Industrial park risk source grade evaluation method and device and electronic equipment

Abstract

The invention provides an industrial park risk source grade assessment method and device and electronic equipment, and the method comprises the steps: obtaining to-be-assessed risk source data of an industrial park, and classifying the to-be-assessed risk source data into an invariant, a continuous variable and a break variable; constructing and training a neural network model which takes the risk source data as input and takes the alarm result of each type of risk source as output; comprising an input layer used for inputting risk source data; the feature extraction unit comprises a first feature extraction unit used for extracting invariant features according to the invariants; the second feature extraction unit is used for extracting continuous features according to the continuous variables; the third feature extraction unit is used for extracting mutation features according to the mutation quantity; the output layer is used for obtaining alarm results of various types of risk sources according to the invariant features, the continuity features and the mutation features; and S3, acquiring to-be-evaluated current risk source data of the industrial park, inputting the trained neural network model, and evaluating an alarm result of each risk source. According to the method, potential risks can be effectively identified, evaluated and controlled, and the safety of production operation is guaranteed, so that an effective risk management strategy is made, and sustainable development is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and specifically relates to a method, device and electronic equipment for assessing the level of risk sources in an industrial park. Background Art

[0002] With the acceleration of industrialization, industrial parks, as an important carrier of economic development, are receiving increasing attention for their safety issues. Especially for chemical industrial parks, since they involve a large number of hazardous chemicals and high-risk operations, once an accident occurs, it will have a serious impact on personnel safety, environmental protection and even social stability. Therefore, establishing a scientific and systematic risk source level assessment method is crucial to prevent and reduce the occurrence of accidents and ensure the sustainable development of the park.

[0003] In the process of risk assessment of industrial parks, industrial parks often contain a variety of processes, equipment and chemicals. The interaction of these complex factors makes risk assessment very complicated. Risk assessment relies on previous data and assumptions, but the uncertainty in the actual environment may lead to deviations in the predicted results. For some highly specialized processes or new materials, there is a lack of mature risk assessment methods and standards, which increases the technical difficulty of the assessment. Human judgment and decision-making in the risk assessment process may be subjective and biased, affecting the objectivity and accuracy of the assessment. Risks are constantly changing and require continuous monitoring and updating, but in actual operations, risk assessments may be ignored or delayed for various reasons. Summary of the invention

[0004] In order to solve the problem that the existing risk assessment is difficult and the assessment accuracy is low, the present invention provides an industrial park risk source level assessment method, comprising:

[0005] S1: Obtain the risk source data to be assessed in the industrial park and classify them into invariable variables, continuous variables and mutation variables;

[0006] S2: Construct and train a neural network model that takes risk source data as input and outputs the warning results of various types of risk sources; including:

[0007] Input layer, used to input risk source data;

[0008] Feature extraction unit, including:

[0009] A first feature extraction unit is used to extract an invariant feature based on an invariant variable;

[0010] A second feature extraction unit is used to extract continuous features based on continuous variables;

[0011] A third feature extraction unit is used to extract mutation features according to the mutation amount;

[0012] The output layer is used to obtain the alarm results of various types of risk sources based on the invariant characteristics, continuity characteristics and mutation characteristics. S3: Obtain the current risk source data to be evaluated in the industrial park, input the trained neural network model, and evaluate the alarm results of each risk source.

[0013] Furthermore, the feature extraction unit includes: a first convolution layer, a first pooling layer, a hidden layer, a second convolution layer, a random dropout layer, a third convolution layer and a second pooling layer connected in sequence;

[0014] The first convolutional layer and the first pooling layer, as the first feature extraction unit, effectively extract invariant features;

[0015] The hidden layer, the second convolutional layer and the random dropout layer, as the second feature extraction unit, effectively extract the continuity features;

[0016] The third convolutional layer and the second pooling layer, as the third feature extraction unit, effectively extract mutation features.

[0017] Furthermore, the convolution kernel of the first convolution layer is 3, the stride is 3, the padding is 2, and the activation function is rule; the convolution kernel of the second convolution layer is 3, the stride is 7, the padding is 3, and the activation function is rule; the convolution kernel of the third convolution layer is 5, the stride is 3, the padding is 2, and the activation function is rule.

[0018] Furthermore, step S3 further includes: obtaining historical risk source data to be assessed in the industrial park;

[0019] The overall evaluation model includes: an overall input layer, two neural network models, a fully connected layer and an overall output layer;

[0020] The total input layer is used to obtain the current risk source data and historical risk source data to be assessed in the industrial park;

[0021] The first neural network model is connected to the total input layer and is used to output current features according to current risk source data;

[0022] The second neural network model is connected to the total input layer and is used to output historical features based on historical risk source data;

[0023] The fully connected layer is connected to the output of the first group of neural network models and the second group of neural network models, and is used to determine the overall alarm result based on current features and historical features, and output it through the overall output layer.

[0024] Furthermore, the total alarm result is determined according to the current features and the historical features, specifically: the current features and the historical features are weighted and added to obtain the total alarm result.

[0025] Furthermore, the training of the neural network model includes:

[0026] Obtain previous data, including key factors corresponding to each risk source, the likelihood and severity of each risk source;

[0027] Based on the key factors, the likelihood and severity of each risk source, each risk source is graded using expert experience, the alarm results of each risk source are obtained, and a training data set is constructed based on previous data and alarm results;

[0028] The neural network model is trained according to the training data set to obtain a trained neural network model.

[0029] Furthermore, based on the risk source data, variables with only two possible answers, yes or no, are classified as unpredictable variables; variables with consistent and predictable change directions are classified as continuous variables; and variables with inconsistent and uncontrollable change directions are classified as mutation variables.

[0030] Furthermore, the risk sources include: mechanical risk sources, electrical risk sources and operational risk sources;

[0031] Mechanical risk sources include: risk sources related to transmission devices, risk sources related to drive devices, risk sources related to bearings, and risk sources related to fixing devices;

[0032] Electrical risk sources include: risk sources related to power supply, risk sources related to control system, risk sources related to electrical equipment and risk sources related to ground wire;

[0033] Operational risk sources include: risk sources related to operational control, risk sources related to safety warnings, and risk sources related to cleaning and maintenance.

[0034] On the other hand, the present invention also provides an industrial park risk source level assessment device, comprising a data acquisition unit, a model processing unit and a risk assessment unit connected in sequence; and used to respectively execute any of the above-mentioned industrial park risk source level assessment methods.

[0035] On the other hand, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned industrial park risk source level assessment methods is implemented.

[0036] The beneficial effects of the present invention are: as can be seen from the above, the industrial park risk source level assessment method, device and electronic equipment provided by the embodiment of the present invention can effectively identify, assess and control potential risks, ensure the safety of production operations, so as to formulate effective risk management strategies to achieve sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 Schematic diagram of the process of the industrial park risk source level assessment method in an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of the structure of a neural network model in an embodiment of the present invention;

[0040] Figure 3 Schematic diagram of a long short-term memory network in an embodiment of the present invention;

[0041] Figure 4 Schematic diagram of the structure of the industrial park risk source level assessment device in an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of an electronic device in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0043] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0044] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0045] The embodiment of the present invention provides a method for assessing the risk level of an industrial park. The method for assessing the risk level of an industrial park according to the embodiment of the present invention is provided. Figure 1 As shown in the figure, the risk source level assessment method for industrial parks includes:

[0046] Step S1: Obtain risk source data to be assessed in the industrial park and classify them into invariable variables, continuous variables and mutation variables;

[0047] In the embodiment of the present invention, all possible risk sources in the park are comprehensively sorted out, including but not limited to the use, storage, transportation, waste disposal and other links of chemicals; flow charts, technical specifications, record books, etc. are optionally used to record in detail the operation process and potential risk factors of each link.

[0048] More specifically, risk sources may include mechanical risk sources, electrical risk sources, and operational risk sources. Mechanical risk sources include, but are not limited to, risk sources related to transmission devices, risk sources related to drive devices, risk sources related to bearings, and risk sources related to fixtures. Electrical risk sources include, but are not limited to, risk sources related to power supplies, risk sources related to control systems, risk sources related to electrical equipment, and risk sources related to ground wires. Operational risk sources include, but are not limited to, risk sources related to operational control, risk sources related to safety warnings, and risk sources related to cleaning and maintenance.

[0049] Take the coal mine belt transportation assessment as an example. In order to ensure the safety and stability of the coal mine belt transportation process, it is necessary to identify and control the risk sources. The following is the identification of risk sources in the coal mine belt transportation profession:

[0050] Mechanical risk sources include:

[0051] 1. Risk sources involved in the belt itself: wear, stretching, aging, rupture, blistering, insufficient or excessive tension, breakage, etc.

[0052] 2. Risk sources involved in the drive device: drive motor failure, electronic control failure, heat increase caused by dust accumulation in the drive transmission chain, etc.

[0053] 3. Risk sources involved in bearings: poor lubrication, wear, overload, vibration, looseness, etc.

[0054] 4. Risk sources involved in the roller: thermal expansion and contraction, peeling, diameter change, rust, foreign objects such as stones and wood stuck in the roller, etc.

[0055] 5. Risk sources involved in belt brackets: insufficient firmness, deformation, rust, cracking, etc.; Risk sources involved: belt overlap, excessive tension, deviation, etc.

[0056] Sources of electrical risks include:

[0057] 1. Risk sources involved in the power supply: power short circuit, poor cable quality, overload, etc.

[0058] 2. Risk sources involved in the control system: control line short circuit, poor wiring, failure of electrical protection relays, etc.

[0059] 3. Risk sources involved in electrical equipment: motor failure, loose terminal blocks, excessive temperature, etc.

[0060] 4. Risk sources involved in the ground wire: ground wire breakage, poor grounding, etc.

[0061] Sources of operational risk include:

[0062] 1. Risk sources involved in operation control: unbalanced load, running too fast or too slow.

[0063] 2. Risk sources involved in the discharge pedal: residues, dropped objects, unbalanced discharge flow, etc.

[0064] 3. Risk sources involved in safety warnings: warning signals, emergency stops, start buttons, etc.

[0065] 4. Risk sources involved in cleaning and maintenance: incorrect shutdown procedures, improper personnel during cleaning and maintenance, and incorrect operations in maintenance and simulation.

[0066] The evaluation is based on 12 key factors from the production nodes in coal mine belt transportation.

[0067] After collecting the above risk source data, it can be optionally converted into digital data for recording and storage. For example, serious variables such as short circuit, warning, emergency stop, etc., which have only two conditions of yes or no, are regarded as non-variable variables, that is, if the variable changes, it is directly identified as a high risk level, and if it does not change, it is identified as a low risk; it is converted to 0 / 1, 0 means no occurrence, 1 means occurrence, and irreversible quantities such as belt wear, aging, cracking, etc. that continue to grow over time are regarded as continuous variables. This type of continuous variable not only has a consistent change direction, but also has controllable and predictable changes. For example, wear and aging, etc., can roughly predict the risks and high risks. ; Specifically expressed by its specific wear and aging amount, etc., when it is within a certain range, it is tolerable and identified as low risk, exceeding a certain range is identified as medium risk, and exceeding a higher range is identified as high risk; the unstable quantities such as temperature and load that fluctuate up and down are regarded as mutation quantities, and the change direction of the mutation variable is inconsistent. The key is that its change amount is uncontrollable, and it is impossible to predict the time when it may occur and the specific change amount. When the change amplitude is within a certain range, that is, the change amount is less than the set threshold, it is tolerable and identified as low risk, exceeding a certain range is identified as medium risk, and exceeding a higher range is identified as high risk.

[0068] Step S2: Construct and train a neural network model that takes risk source data as input and the warning results of various types of risk sources as output; including:

[0069] Input layer, used to input risk source data;

[0070] Feature extraction unit, including:

[0071] A first feature extraction unit is used to extract an invariant feature based on an invariant variable;

[0072] A second feature extraction unit is used to extract continuous features based on continuous variables;

[0073] A third feature extraction unit is used to extract mutation features according to the mutation amount;

[0074] The output layer is used to obtain the warning results of various types of risk sources based on invariant features, continuity features and mutation features.

[0075] In this embodiment, according to the type of risk source data, invariant features, continuity features and mutation features are extracted in sequence, so that each risk source data can be processed according to its type, and the unique features of each type of data can be extracted more effectively and accurately; more importantly, according to the different features, invariant features, continuity features and mutation features are extracted in sequence, so that the importance of each feature can be brought into play in order, thereby further improving the accuracy of subsequent evaluations.

[0076] More preferably, the feature extraction unit includes, in sequence: a first convolutional layer, a first pooling layer, a hidden layer, a second convolutional layer, a random inactivation layer, a third convolutional layer, and a second pooling layer; the first convolutional layer and the first pooling layer, as the first feature extraction unit, effectively extract invariant features; the hidden layer, the second convolutional layer, and the random inactivation layer, as the second feature extraction unit, are good at extracting continuous values ​​and effectively extracting continuity features; the third convolutional layer and the second pooling layer, as the third feature extraction unit, are placed at the end of the model to effectively extract mutation features. Specifically, the structure of the feature extraction unit, the present invention provides a preferred embodiment, which takes advantage of each structure, first extracts invariant features, then extracts continuous variable features, and finally extracts mutation features, and finally subsequently fuses and outputs the alarm results of various types of risk sources, thereby determining the overall risk level.

[0077] More preferably, the convolution kernel of the first convolution layer is 3, the step size is 3, the padding is 2, and the activation function is rule. The convolution kernel of the second convolution layer is 3, the step size is 7, the padding is 3, and the activation function is rule. The convolution kernel of the third convolution layer is 5, the step size is 3, the padding is 2, and the activation function is rule.

[0078] More preferably, the hidden layer adopts Figure 3 The Long Short-Term Memory (LSTM) network model shown in Figure 1 is used to further extract continuous features, where the forget gate f tActing on the previous temporal hidden state C t-1 . Determines the effect of the previous hidden unit state on the current hidden unit: f t *C t-1 .

[0079] Input Gate i t The input to the current hidden unit (The input is the output h of the previous time series hidden unit t-1 and the input of the current hidden unit). Determines the effect of the current input on the current hidden unit:

[0080]

[0081] The forget gate and the input gate determine the current hidden layer state output C t , output gate o t Acting on the current output, produces the current hidden unit output:

[0082]

[0083] h t =tanh(o t *C t )

[0084] Among them, f t 、i t , o t They are all linear combinations of the inputs of the current hidden units, and then normalized by the sigmoid function:

[0085] f t =σ(W f X t +U f h t-1 +b f )

[0086] i t =σ(W i X t +U i h t-1 +b i )

[0087] o t =σ(W o X t +U o h t-1 +b o )

[0088] More preferably, after the neural network model is constructed, before step S3, the previous data of the risk source is obtained, the previous data includes the key factors corresponding to each risk source, the possibility and severity of each risk source, and the three types of risk source data, invariant, continuous variable and mutation, at least one key factor. According to the key factors, the possibility and severity of each risk source, each risk source is graded using expert experience, the alarm result of each risk source is obtained, and a training data set is constructed based on the previous data and the alarm result; the neural network model is trained based on the training data set to obtain a trained neural network model.

[0089] S3: Obtain the current risk source data to be evaluated in the industrial park, input it into the trained neural network model, and evaluate the alarm results of each risk source.

[0090] Specifically, the current risk source data to be evaluated in the industrial park is obtained, and the neural network model trained in step S2 is input to evaluate the alarm results of each risk source. Specifically, the trained neural network model is used to conduct an in-depth analysis of each key factor to evaluate the alarm results of various risk sources.

[0091] More preferably, step S3 further includes: obtaining historical risk source data to be assessed in the industrial park;

[0092] like Figure 2 As shown, the overall evaluation model includes: an overall input layer, two of the above-mentioned neural network models, a fully connected layer and an overall output layer;

[0093] The total input layer is used to obtain the current risk source data and historical risk source data to be assessed in the industrial park;

[0094] The first neural network model is connected to the total input layer and is used to output current features according to current risk source data;

[0095] The second neural network model is connected to the total input layer and is used to output historical features based on historical risk source data;

[0096] The fully connected layer is connected to the output of the first and second neural network models to add the weights of current features and historical features, and outputs the total alarm results of each risk source through the total output layer.

[0097] For example, the key factors of the current day are input into the first neural network model for processing, and the key factors of several days before the current day are input into the second neural network model for processing; the first neural network model processes and outputs the feature vector of the current day, and the second neural network model processes and outputs the feature vector of several days before the current day; the feature vector of the current day is weighted added with the feature vectors of several days before the current day, and the alarm results of each risk source are output after processing by the fully connected layer.

[0098] For example, after the first neural network model and the second neural network model output the corresponding feature vector, the feature vector of w(1*n) is added, w1*0.65+w2*0.35. According to the possibility of risk occurrence and the severity of the consequences, the risks are graded, including low risk, medium risk, and high risk. That is, the neural network model outputs the alarm results, including low risk, medium risk, and high risk.

[0099] Preferably, the industrial park risk source level assessment method of the present invention further includes: Step S4: comprehensively assessing the overall risk level based on the alarm results of each risk source combined with the interrelationships and combined effects between the risk sources.

[0100] In the embodiment of the present invention, the risks are quantitatively or qualitatively described using previous data and industry standards, and the overall risk level is comprehensively evaluated by considering the interrelationships and combined effects between risks.

[0101] In the embodiment of the present invention, the risk source level assessment of the industrial park also includes: formulating corresponding prevention and control measures for different risk levels; such as improving processes, strengthening monitoring, and setting up protective facilities.

[0102] Develop specific operating procedures and emergency plans based on the risk level of each risk source to ensure a rapid response when risks occur.

[0103] In an embodiment of the present invention, the industrial park risk source level assessment method also includes: establishing a monitoring system to regularly check the status of risk sources and the implementation of safety measures; regularly reviewing the results of risk assessments, and timely updating the risk database to ensure the timeliness and accuracy of the assessment.

[0104] Timely report the results of risk assessment to management, employees and relevant stakeholders to ensure information transparency. Provide safety risk education and training to employees to improve their safety awareness and emergency response capabilities.

[0105] The embodiment of the present invention classifies and extracts features from risk sources according to their characteristics, and then identifies and evaluates them, thereby discovering potential safety hazards in advance and taking preventive measures to prevent accidents. Through quantitative risk management, decision makers can make more scientific safety management decisions based on data and analysis results. Clear risk levels help to formulate targeted emergency plans, improve the ability to respond to emergencies, and help strengthen cooperation with external professional organizations to ensure the effectiveness and practicality of risk assessment work.

[0106] The embodiments of the present invention introduce new safety management technologies and best practices to continuously improve the level of risk management. Through this series of steps, the industrial park can effectively identify, evaluate and control potential risks, ensure the safety of production operations, and comply with national and local laws and regulations to promote the sustainable development of the park. Safety risk assessment can timely detect potential dangerous factors, such as belt deviation, slippage, overload, etc. By taking corresponding safety measures, accidents can be effectively prevented and the life safety and health of employees can be guaranteed. In addition, good safety management can also enhance the corporate image and enhance employees' trust and satisfaction with the company. Coal mine belt transportation risk assessment is of great significance to ensuring safe production, improving efficiency and protecting the environment. Enterprises should fully recognize its importance and formulate effective risk management strategies to achieve sustainable development. It can effectively identify, evaluate and control potential risks and ensure the safety of production operations, so as to formulate effective risk management strategies to achieve sustainable development.

[0107] The above specific embodiments of the present invention are described. In some cases, the actions or steps recorded in the embodiments of the present invention can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] Based on the same concept, an embodiment of the present invention also provides an industrial park risk source level assessment device. Applicable to a processor. Figure 4 As shown, the industrial park risk source level assessment device includes: a data acquisition unit, a model processing unit and a risk assessment unit, which respectively execute the above steps S1-S3.

[0109] For the convenience of description, the above device is described as various modules according to their functions. Of course, when implementing the embodiment of the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0110] The device of the above embodiment is applied to the corresponding method in the above embodiment, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0111] Based on the same inventive concept, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method of any of the above embodiments when executing the program.

[0112] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the method in any of the above embodiments.

[0113] Figure 5 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 501, a memory 502, an input / output interface 503, a communication interface 504, and a bus 505. The processor 501, the memory 502, the input / output interface 503, and the communication interface 504 are connected to each other through the bus 505 in the device.

[0114] The processor 501 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided by the method embodiment of the present invention.

[0115] The memory 502 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 502 can store an operating system and other application programs. When the technical solution provided by the method embodiment of the present invention is implemented by software or firmware, the relevant program code is stored in the memory 502 and called and executed by the processor 501.

[0116] The input / output interface 503 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0117] The communication interface 504 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0118] The bus 505 includes a path for transmitting information between various components of the device (eg, the processor 501 , the memory 502 , the input / output interface 503 , and the communication interface 504 ).

[0119] It should be noted that, although the above device only shows the processor 501, the memory 502, the input / output interface 503, the communication interface 504 and the bus 505, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present invention, and does not necessarily include all the components shown in the figure.

[0120] It should be appreciated that the method steps in the embodiments of the present invention can be implemented or implemented by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer readable memory. The method can use standard programming techniques. Each program can be implemented in a high-level process or object-oriented programming language to communicate with a computer system. However, if necessary, the program can be implemented in an assembly or machine language. In any case, the language can be a compiled or interpreted language. In addition, the program can be run on a programmed ASIC for this purpose.

[0121] Furthermore, the operations of the processes described herein may be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes described herein (or variations and / or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) that is executed collectively on one or more processors, by hardware, or a combination thereof. A computer program includes a plurality of instructions that may be executed by one or more processors.

[0122] Further, the method can be implemented in and operably connected to any type of suitable computing platform, including but not limited to a personal computer, a minicomputer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, etc. Various aspects of the present invention can be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, an optical read and / or write storage medium, a RAM, a ROM, etc., so that it can be read by a programmable computer, and when the storage medium or device is read by the computer, it can be used to configure and operate the computer to perform the process described herein. In addition, the machine-readable code, or a portion thereof, can be transmitted via a wired or wireless network. When such media includes instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor, the invention herein includes these and other different types of non-transitory computer-readable storage media. When programmed according to the methods and techniques of the present invention, the present invention can also include the computer itself.

[0123] The computer program can be applied to input data to perform the functions herein, thereby converting the input data to generate output data stored in a non-volatile memory. The output information can also be applied to one or more output devices such as a display. In a preferred embodiment of the present invention, the converted data represents physical and tangible objects, including specific visual depictions of physical and tangible objects produced on the display.

[0124] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of the present application as above, which are not provided in detail for the sake of simplicity.

[0125] This application is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the embodiments of the present invention. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the scope of protection of this application.

Claims

1. A method for assessing the risk level of an industrial park, characterized in that: Methods include: S1: Obtain the risk source data to be assessed in the industrial park and classify them into invariable variables, continuous variables and mutation variables; S2: Construct and train a neural network model that takes risk source data as input and outputs the warning results of various types of risk sources; including: Input layer, used to input risk source data; Feature extraction unit, including: A first feature extraction unit is used to extract an invariant feature based on an invariant variable; A second feature extraction unit is used to extract continuous features based on continuous variables; A third feature extraction unit is used to extract mutation features according to the mutation amount; The output layer is used to obtain the warning results of various types of risk sources based on the invariant features, continuity features and mutation features; S3: Obtain the current risk source data to be evaluated in the industrial park, input it into the trained neural network model, and evaluate the alarm results of each risk source.

2. The industrial park risk source level assessment method according to claim 1 is characterized in that: A feature extraction unit, comprising: a first convolutional layer, a first pooling layer, a hidden layer, a second convolutional layer, a random dropout layer, a third convolutional layer, and a second pooling layer connected in sequence; The first convolutional layer and the first pooling layer, as the first feature extraction unit, effectively extract invariant features; The hidden layer, the second convolutional layer and the random dropout layer, as the second feature extraction unit, effectively extract the continuity features; The third convolutional layer and the second pooling layer, as the third feature extraction unit, effectively extract mutation features.

3. The industrial park risk source level assessment method according to claim 2 is characterized in that: The convolution kernel of the first convolution layer is 3, the stride is 3, the padding is 2, and the activation function is rule; the convolution kernel of the second convolution layer is 3, the stride is 7, the padding is 3, and the activation function is rule; the convolution kernel of the third convolution layer is 5, the stride is 3, the padding is 2, and the activation function is rule.

4. The industrial park risk source level assessment method according to claim 1 is characterized in that: Step S3 also includes: obtaining historical risk source data to be assessed in the industrial park; The overall evaluation model includes: an overall input layer, two neural network models, a fully connected layer and an overall output layer; The total input layer is used to obtain the current risk source data and historical risk source data to be assessed in the industrial park; The first neural network model is connected to the total input layer and is used to output current features according to current risk source data; The second neural network model is connected to the total input layer and is used to output historical features based on historical risk source data; The fully connected layer is connected to the output of the first group of neural network models and the second group of neural network models, and is used to determine the overall alarm result based on current features and historical features, and output it through the overall output layer.

5. The industrial park risk source level assessment method according to claim 4 is characterized in that: The total alarm result is determined according to the current features and the historical features, specifically: the current features and the historical features are weighted and added to obtain the total alarm result.

6. The industrial park risk source level assessment method according to claim 1 is characterized in that: Training of neural network models, including: Obtain previous data, including key factors corresponding to each risk source, the likelihood and severity of each risk source; Based on the key factors, the likelihood and severity of each risk source, each risk source is graded using expert experience, the alarm results of each risk source are obtained, and a training data set is constructed based on previous data and alarm results; The neural network model is trained according to the training data set to obtain a trained neural network model.

7. The industrial park risk source level assessment method according to claim 1 is characterized in that: According to the risk source data, variables with only two possible answers, yes or no, are classified as unchangeable variables; variables with consistent and predictable change directions are classified as continuous variables; and variables with inconsistent and uncontrollable change directions are classified as mutation variables.

8. The industrial park risk source level assessment method according to claim 1 is characterized in that: Risk sources include: mechanical risk sources, electrical risk sources and operational risk sources; Mechanical risk sources include: risk sources related to transmission devices, risk sources related to drive devices, risk sources related to bearings, and risk sources related to fixing devices; Electrical risk sources include: risk sources related to power supply, risk sources related to control system, risk sources related to electrical equipment and risk sources related to ground wire; Operational risk sources include: risk sources related to operational control, risk sources related to safety warnings, and risk sources related to cleaning and maintenance.

9. An industrial park risk source level assessment device, characterized in that: It comprises a data acquisition unit, a model processing unit and a risk assessment unit which are connected in sequence; and is used to respectively execute the industrial park risk source level assessment method described in any one of claims 1 to 8.

10. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the industrial park risk source level assessment method as described in any one of claims 1 to 8 is implemented.

Images