According to the evaluation index system in Table 1, construct the factor set U of evaluation factors i ={u i1 ,u i2 ,...,u im};

[0083] (1.2) Single-factor evaluation, evaluate each factor separately, and have industry experts compare the importance among the secondary indexes in the evaluation indexes, and establish a weight judgment matrix R i :

[0084]

[0085] Among them, r imkIndicates the importance degree of the m-th secondary index compared to the k-th secondary index in the i-th primary index;

[0086] (1.3) Conduct a consistency test on the weight judgment matrix to determine whether CR satisfies "CR ≤ 0.1". The RI values are shown in Table 2:

[0087]

[0088] Among them, n represents the number of samples to be measured; CR is the consistency rate; CI is the consistency index; RI represents the random consistency index; λ max is the maximum eigenvalue; ω is the normalized weight vector; ω i is the weight value of the i-th primary evaluation index.

[0089] Table 2 RI value table

[0090]

[0091] (1.4) After the weight judgment matrix passes the consistency test, use the eigenvalue method to obtain the subjective weight value of the i-th primary index:

[0092]

[0093] Among them, ω in is the subjective weight of the n-th secondary evaluation index of evaluation index i, and ω zi is the subjective weight set of the i-th primary evaluation index.

[0094] According to the basic principle of the analytic hierarchy process method, by consulting the literature and consulting 5 experts and scholars, it is divided into 1-9 scale values to compare the importance degree of indicators, and the questionnaire results of each indicator comparison are obtained, and the weight judgment matrices of primary and secondary indicators are established.

[0095] The objective weight method uses the entropy weight method, and the specific calculation steps are as follows:

[0096] (2.1) Data standardization: Since the value ranges of each indicator are different, it is necessary to change them to the same range for comparison, and calculate the indicator values according to the maximum and minimum ideal values of the indicators:

[0097]

[0098] Among them, x ij represents the initial indicator data, y ij represents the standard indicator data, man(x ij ) represents the maximum indicator, and min(x ij ) represents the minimum indicator. When x ij is greater than min(x ijWhen [condition], use the first formula in step (2.1) to calculate y ij , when x ij is less than man(x ij ), use the second formula in step (2.1) to calculate y ij .

[0099] (2.2) Calculate the probability distribution of the j-th secondary indicator under the i-th primary indicator in each sample:

[0100]

[0101] Furthermore, obtain the information entropy E j :

[0102]

[0103] According to the information entropy E j obtain the objective weight ω ij :

[0104]

[0105] Furthermore, obtain the objective weight value of the i-th primary indicator:

[0106] ω ki =(ω i1 ,ω i2 ,...,ω ij )(10)

[0107] where ω ij is the objective weight of the j-th secondary evaluation indicator of the evaluation indicator i, and ω ki represents the subjective and objective weight set;

[0108] Finally, according to the weights obtained by the subjective weighting method and the objective weighting method, use the linear weighting method to calculate the mixed weight:

[0109] W = αW z +(1 - α)W k (0 ≤ α ≤ 1)(11)

[0110] where α represents the subjective weight coefficient.

[0111] Considering that there is little research on safety in this field, appropriately reduce the subjective weight. In this embodiment, α = 0.4 is taken.

[0112] For the evaluation values of each indicator, determine the scores and grades of the indicator values by calculating the Euclidean distance between each indicator and its ideal value.

[0113] S103. Input risk events, indicator values, and optimal values to evaluate the safety situation of the emergency rescue drill site.

[0114] The risk events include: rockfall events, collapse events, water inrush events, gas poisoning events, asphyxiation events, and methane explosion events. The safety conditions of the emergency rescue drill site are evaluated to provide a basis and reference for decision-making.

[0115] The safety levels of each risk event are as follows:

[0116] The environmental safety status levels of rockfall events are respectively level I, I, I, I, I, II, III, II, II;

[0117] The safety levels of collapse events are level I, I, I, I, II, II, II, III, III. During the later rescue, due to the increase in the disturbance of the large-diameter drill rig, the deformation of the surrounding rock accumulates over time, resulting in an increase in the collapse risk. The evaluation results are consistent with the actual monitoring results. It is recommended to pay attention to the reinforcement of the drill site during rescue;

[0118] The safety levels of water inrush events are respectively level I, I, I, II, I, I, II, II, I, which is relatively consistent with the situation that the water output of the drill site observed is small and there is no large amount of accumulated water;

[0119] The safety levels of hydrogen sulfide gas poisoning events are respectively level II, II, I, I, I, I, I, I, I, which reflects that the jet roadway ventilation was introduced into the rescue drill site on the afternoon of the first day, increasing the ventilation volume and reducing the hydrogen sulfide concentration, thus improving the safety level;

[0120] The safety levels of carbon monoxide poisoning events are respectively level II, II, III, III, III, III, III, III, III. A stable leakage source was found to continuously protrude in the drill site. Although the carbon monoxide concentration was reduced to a certain extent by increasing ventilation, the safety level was not improved. Attention should be paid to continuing to strengthen ventilation;

[0121] The safety levels of asphyxiation events are respectively level II, II, II, II, II, I, I, I, I. Half a day after the accident, the rescue personnel gathered and the number of people in the drill site increased, and ventilation devices were added. The evaluation results basically conform to the situation of the drill site;

[0122] The safety levels of methane explosion events are respectively level I, I, I, I, I, I, I, I, I, which is relatively safe.

[0123] According to the risk events: select the corresponding indicators, substitute the data of each indicator into the calculation, and then combine the weights to obtain the current safety conditions of each risk event in the drill site. Furthermore, corresponding treatment measures are proposed according to the situation of each risk indicator.

[0124] S104. According to the data change situation, use the ARIMAX multi-factor time series prediction model to measure the surrounding rock deformation and gas concentration data monitored, continuously predict the monitoring data through dynamic adjustment, predict the safety situation of the emergency rescue drill site, and give an early warning when it exceeds the safety setting requirements. The ARIMAX multi-factor time series prediction model is an autoregressive integrated moving average model with exogenous variables, which is used to predict the time series of monitoring data and then predict the monitoring data.

[0125] The calculation steps for predicting the surrounding rock deformation and gas concentration data monitored by using the ARIMAX multi-factor time series prediction model are as follows:

[0126] (1) Calculate the orders p and q of the autoregressive moving average model (ARMA):

[0127]

[0128] Among them, X1, X2,..., X t is the time series of sample data, μ is the constant term, p is the order, γ i is the autocorrelation coefficient, is the first error term, and the formula of the time series of sample data satisfies:

[0129] X t = a1X t-1 + a2X t-2 +......+ a p X t-p + u t (13)

[0130] Among them, a is its coefficient, u t is the second error term. When the noise in the sample data is not white noise, u t is regarded as a moving average of order q, and the formula of u t is:

[0131]

[0132] In the formula, δ t represents the white noise sequence; in the ARMA model

[0133] When the current value of the time series has nothing to do with the historical value but is related to the linear superposition of historical white noise, that is, X t = u t the formula of the time series of sample data is expressed as:

[0134]

[0135] The impact of white noise in the AR model on the predicted value is obtained in the MA, and the purpose of predicting future data based on past sequences is achieved through accumulation; the formula definition of the q-order MA model is as follows:

[0136]

[0137] where θ i is the coefficient of the white noise value δ i , and δ i-1 represents the white noise value of the (i - 1)-order MA model;

[0138] (2) Determine whether the time series of sample data is stationary. Use the autocorrelation function value to judge the stationarity. The calculation of the autocorrelation coefficient is as follows:

[0139]

[0140] where is the mean value of the time series of sample data; k is a certain moment within time t. If it decreases rapidly or even becomes zero when k increases, the time series of sample data is stationary. Otherwise, the time series of sample data is non-stationary and needs to be differenced;

[0141] (3) If the time series of sample data is non-stationary, then the time series needs to be differenced:

[0142] Δ d X t =(1 - B) d X t (18)

[0143] where B is the lag operator, BX t =X t-1 , d is the differencing order, and Δ d X t is the differenced sample value;

[0144] (4) After the sample processing is completed, the order of the prediction model needs to be determined: Perform a parameter grid search according to the AIC principle. The specific formula is as follows:

[0145] AIC = 2k - 2ln(L)(19)

[0146] where k represents the number of parameters to be estimated. The smaller the AIC value, the better the prediction model effect; Define the maximum factorial L. Calculate the AIC values from 1 to L respectively, and construct the optimal prediction model with the corresponding p and q when the AIC value is the smallest;

[0147] (5) Parameter estimation:

[0148]

[0149] where: φ(B) = 1 - φ1B - … - φ p B p is the autoregressive polynomial of a stationary invertible ARMA model, and φ p is the AR coefficient of the p-th term; θ(B) = 1 + θ1B + … + θ q B q is the moving average polynomial of a stationary invertible ARMA, and θ q is the MA coefficient of the q-th term; c is the constant term used to adjust the baseline level of the model; β i represents the regression coefficient corresponding to the exogenous variable; X it is the observed value of the i-th exogenous variable at time t; B p and B q represent the lag operator of the ARMA model, and B p *φ t = φ t-p .

[0150] (6) Data prediction. Using the calculation formula in step (5), the data at time t + 1 can be predicted:

[0151]

[0152] where φ1 represents the AR coefficient of the first term, β1 represents the regression coefficient of the first exogenous variable, θ1 represents the MA coefficient of the first term, represents the data at time t + 1, and X 1,t+1 represents the value of the first exogenous variable at time t + 1.

[0153] (7) Use the root mean square error (RMSE) to take n sample values to compare the error between the fitted value and the actual observed value:

[0154]

[0155] Substitute the predicted data into the above evaluation index system for evaluation, so as to achieve the purpose of early warning.

[0156] The advantages of the present invention are as follows: First, by adopting the method of combining the monitoring program with the main interface, eliminating irrelevant components and adding necessary connection channels, a software model capable of evaluating the safety situation of the emergency rescue drill site and the safety situation in the next stage is established, thus solving the problem that the safety situation of the emergency rescue drill site lacks theoretical support based on the experience of on-site personnel. Second, by using the combination of theoretical algorithms and monitoring data, the predicted monitoring data is dynamically adjusted to achieve real-time update of the prediction model and gradual improvement of the prediction accuracy. The indicators in the software can also be manually updated, gradually improving the prediction accuracy of the safety situation of the emergency rescue drill site, which is of great significance for preventing the occurrence of secondary accidents at the emergency rescue drill site and ensuring the safety of rescue personnel.

[0157] For the combined system, the specific implementation steps of the early warning method in this embodiment are as follows:

[0158] 1) Gas monitoring data is wirelessly transmitted to the host computer through the 433 module and automatically read into the software by arranging electrochemical gas sensors at the monitoring positions.

[0159] 2) For surrounding rock monitoring data, by setting the positions of the measuring points, setting the initial positions of the measuring points, and calculating the relative positions, the calculation module in the distribution box can calculate the deformation amount of the surrounding rock, and it is automatically read into the software through wifi transmission.

[0160] 3) In the main interface of the self-developed evaluation and early warning software, risk events can be selected for evaluation.

[0161] During the use of the system, some data required for establishing the index system are as follows:

[0162] The original terrain has large undulations, the burial depth is 701 - 723 m, and the maximum burial depth is 784 m; the surrounding rock lithology is mainly sandstone, with relatively hard lithology and relatively complete rock mass; both surface water and groundwater are relatively developed, mainly pore water, bedrock fissure water, and karst water. At 4:20 am on a certain day in 2023, a collapse accident occurred in some roadways of the coal mine, and many people were trapped, and toxic gases and hot water were emitted. According to the monitoring results of on-site gas sensors, the main components of the emitted gas are carbon dioxide, accompanied by methane, carbon monoxide, hydrogen sulfide, etc. Fill in the four pieces of information: index name, index category, classification category, and optimal value based on the above information.

[0163] The data import interface can select to manually import data by itself or automatically import monitoring data.

[0164] During the emergency rescue process, determine the situation of the survivors, open up the transportation channels for survival supplies, and open up the rescue channels. For 3 key nodes, namely 8 am on the first day (denoted as day1 - 8, the same below), 17 pm on the first day (day1 - 17), and 3 am on the third day (day3 - 3), and the situations one hour before and after are evaluated for safety. The data is shown in Table 3:

[0165] Table 3 Monitoring Data

[0166]

[0167] After the obtained monitoring data and environmental conditions are input into the software, the weight construction interface can be entered to add the importance comparison to generate weights by oneself, or use the built-in weights for calculation. In the case, the environmental safety status levels of the rockfall events at each moment are calculated using the built-in weights as I, I, I, I, I, II, III, II, II respectively; the safety level of the collapse events is I, I, I, I, II, II, II, III, III. During the later rescue, due to the increase in the disturbance of the large-diameter drill rig, the surrounding rock deformation accumulates with time, resulting in an increase in the collapse risk. The evaluation results are consistent with the actual monitoring results. It is recommended to pay attention to the reinforcement of the drill site during the rescue; the safety levels of the water inrush events are I, I, I, II, I, I, II, II, I respectively, which is relatively consistent with the situation that the water output of the drill site observed in the drill site is small and there is no large amount of accumulated water; the safety levels of the hydrogen sulfide gas poisoning events are II, II, I, I, I, I, I, I, I respectively, which reflects that the jet roadway ventilation was introduced into the rescue drill site on the afternoon of the first day, increasing the ventilation volume and reducing the hydrogen sulfide concentration, thus improving the safety level; the safety levels of the carbon monoxide poisoning events are II, II, III, III, III, III, III, III, III respectively. A stable leakage source is continuously protruding in the drill site. Although the carbon monoxide concentration has been reduced to a certain extent by increasing the ventilation, the safety level has not been improved. Attention should be paid to continuing to strengthen the ventilation; the safety levels of the asphyxiation events are II, II, II, II, II, I, I, I, I respectively. Half a day after the accident, the rescue personnel gathered and the number of people in the drill site increased, and the ventilation device was added. The evaluation results basically conform to the situation of the drill site; the safety levels of the methane explosion events are I, I, I, I, I, I, I, I, I respectively, which is relatively safe. To sum up, the risk of carbon monoxide poisoning is relatively high during the rescue process, and the risks of collapse and rockfall are relatively high in the later stage of the rescue. It is recommended to strengthen the ventilation and the later support to prevent the occurrence of secondary accidents.

[0168] 4) After the assessment is completed, the program can predict the monitoring data for a period of time in the future based on historical data, and dynamically calibrate and update according to the real-time refreshed monitoring data; the predicted monitoring data is brought into the assessment module again for calculation to obtain the predicted risk level, achieving the purpose of early warning. The data prediction results are shown in the figure, and the early warning is shown in the figure.

[0169] 5) The monitoring data is uploaded to the cloud at the same time. The core algorithm of the software is built into the server, which is convenient for remotely viewing the monitoring data, safety conditions and the next safety situation of the emergency rescue drill site.

[0170] A further improvement of the present invention lies in that in step 4), an evaluation and early warning algorithm self-developed and applicable to the emergency rescue drill site is integrated into the self-developed evaluation and early warning software, achieving the purpose of evaluating the safety of the emergency rescue drill site, obtaining the risk events that are more likely to occur, and corresponding disposal measures can be made according to the evaluation information to prevent the occurrence of secondary disasters.

[0171] Generally, the rescue party at the emergency rescue site pays little attention to the safety evaluation and prediction of the site environment, and often overly relies on the experience of construction workers, resulting in certain problems in the current rescue: on the one hand, limited by the professional level of construction workers, due to the incomplete consideration of risk factors, the reliability of the evaluation results is low; on the other hand, it is difficult to predict the probability of accidents in the next stage, resulting in the lack of rescue pre-set measures or the inability to improve the protection measures at the rescue site in a timely manner, putting rescue personnel in danger. Therefore, referring to relevant accident literature and reports, and combining expert opinions, a safety risk assessment index system for the emergency rescue drill site is established, and then an evaluation and prediction and early warning method is established.

[0172] A further improvement of the present invention lies in integrating gas monitoring data and surrounding rock monitoring data in the environment of the emergency rescue drill site, and comprehensively considering various index factors that may cause secondary disasters according to different risk events.

[0173] Generally, surrounding rock deformation monitoring is usually used in the construction excavation process of tunnels; gas monitoring is used in the excavation of resources such as coal and oil, where there is an environment with harmful gas volatilization; the present invention comprehensively considers the environmental characteristics of the emergency rescue drill site, combines gas monitoring and surrounding rock deformation monitoring, is applicable to the new environment, and has good effects.

[0174] The advantages of the present invention are that each module forms a closed loop, supporting real-time monitoring, risk assessment, intelligent early warning and dynamic decision-making. Through the functional division and collaborative work of each module, the system can comprehensively support the environmental safety assessment and early warning of the emergency rescue drill site, providing efficient technical support for on-site management and decision-making.

[0175] The present invention integrates gas monitoring, surrounding rock monitoring and a safety assessment and early warning system, providing a comprehensive solution. Through intelligent data processing and algorithm models, the platform can not only monitor various safety data in real time, but also provide accurate early warnings and emergency response measures to ensure the safety during the engineering operation process. With the progress of technology and the continuous accumulation of data, the platform will be continuously optimized to gradually achieve more efficient and intelligent safety management.

[0176] Example 3

[0177] A computer device includes a memory, a processor, and a computer program stored on the memory, and the processor executes the computer program to implement the steps of the above method.

[0178] Example 4

[0179] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above method.

[0180] Example 5

[0181] A computer program product includes a computer program or instructions, which, when executed by a processor, implement the steps of the above method.

[0182] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. An environmental safety assessment and early warning system applied to an emergency rescue drilling site, characterized in that, Including: A gas monitoring module, a surrounding rock deformation monitoring module, and an evaluation and early warning software module; The gas monitoring module includes an electrochemical gas sensor, a gas concentration data acquisition module, a 433 wireless transmission module, and a GPRS wireless transmission module; the gas concentration data acquisition module is used to receive data and supply power to the electrochemical gas sensor, the 433 wireless transmission module, and the GPRS wireless transmission module; the 433 wireless transmission module includes receiving and transmitting antennas and is used to wirelessly transmit data to a local host computer; the GPRS wireless transmission module includes a transmitting antenna and a built-in wireless network card and is used to wirelessly transmit data to the cloud; The surrounding rock deformation monitoring module includes a target lamp, a power supply module, a monitoring camera, and a distribution box. The target lamp is used to emit laser light; the power supply module is used to supply power to the target lamp or charge the battery; the monitoring camera is used to receive the laser light of the target lamp; the distribution box includes a data acquisition module and a wireless transmission module and supplies power to the monitoring camera; The evaluation and early warning software module is used to receive the data monitored by the gas monitoring module and the surrounding rock deformation monitoring module and realize functions such as data reception and recording, environmental data input, index level determination, and overall safety assessment and prediction of the emergency rescue drill site environment.

2. The environmental safety assessment and early warning system for an emergency rescue drill site according to claim 1, characterized in that The electrochemical gas sensor includes four gas probes: a methane gas electrochemical gas sensor probe, a carbon monoxide electrochemical gas sensor probe, a hydrogen sulfide electrochemical gas sensor probe, and an oxygen electrochemical gas sensor probe, which are respectively used to monitor methane gas, carbon monoxide, hydrogen sulfide, and oxygen.

3. An environmental safety assessment and early warning method applied to an emergency rescue drill site, which is implemented based on the environmental safety assessment and early warning system described in any one of claims 1-2, characterized in that, The method includes the following steps: S101, collecting and integrating gas monitoring and surrounding rock deformation monitoring data; S102, establishing an evaluation index system for the emergency rescue drill site; The index system includes two primary indexes: pregnancy risk environment and risk-causing factors. The primary indexes are further subdivided into the following secondary indexes and index systems: Secondary indexes of the pregnancy risk environment: surrounding rock level, rock mass fragmentation degree, rock mass hardness, underground water volume, original surrounding rock support measures, drill site reinforcement, fault fracture zone; Secondary indexes of risk-causing factors: ventilation volume, drill site vibration speed, surrounding rock deformation amount, surrounding rock deformation rate, oxygen concentration, hydrogen sulfide concentration, methane concentration, carbon monoxide concentration, and on-site personnel operation specification degree; After establishing the risk evaluation index system, use the AHP hierarchical analysis method to discretize the indexes; and combine the questionnaire form of expert questionnaire scoring to fuzzify the evaluation indexes, and calculate the weights of each index through the method of combined weighting; S103, evaluating the safety situation of the emergency rescue drill site according to risk events, index values, and optimal values; S104, according to the data change situation, use the ARIMAX multi-factor time series prediction model to predict the monitored surrounding rock deformation and gas concentration data. According to the continuously predicted monitored data, combined with the evaluation indexes, predict the safety situation of the emergency rescue drill site and give an early warning when it exceeds the safety setting requirements.

4. The environmental safety assessment and early warning method for an emergency rescue drill site according to claim 3, wherein The method of combined weighting includes a subjective weighting method and an objective weighting method. The specific steps of the subjective weighting method are as follows: (1.1) According to the established evaluation index system, construct the factor set U of evaluation factors i ={u i1 , u i2 ,..., u im}, where u im represents each evaluation factor; i is the number of primary indexes; m is the number of secondary evaluation indexes in the primary index i; (1.2) Evaluate each factor individually. Let industry experts compare the importance among secondary indicators in the evaluation index, and establish a weight judgment matrix R according to the comparison results. i ; where r imk represents the importance degree of the m-th secondary index compared to the k-th secondary index in the i-th primary index; (1.3) Conduct a consistency test on the weight judgment matrix to determine whether CR meets the requirement of "CR ≤ 0.1". Among them, n represents the number of samples to be measured; CR is the harmonization rate; CI is the consistency index; RI represents the random consistency index; λ max is the maximum eigenvalue; ω is the normalized weight vector; ω i is the weight value of the i-th primary evaluation index; (1.4) After the weight judgment matrix passes the consistency test, use the eigenvalue method to obtain the subjective weight value of the i-th first-level index: where, ω in is the subjective weight of the n-th secondary evaluation index of evaluation index i, and ω zi is the subjective weight set of the i-th primary evaluation index; The objective weight method adopts the entropy weight method, and the specific calculation steps are as follows: (2.1) Data standardization: Since the value ranges of each index are different, it is necessary to change them to the same range for comparison. Calculate the index values according to the maximum and minimum ideal values of the index: where x ij represents the initial index data, and y ij represents the standard index data. man(x ij ) represents the maximum index, and min(x ij ) represents the minimum index; when x ij is greater than min(x ij ), the first formula in step (2.1) is used to calculate y ij , and when x ij is less than man(x ij ), the second formula in step (2.1) is used to calculate y ij ; (2.2) Calculate the probability distribution of the j-th second-level index under the i-th first-level index in each sample: Furthermore, the information entropy E is obtained. j : According to the information entropy E j the objective weight ω is obtained ij : Furthermore, obtain the objective weight value of the i-th first-level index: ω ki = (ω i1 , ω i2 ,..., ω ij )(10) Among them, ω ij is the objective weight of the j-th secondary evaluation index of the evaluation index i, and ω ki represents the set of objective weights of the i-th primary evaluation index; Finally, calculate the mixed weight using the linear weighting method based on the weights obtained by the subjective weighting method and the objective weighting method: ω i = αω ki + (1 - α)ω zi (0 ≤ α ≤ 1)(11) Among them, α represents the subjective weight coefficient; For the evaluation values of each index, determine the scores and grades of each index value by calculating the Euclidean distance between each index and its ideal value.

5. The environmental safety assessment and early warning method for an emergency rescue drill site according to claim 3, wherein In S103, according to the surrounding rock deformation monitoring data and gas concentration monitoring data, the risk events include: rockfall events, collapse events, water inrush events, gas poisoning events, asphyxiation events, methane explosion events, and evaluate the safety conditions of the emergency rescue drill site to provide a basis and reference for decision-making; The safety levels of each risk event are: The environmental safety status levels of rockfall events are I, I, I, I, I, II, III, II, II levels respectively; The safety levels of collapse events are I, I, I, I, II, II, II, III, III levels. During the later rescue, due to the increase in the disturbance of the large-diameter drill rig, the deformation of the surrounding rock accumulates over time, resulting in an increase in the collapse risk. The evaluation results are consistent with the actual monitoring results. It is recommended to pay attention to the reinforcement of the drill site during rescue; The safety levels of water inrush events are I, I, I, II, I, I, II, II, I levels respectively, which is relatively consistent with the situation that the water inflow of the drill site observed in the drill site is small and there is no large amount of accumulated water; The safety levels of hydrogen sulfide gas poisoning events are II, II, I, I, I, I, I, I, I levels respectively, which reflects that the jet roadway ventilation was introduced into the rescue drill site on the afternoon of the first day, increasing the ventilation volume and reducing the hydrogen sulfide concentration, resulting in an improvement in the safety level; The safety levels of carbon monoxide poisoning events are II, II, III, III, III, III, III, III, III levels respectively. A stable leakage source was found to be continuously protruding in the drill site. Although the concentration of carbon monoxide was reduced to a certain extent by increasing ventilation, the safety level was not improved. Attention should be paid to continuing to strengthen ventilation; The safety levels of asphyxiation events are II, II, II, II, II, I, I, I, I levels respectively. Half a day after the accident, rescue personnel gathered, the number of people in the drill site increased, and ventilation devices were added. The evaluation results basically conform to the situation of the drill site; The safety levels of methane explosion events are I, I, I, I, I, I, I, I, I levels respectively, which are relatively safe.

6. The environmental safety assessment and early warning method for an emergency rescue drill site according to claim 3, characterized in that, In S103, corresponding indicators are selected according to risk events. After combining with the calculated weights, the safety conditions of current risk events in the drilling site can be obtained, and then corresponding treatment measures can be proposed based on the conditions of each risk indicator.

7. The environmental safety assessment and early warning method for an emergency rescue drilling site according to claim 3, characterized in that, In S104, the calculation steps for predicting the monitored surrounding rock deformation and gas concentration data using the ARIMAX multi-factor time series prediction model are as follows: (1) Calculate the orders p and q of the ARMA autoregressive moving average model: where X1, X2, ..., X t is the time series of sample data, μ is the constant term, p is the order, and γ i is the autocorrelation coefficient, is the first error term, and the formula for the time series of sample data satisfies: X t = a1X t-1 + a2X t-2 +......+ a p X t-p + u t (13) where a is its coefficient, u t is the second error term. When the noise in the sample data is not white noise, u t is regarded as a moving average of order q, and u t The formula is: where δ t represents a white noise sequence; in the ARMA model When the current value of the time series is independent of the historical values and related to the linear superposition of historical white noise, i.e., X t = u t , the formula representation of the time series of the sample data is: The influence of white noise in the AR model on the predicted value is obtained in the MA, and the purpose of predicting future data based on past sequences is achieved through accumulation; the formula definition of the q-order MA model is: where θ i is the coefficient of the white noise value δ i , and δ i-1 represents the white noise value of the (i - 1)-th order MA model; (2) Determine whether the time series of sample data is stationary, and use the autocorrelation function value for stationarity judgment. The calculation of the autocorrelation coefficient is as follows: where is the mean of the sample data time series; k is a certain moment within time t. If it decreases rapidly or even becomes zero when k increases, the sample data time series is stationary; otherwise, the sample data time series is non-stationary and needs to be differenced. (3) If the time series of sample data is not stationary, the time series needs to be differenced: Δ d X t =(1 - B) d X t (18) where B is the lag operator, BX t = X t-1 , d is the order of differencing, Δ d X t is the sample value after differencing; (4) After the sample processing is completed, the order of the prediction model needs to be determined: parameter grid search is performed according to the AIC principle, and the specific formula is as follows: AIC = 2k - 2ln(L)(19) Among them, k represents the number of parameters to be estimated. The smaller the AIC value, the better the prediction model effect; define the maximum factorial L, calculate the AIC values from 1 to L respectively, and construct the optimal prediction model with the corresponding p and q when the AIC value is the smallest; (5) Parameter estimation: where, φ(B) = 1 - φ1B - … - φ p B p is the autoregressive polynomial of a stationary invertible ARMA model, and φ p is the AR coefficient of the p-th term; θ(B) = 1 + θ1B + … + θ q B q is the moving average polynomial of a stationary invertible ARMA model, and θ q is the MA coefficient of the q-th term; c is a constant term used to adjust the baseline level of the model; β i represents the regression coefficient of the i-th exogenous variable; X it represents the observed value of the i-th exogenous variable at time t; B p and B q represent the lag operator of the ARMA model, and B p *φ t = φ t-p ; ε t represents the white noise error term at time t; (6) Data prediction, the data at time t + 1 can be predicted using the calculation formula in step (5): Among them, φ1 represents the AR coefficient of the first term, β1 represents the regression coefficient of the first exogenous variable, and θ1 represents the MA coefficient of the first term. represents the data at time t + 1, X 1,t+1 represents the value of the first exogenous variable at time t + 1; (7) Use the root mean square error RMSE to take n sample values to compare the error between the fitted value and the actual observed value: The predicted data is brought into the evaluation index system for evaluation, so as to achieve the purpose of early warning.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes a computer program to implement the steps of the environmental safety assessment and early warning method described in any one of claims 3-7.

9. A computer-readable storage medium, characterized in that, There is a computer program stored, and when the program is executed by the processor, the steps of the environmental safety assessment and early warning method described in any one of claims 3-7 are implemented.

10. A computer program product, characterized in that, It includes a computer program or instruction, and when the program or instruction is executed by the processor, the steps of the environmental safety assessment and early warning method described in any one of claims 3-7 are implemented.

Images