Risk assessment processing method and device for power operation
Through machine learning training, the power operation risk assessment model, combined with the BP neural network and WOA algorithm, the problem of human factors in the operation safety risk assessment of power enterprises is solved, the reliability and safety of the assessment is improved, and the high risk is achieved Optimization processing of jobs.
Patent Information
- Application Number
- CN202410976197.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-11-12
AI Technical Summary
When conducting operation safety risk assessment, power enterprises are affected by human factors, resulting in low reliability of evaluation results. The accuracy and speed of the evaluation methods of the prior art cannot meet daily operation requirements, and historical data has a relatively good result for evaluation results. Great impact, lack of reverse optimization processing.
The power operation risk assessment model obtained by machine learning training is adopted. By obtaining work site data, calculating the risk coefficient of power operation, and comparing it with the reference risk coefficient, generating optimization strategies to reduce risks, using BP neural network and WOA algorithm for training, improving Accuracy and independence of the assessment.
It reduces the impact of human factors on evaluation, improves the safety of power operations and the objectivity of evaluation, ensures the reliability and speed of evaluation results, and realizes the optimization of high-risk operations.
Smart Images

Figure CN118941077A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of power grid safety technology, and in particular to a risk assessment processing method and device for power operations. Background Art
[0002] At present, power companies are constantly optimizing their requirements for power outages and fault elimination. However, the on-site environment for power operations is complex, the operation hazard index continues to reach new highs, and cases of personal accidents occurring during operations are common.
[0003] In the past, the operational safety risk assessment of power companies was greatly affected by human factors, and the influencing factors of some influencing factors at that time would be applied to the next risk assessment. In addition, the accuracy and speed of the current safety assessment method still cannot meet the requirements of daily operations, and the risk influencing factors considered are limited, which has brought certain obstacles to the subsequent optimization of work methods.
[0004] Although there are some intelligent operation risk assessment methods in the prior art, their calculation methods are fixed, and the correlation between the various influencing factors and the safety risk assessment algorithm are still greatly affected by the human factors of the assessor, and the accuracy of the assessment structure cannot be guaranteed. Moreover, the historical safety risk assessment data will have a greater correlation impact on the next risk assessment, which is not conducive to ensuring the independence and customer-oriented nature of the safety risk assessment before each operation. Moreover, the application of the operation risk assessment system in the current working method is limited to adjusting the type of assigned personnel for the operation, and lacks the link of reverse optimization of the data.
[0005] With regard to the problem in the above-mentioned related technologies that power companies are affected by human factors when conducting operational safety risk assessments, resulting in low reliability of the results of operational safety risk assessments, no effective solution has been proposed yet. Summary of the invention
[0006] The embodiments of the present invention provide a method and device for risk assessment of power operations, so as to at least solve the technical problem in the related art that power enterprises are affected by human factors when conducting operation safety risk assessment, resulting in low reliability of operation safety risk assessment results.
[0007] According to one aspect of an embodiment of the present invention, a risk assessment processing method for an electric power operation is provided, comprising: obtaining work site data of a target electric power operation, wherein the target electric power operation is an electric power operation that needs to be risk assessed; inputting the work site data into an electric power operation risk assessment model, so as to process the work site data using the electric power operation risk assessment model to obtain an electric power operation risk coefficient corresponding to the work site data, wherein the electric power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data comprising: sample work site data and a sample electric power operation risk coefficient corresponding to the sample work site data, wherein the sample work site data comprises safety risk influencing factors of the work site where the target electric power operation is located; comparing the electric power operation risk coefficient with a reference electric power operation risk coefficient, and determining that the target electric power operation is a high-risk electric power operation when the risk coefficient in the target electric power operation is higher than a risk coefficient threshold based on the comparison result, wherein the reference electric power operation risk coefficient is a risk coefficient obtained by risk assessment of historical electric power operations whose similarity to the target electric power operation is greater than a similarity threshold; generating an optimization strategy for the high-risk electric power operation; and optimizing the high-risk electric power operation according to the optimization strategy.
[0008] Optionally, before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, the power operation risk assessment processing method also includes: acquiring a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data; and training using the plurality of sets of training data including the plurality of the sample work site data and the plurality of the sample power operation risk coefficients to obtain the power operation risk assessment model.
[0009] Optionally, obtaining a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data comprises: converting the sample operation data corresponding to the sample work site into a vector format according to the sample risk coefficient of the sample work site and the sample risk coefficient influencing factor of the sample risk coefficient, to obtain the sample operation data in the vector format, wherein the sample operation data in the vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the sample operation data of the i-th group, represents the humidity corresponding to the sample operation data of the i-th group, represents the number of workers corresponding to the sample operation data of the i-th group, represents the climbing height corresponding to the sample operation data of the i-th group, represents the environmental visibility corresponding to the sample operation data of the i-th group, represents the fatigue level of the operator corresponding to the sample operation data of the i-th group, Representing the operation time corresponding to the i-th group of sample operation data; normalizing the sample operation data in the vector format to obtain the normalized sample operation data in the vector format; assigning the numerical value corresponding to the normalized sample operation data in the vector format to the influencing factor to obtain a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data.
[0010] Optionally, the multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients are used for training to obtain the power operation risk assessment model, including: constructing a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, the input layer is 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; assigning and encoding the network weights and thresholds of the BP neural network structure to obtain an initialized population; processing the initialized population by the WOA algorithm to obtain a target BP neural network structure with target weights and target thresholds; forward propagating the target BP neural network structure to obtain an output value; calculating the calculation error of each layer of neurons of the target BP neural network structure starting from the output layer, and updating the parameters of the target BP neural network structure according to the calculation error to obtain the updated target BP neural network structure; when the number of training times reaches the training times threshold or the calculation error is less than the error threshold, stopping the training to obtain the power operation risk assessment model.
[0011] Optionally, after training using the multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain the power operation risk assessment model, the power operation risk assessment processing method also includes: using the calculated error as a fitness function to calculate the individual fitness and update the whale position to obtain a new whale position; calculating the individual fitness, individual extreme value and global extreme value of the new whale position, and recording the optimal fitness and the new whale position; when it is determined that the iteration termination condition is met, outputting the optimal weight and the optimal threshold to the target BP neural network structure for training.
[0012] Optionally, the power operation risk coefficient is compared with a reference power operation risk coefficient, and when it is determined based on the comparison result that the risk coefficient of the target power operation is higher than a risk coefficient threshold, the target power operation is determined to be a high-risk power operation, including: determining the historical power operation whose similarity with the target power operation is greater than the similarity threshold; obtaining power-related data of the historical power operation and performing comparative analysis with the power data of the target power operation to determine the comparison result of the power operation risk coefficient of the historical power operation and the reference power operation risk coefficient of the target power operation; when it is determined that the comparison result indicates that the risk coefficient of the target power operation is higher than the risk coefficient threshold, the target power operation is determined to be the high-risk power operation.
[0013] Optionally, generating an optimization strategy for the high-risk power operation includes: determining high-risk power data for the high-risk power operation; inputting the high-risk power data into an optimization model, so as to process the high-risk power data using the optimization model to obtain the optimization strategy, wherein the optimization model is a model obtained by machine training using multiple sets of training data, and each of the multiple sets of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
[0014] According to another aspect of an embodiment of the present invention, there is also provided a risk assessment and processing device for power operations, comprising: a first acquisition unit, used to acquire work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment; a first processing unit, used to input the work site data into a power operation risk assessment model, so as to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data comprising: sample work site data and risk factors corresponding to the sample work site data a corresponding sample power operation risk coefficient, the sample work site data including safety risk influencing factors of the work site where the target power operation is located; a determination unit, used to compare the power operation risk coefficient with the reference power operation risk coefficient, and determine based on the comparison result that when the risk coefficient in the target power operation is higher than the risk coefficient threshold, the target power operation is determined to be a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than the similarity threshold; a generation unit, used to generate an optimization strategy for the high-risk power operation; a second processing unit, used to optimize the high-risk power operation according to the optimization strategy.
[0015] Optionally, the risk assessment and processing device for power operations also includes: a second acquisition unit, used to acquire a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain the power operation risk coefficient corresponding to the work site data; a training unit, used to perform training using the plurality of sets of training data including the plurality of the sample work site data and the plurality of the sample power operation risk coefficients to obtain the power operation risk assessment model.
[0016] Optionally, the second acquisition unit includes: a format conversion module, configured to convert the sample operation data corresponding to the sample work site into a vector format according to the sample risk coefficient of the sample work site and the sample risk coefficient influencing factor of the sample risk coefficient, so as to obtain the sample operation data in the vector format, wherein the sample operation data in the vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the sample operation data of the i-th group, represents the humidity corresponding to the sample operation data of the i-th group, represents the number of workers corresponding to the sample operation data of the i-th group, represents the climbing height corresponding to the sample operation data of the i-th group, represents the environmental visibility corresponding to the sample operation data of the i-th group, represents the fatigue level of the operator corresponding to the sample operation data of the i-th group, Represents the operation time corresponding to the i-th group of sample operation data; a normalization processing module, used to normalize the sample operation data in the vector format to obtain the sample operation data in the vector format after normalization; an assignment module, used to assign the numerical value corresponding to the sample operation data in the vector format after normalization to the influencing factor, so as to obtain multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data.
[0017] Optionally, the training unit includes: a construction module, which is used to construct a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer is 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; an assignment coding module, which is used to assign and code the network weights and thresholds of the BP neural network structure to obtain an initialized population; a first processing module, which is used to process the initialized population through a WOA algorithm to obtain a target BP neural network structure with target weights and target thresholds; a propagation processing module, which is used to perform forward propagation processing on the target BP neural network structure to obtain an output value; an update module, which is used to calculate the calculation error of each layer of neurons in the target BP neural network structure starting from the output layer, and update the parameters of the target BP neural network structure according to the calculation error to obtain the updated target BP neural network structure; a first training module, which is used to stop training when the number of training times reaches a training times threshold or the calculation error is less than an error threshold, to obtain the power operation risk assessment model.
[0018] Optionally, the risk assessment processing device for power operations also includes: an updating module, which is used to calculate the individual fitness and update the whale position using the calculation error as a fitness function to obtain a new whale position after training with the multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain the power operation risk assessment model; a calculation module, which is used to calculate the individual fitness, individual extreme value and global extreme value of the new whale position, and record the optimal fitness and the new whale position; a second training module, which is used to output the optimal weight and the optimal threshold to the target BP neural network structure for training when it is determined that the iteration termination condition is met.
[0019] Optionally, the determination unit includes: a first determination module, used to determine the historical power operation whose similarity with the target power operation is greater than the similarity threshold; an acquisition module, used to acquire the power-related data of the historical power operation and compare and analyze the power data of the target power operation to determine the comparison result of the power operation risk coefficient of the historical power operation and the reference power operation risk coefficient of the target power operation; a second determination module, used to determine that the target power operation is the high-risk power operation when it is determined that the comparison result indicates that the risk coefficient of the target power operation is higher than the risk coefficient threshold.
[0020] Optionally, the generation unit comprises: a third determination module, used to determine the high-risk power data of the high-risk power operation; a second processing module, used to input the high-risk power data into an optimization model, so as to process the high-risk power data using the optimization model to obtain the optimization strategy, wherein the optimization model is a model obtained by machine training using multiple groups of training data, and each of the multiple groups of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
[0021] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein the program executes any one of the above-mentioned risk assessment processing methods for power operations.
[0022] According to another aspect of an embodiment of the present invention, a processor is further provided, wherein the processor is used to run a program, wherein when the program is run, any one of the above-mentioned risk assessment processing methods for power operations is executed.
[0023] According to another aspect of an embodiment of the present invention, a computer program product is provided, comprising computer instructions, wherein when the computer instructions are executed by a processor, any one of the above-mentioned methods for risk assessment and processing of electric power operations is executed.
[0024] In an embodiment of the present invention, work site data of a target power operation is obtained, wherein the target power operation is a power operation that requires risk assessment; the work site data is input into a power operation risk assessment model, so as to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, and the sample work site data includes safety risk influencing factors of the work site where the target power operation is located; the power operation risk coefficient is compared with a reference power operation risk coefficient, and when it is determined based on the comparison result that the risk coefficient in the target power operation is higher than a risk coefficient threshold, the target power operation is determined to be a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than a similarity threshold; an optimization strategy for high-risk power operations is generated; and the high-risk power operations are optimized according to the optimization strategy. Through the technical solution provided by the present invention, it is possible to obtain a power operation risk assessment model through training with power operation data within a historical time period, which is used to determine the risk coefficient of the power operation, and when the power operation is determined to be a high-risk power operation based on the risk coefficient of the power operation, an optimization strategy is generated to optimize the power operation, thereby achieving the technical effect of improving the safety of the power operation, and further solving the technical problem in the related technology that the power enterprises are affected by human factors when conducting operation safety risk assessment, resulting in low reliability of the results of operation safety risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0026] Figure 1 It is a hardware structure block diagram of a mobile terminal of a risk assessment processing method for electric power operation according to an embodiment of the present invention;
[0027] Figure 2 is a flow chart of a risk assessment processing method for power operations according to an embodiment of the present invention;
[0028] Figure 3 is a workflow diagram of a data acquisition module according to an embodiment of the present invention;
[0029] Figure 4 is a schematic diagram of a BP neural network model according to an embodiment of the present invention;
[0030] Figure 5 is a schematic diagram of constructing a BP network structure according to an embodiment of the present invention;
[0031] Figure 6 is a flow chart of a WOA-BP neural network structure according to an embodiment of the present invention;
[0032] Figure 7 is a flow chart of a reverse optimization processing method according to an embodiment of the present invention;
[0033] Figure 8 is a flow chart of an optional risk assessment processing method for power operations according to an embodiment of the present invention;
[0034] Fig. 9 is a schematic diagram of a risk assessment processing device for power operations according to an embodiment of the present invention. DETAILED DESCRIPTION
[0035] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0036] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0037] As introduced in the background technology, the calculation method of the power operation assessment system in the related technology is fixed. The correlation between the influencing factors and the safety risk assessment algorithm are greatly affected by human factors, and the accuracy of the assessment results cannot be guaranteed. The historical safety risk assessment data will have a greater correlation impact on the next risk assessment, which is not conducive to ensuring the independence and objectivity of the safety risk assessment before the next operation. In view of the above defects, a risk assessment processing method and device for power operations are provided in an embodiment of the present invention.
[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0039] The method embodiments provided in the embodiments of the present invention can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for risk assessment of electric power operation according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.
[0040] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the risk assessment processing method of the power operation in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific example of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.
[0041] According to an embodiment of the present invention, a method embodiment of a risk assessment processing method for power operations is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0042] Figure 2 is a flow chart of a risk assessment processing method for power operations according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:
[0043] Step S202, obtaining work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment.
[0044] Optionally, the work site data of the above-mentioned target power operation may include but is not limited to: the power operation location, the start time and end time of the power operation, the power equipment used, and the working status of the operator.
[0045] Step S204, input the work site data into the power operation risk assessment model, so as to process the work site data using the power operation risk assessment model to obtain the power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model trained by machine learning using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, and the sample work site data includes the safety risk influencing factors of the work site where the target power operation is located.
[0046] In this embodiment, historical work site data and historical power operation risk coefficients within a historical time period can be first obtained as training data to obtain sample work site data and sample power operation risk coefficients corresponding to the sample work site data, and the sample work site data and the sample power operation risk coefficients corresponding to the sample work site data are trained to obtain a power operation risk assessment model, so that in subsequent use, the work site data can be directly input into the power operation risk assessment model, so that the power operation risk assessment model can be used to process the power operation risk assessment model to obtain the corresponding power operation risk coefficient, thereby improving the efficiency and accuracy of obtaining the power operation risk coefficient.
[0047] Step S206, compare the power operation risk coefficient with the reference power operation risk coefficient, and when it is determined based on the comparison result that the risk coefficient of the target power operation is higher than the risk coefficient threshold, determine the target power operation as a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than the similarity threshold.
[0048] In this embodiment, whether the target power operation is a high-risk power operation can be determined based on the power operation risk coefficient obtained above, so that subsequent processing can be better performed.
[0049] Step S208: generating an optimization strategy for high-risk power operations.
[0050] In this embodiment, an optimization strategy may be generated for high-risk power operations, and the high-risk power operations may be optimized through the optimization strategy to reduce the risk of the high-risk power operations.
[0051] Step S210: Optimize the high-risk power operations according to the optimization strategy.
[0052] As can be seen from the above, in an embodiment of the present invention, the work site data of the target power operation is obtained, wherein the target power operation is a power operation that requires risk assessment; the work site data is input into the power operation risk assessment model, so as to use the power operation risk assessment model to process the work site data and obtain the power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, the sample work site data includes safety risk influencing factors at the work site where the target power operation is located; the power operation risk coefficient is compared with the reference power operation risk coefficient A comparison is made, and when it is determined based on the comparison result that the risk coefficient of the target power operation is higher than the risk coefficient threshold, the target power operation is determined to be a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than the similarity threshold; an optimization strategy for the high-risk power operation is generated; and the high-risk power operation is optimized according to the optimization strategy, thereby achieving a power operation risk assessment model obtained by training the power operation data within a historical time period, which is used to determine the risk coefficient of the power operation, and when the power operation is determined to be a high-risk power operation according to the risk coefficient of the power operation, an optimization strategy is generated to optimize the power operation, thereby achieving the technical effect of improving the safety of the power operation.
[0053] It is worth noting that since the assessment of power operations is converted from a manual method to a machine learning method in the embodiment of the present invention, the influence of human factors in the risk assessment of power operations is greatly reduced, making the risk assessment of power operations more customized and improving the safety of power operations.
[0054] Therefore, the technical solution provided by the embodiment of the present invention solves the technical problem in the related art that power companies are affected by human factors when conducting operation safety risk assessment, resulting in low reliability of the results of operation safety risk assessment.
[0055] According to the above-mentioned embodiment of the present invention, before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain the power operation risk coefficient corresponding to the work site data, the power operation risk assessment processing method also includes: obtaining multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data; and training using multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain the power operation risk assessment model.
[0056] In this embodiment, the data acquisition module can be used to acquire data related to the safety risk coefficient of historical work, extract data features from the data, and use the extracted features as influencing factors of the safety risk coefficient of the work site.
[0057] In the above embodiment, obtaining a plurality of sample work site data and a plurality of sample power operation risk coefficients corresponding to the plurality of sample work site data includes: converting the sample operation data corresponding to the sample work site into a vector format according to the sample risk coefficient of the sample work site and the sample risk coefficient influencing factor of the sample risk coefficient, to obtain the sample operation data in the vector format, wherein the sample operation data in the vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the i-th group of sample operation data, represents the humidity corresponding to the i-th group of sample operation data, represents the number of workers corresponding to the i-th group of sample operation data, represents the climbing height corresponding to the i-th group of sample operation data, represents the environmental visibility corresponding to the i-th group of sample operation data, represents the fatigue level of the operator corresponding to the i-th group of sample operation data, Represents the operation time corresponding to the i-th group of sample operation data; normalizes the sample operation data in vector format to obtain the sample operation data in vector format after normalization; assigns the numerical value corresponding to the sample operation data in vector format after normalization to the influencing factor to obtain multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data.
[0058] In this embodiment, the vector conversion module in the data acquisition module can be used to convert the historical operation data into a vector form according to the corresponding relationship between the risk coefficient and the risk coefficient influencing factor, which can be specifically expressed by the following formula:
[0059] Figure 3 is a workflow diagram of a data acquisition module according to an embodiment of the present invention, such as Figure 3 As shown, the vector conversion module can be used to convert historical operation data into vector form according to the corresponding relationship between risk coefficient and risk coefficient influencing factors; then, the data normalization submodule can be used to normalize the data. The normalization can use min-max standardization, which is to transform the linearity of the original data so that the result falls within a specific range. The normalization can use the following formula: Where: x i is a certain characteristic value; x max is the maximum value of the eigenvalue; x min The minimum value of this eigenvalue.
[0060] like Figure 3 As shown, the initialization parameter submodule can also be used to assign initialization parameters to influencing factors.
[0061] According to the above embodiment of the present invention, a plurality of groups of training data including a plurality of sample work site data and a plurality of sample power operation risk coefficients are used for training to obtain a power operation risk assessment model, including: constructing a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer is 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; assigning and encoding the network weights and thresholds of the BP neural network structure to obtain an initialization population; processing the initialization population by the WOA algorithm to obtain a target BP neural network structure of target weights and target thresholds; forward propagating the target BP neural network structure to obtain an output value; calculating the calculation error of each layer of neurons of the target BP neural network structure starting from the output layer, and updating the parameters of the target BP neural network structure according to the calculation error to obtain an updated target BP neural network structure; when the number of training times reaches the training times threshold or the calculation error is less than the error threshold, stopping the training to obtain a power operation risk assessment model.
[0062] In this embodiment, a BP neural network structure can be constructed first, and the influencing factors and risk coefficient data can be divided into a training set and a test set in a ratio of 8:2 to train and test the neural network. Each neuron in the BP neural network model receives input signals from other neurons. Figure 4 is a schematic diagram of a BP neural network model according to an embodiment of the present invention, such as Figure 4 As shown, each signal (x1, x2…x i …x n ) are all passed through a weighted (w1, w2…w i …w n ) connection transmission, the neuron adds up these signals to get a total input value, then compares the total input value with the threshold of the neuron (simulating the threshold potential), and then processes it through an activation function to get the final output (simulating the activation of the cell), which will be passed down layer by layer as the input of the subsequent neurons, where the output is θ is the threshold, x i is the input from the ith neuron, ω i is the connection weight of the ith neuron.
[0063] Among them, the specific implementation steps of the above method are as follows: determine the neural network structure as a BP network with an input layer of 7 nodes, a hidden layer of 5 nodes, and an output layer of 1 node. Its network structure is as follows: a BP network with an input layer of 7 nodes, a hidden layer of 5 nodes, and an output layer of 1 node. Figure 5 Schematic diagram of constructing a BP network structure according to an embodiment of the present invention. Figure 5 As shown, the neural network structure is a BP network with 7 nodes in the input layer, 5 nodes in the hidden layer and 1 node in the output layer.
[0064] Then, the weights and thresholds of the neural network are assigned and encoded to obtain the initial population, and then the optimal weights and thresholds are obtained through the WOA algorithm to construct a new network structure; and forward propagation is performed. The transmission direction of the forward propagation process is from the input layer to the output layer, and the data forward propagation is completed through the mapping relationship between input and output. During the propagation process, the input layer and related network parameters need to be adjusted linearly, and the activation function is used to complete the linear to nonlinear conversion, and the result is
[0065]
[0066] To the output value, the forward calculation formula is as follows: j =f(s j ), where i and j are the neuron nodes of the BP neural network respectively; ω ij is the weight between node i and node j; b fis the threshold of node j; x i is the input value of node i; f is the activation function; y j is the output value of the node.
[0067] In addition, error calculation is required. The calculation error of each layer of neurons can be calculated reversely from the output layer and the parameters can be updated. Let the output layer result be y j , a single training sample has n-1 outputs, and the loss function is calculated as follows: Where, d j is the expected output value of the corresponding sample.
[0068] When logical judgment determines that the training accuracy has been reached or the maximum number of training times has been reached, the prediction output of the last step is entered; if the training accuracy has not been reached or the maximum number of training times has not been reached, the weight or threshold is continued to be updated.
[0069] According to the above embodiment of the present invention, after training using multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain a power operation risk assessment model, the power operation risk assessment processing method also includes: using the calculation error as the fitness function to calculate the individual fitness and update the whale position to obtain a new whale position; calculating the individual fitness, individual extreme value and global extreme value of the new whale position, and recording the optimal fitness and the new whale position; when it is determined that the iteration termination condition is met, outputting the optimal weight and the optimal threshold to the target BP neural network structure for training.
[0070] In this embodiment, algorithm-related parameters can be initialized, where the parameters set in this algorithm are as follows: the number of whale populations N = 30, the maximum number of iterations k max = 1000. Then, the error of BP neural network is used as fitness function to calculate individual fitness and update the whale position, where the whale position update formula is: Where: k is the current iteration number of the algorithm; X(k) is the position of the individual whale at the kth iteration; X(k+1) is the position of the individual whale at the k+1th iteration; X p (k) is the prey position; A and C are the adjustment coefficients of the whale position; r1 and r2 are random numbers in [0,1]; a is the control factor, and the value of the control factor is determined by the number of iterations of the algorithm; k max is the maximum number of iterations of the algorithm.
[0071] In addition, it is necessary to find individual and global extreme values: update the whale position; calculate the individual fitness of all new positions; update the individual and global extreme values, record the optimal fitness and updated whale position and replace them; determine whether the iteration termination condition is met; if so, output the optimal weights and thresholds to the BP neural network for training; if not, return to the step of updating the whale position and repeat this step and the steps that follow.
[0072] in, Figure 6 is a flow chart of the WOA-BP neural network structure according to an embodiment of the present invention. Figure 6 As shown, it includes the WOA algorithm part and the BP neural network part. Among them, in the WOA algorithm part, firstly, it is necessary to input the data related to the power operation, and then perform data processing; after the algorithm related parameters are initialized, the result of the data processing is used as the BP neural network error as the fitness function; then find the individual and global extreme values; update the whale position, and calculate the individual fitness through WOA; after updating the individual and global extreme values, determine whether the termination condition is met, if so, obtain the optimal weight and threshold, and assign it to the new network structure; otherwise, return to continue to find the individual and global extreme values; after assigning the obtained optimal weight and threshold to the new network structure, enter the forward propagation process and calculate the error; determine whether the number of training times or accuracy has been reached, if so, predict the output, otherwise update the weight and threshold.
[0073] According to the above embodiment of the present invention, the risk coefficient of an electric power operation is compared with the risk coefficient of a reference electric power operation, and when it is determined based on the comparison result that the risk coefficient of the target electric power operation is higher than the risk coefficient threshold, the target electric power operation is determined to be a high-risk electric power operation, including: determining a historical electric power operation whose similarity to the target electric power operation is greater than the similarity threshold; obtaining electric power-related data of the historical electric power operation and performing comparative analysis with the electric power data of the target electric power operation to determine a comparison result between the electric power operation risk coefficient of the historical electric power operation and the reference electric power operation risk coefficient of the target electric power operation; when it is determined that the comparison result indicates that the risk coefficient of the target electric power operation is higher than the risk coefficient threshold, the target electric power operation is determined to be a high-risk electric power operation.
[0074] In this embodiment, historical power operations whose similarity with the target power operation is greater than a similarity threshold can be found, and then power-related data of the historical power operations are obtained; they are compared and analyzed with the power data of the target power operation to determine whether the target power operation is a high-risk power operation.
[0075] According to the above embodiment of the present invention, an optimization strategy for high-risk power operations is generated, including: determining high-risk power data of high-risk power operations; inputting the high-risk power data into an optimization model, so as to process the high-risk power data using the optimization model to obtain an optimization strategy, wherein the optimization model is a model obtained by machine training using multiple sets of training data, and each set of the multiple sets of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
[0076] In this embodiment, an optimization model can be pre-trained and used to process high-risk power data to obtain an optimization strategy for optimizing high-risk power operations, so that the optimization strategy corresponding to the high-risk power data can be obtained more quickly, thereby improving the safety of power operations and reducing the risks of power operations.
[0077] Figure 7 is a flowchart of a reverse optimization processing method according to an embodiment of the present invention. Figure 7 As shown, the relevant data of this assessment can be interactively compared with the data of similar historical work tasks; for work plans with higher safety risk assessment values, the optimization range and rational modification suggestions of various factors in the work implementation plan are given; and the modification suggestions are fed back to the relevant departments for processing. Here, based on the effective interactive analysis of the safety risk coefficient and historical data, reverse optimization suggestions are made for the adjustment of the working method and various influencing factors, and feedback is sent to the relevant departments, further improving the safety of power operations.
[0078] Figure 8 is a flow chart of an optional risk assessment processing method for power operations according to an embodiment of the present invention. Figure 8 As shown in the figure, after obtaining the safety risk factor related data of historical work, the data features in the data are extracted, and the extracted features are used as the influencing factors of the safety risk factor of the work site; then, the BP neural network structure is constructed, and the influencing factors and risk factor data are divided into training set and test set in the ratio of 8:2, and the neural network is trained and tested; the whale optimization algorithm is used to optimize the weights and thresholds of the BP neural network; the current work site data is input into the BP neural network, and the safety risk factor is output; then, based on the effective interactive analysis of the safety risk factor and historical data, reverse optimization suggestions are made for the adjustment of the working method and various influencing factors, and feedback is sent to the relevant parts.
[0079] Through the above technical solution provided by the embodiment of the present invention, the adaptive enhancement between the input data of the safety risk assessment system and the effective combination of individual and global information are realized through the training of data samples, thereby improving the accuracy of the network in practical applications; the system does not add the connection part between the current output data and the historical data, eliminating the greater influence of historical data on the risk assessment of this operation, so as to ensure the independence and objectivity of this risk coefficient. In addition, the model for reverse optimization of the operation mode and related influencing factors, through the analysis of the safety risk coefficient, proposes reasonable adjustment suggestions for each work and feeds back to the relevant departments, so as to facilitate the operation person in charge to adjust the proportion of each influencing factor affecting the risk coefficient in the on-site work, optimize the work implementation plan, and effectively control the safety risk coefficient.
[0080] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.
[0082] According to an embodiment of the present invention, there is also provided a risk assessment processing device for electric power operations for implementing the risk assessment processing method for electric power operations. Fig. 9 is a schematic diagram of a risk assessment processing device for power operations according to an embodiment of the present invention. Fig. 9 As shown, the device includes: a first acquisition unit 901, a first processing unit 903, a determination unit 905, a generation unit 907 and a second processing unit 909. The risk assessment processing device for power operation is described in detail below.
[0083] The first acquisition unit 901 is used to acquire work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment.
[0084] The first processing unit 903 is used to input the work site data into the power operation risk assessment model so as to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, and the sample work site data includes safety risk influencing factors of the work site where the target power operation is located.
[0085] Determination unit 905 is used to compare the risk coefficient of the power operation with the risk coefficient of the reference power operation, and when it is determined based on the comparison result that the risk coefficient of the target power operation is higher than the risk coefficient threshold, the target power operation is determined to be a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity with the target power operation is greater than the similarity threshold.
[0086] The generating unit 907 is used to generate an optimization strategy for high-risk power operations.
[0087] The second processing unit 909 is used to optimize the high-risk power operations according to the optimization strategy.
[0088] It should be noted here that the above-mentioned first acquisition unit 901, first processing unit 903, determination unit 905, generation unit 907 and second processing unit 909 correspond to steps S202 to S210 in the above-mentioned embodiments, and the five units are the same as the instances and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the above-mentioned embodiments.
[0089] As can be seen from the above, in the scheme recorded in the above-mentioned embodiment of the present invention, the first acquisition unit can be used to acquire the work site data of the target power operation, wherein the target power operation is the power operation that needs to be risk assessed; the first processing unit is used to input the work site data into the power operation risk assessment model, so as to process the work site data using the power operation risk assessment model to obtain the power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, the sample work site data including the safety risk influencing factors of the work site where the target power operation is located; the determination unit is used to compare the power operation risk coefficient with the reference power operation risk coefficient. The power operation risk coefficient is compared with the target power operation risk coefficient, and when it is determined based on the comparison result that the risk coefficient of the target power operation is higher than the risk coefficient threshold, the target power operation is determined to be a high-risk power operation, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than the similarity threshold; the generation unit is used to generate an optimization strategy for the high-risk power operation; and the second processing unit is used to optimize the high-risk power operation according to the optimization strategy, so that a power operation risk assessment model is obtained by training the power operation data within the historical time period, which is used to determine the risk coefficient of the power operation, and when the power operation is determined to be a high-risk power operation according to the risk coefficient of the power operation, an optimization strategy is generated to optimize the power operation, thereby achieving the technical effect of improving the safety of the power operation.
[0090] It is worth noting that since the assessment of power operations is converted from a manual method to a machine learning method in the embodiment of the present invention, the influence of human factors in the risk assessment of power operations is greatly reduced, making the risk assessment of power operations more customized and improving the safety of power operations.
[0091] Optionally, the risk assessment and processing device for power operations also includes: a second acquisition unit, used to acquire multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain the power operation risk coefficient corresponding to the work site data; a training unit, used to perform training using multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain the power operation risk assessment model.
[0092] Optionally, the second acquisition unit includes: a format conversion module, which is used to convert the sample operation data corresponding to the sample work site into a vector format according to the sample risk coefficient of the sample work site and the sample risk coefficient influencing factor of the sample risk coefficient, so as to obtain the sample operation data in the vector format, wherein the sample operation data in the vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the i-th group of sample operation data, represents the humidity corresponding to the i-th group of sample operation data, represents the number of workers corresponding to the i-th group of sample operation data, represents the climbing height corresponding to the i-th group of sample operation data, represents the environmental visibility corresponding to the i-th group of sample operation data, represents the fatigue level of the operator corresponding to the i-th group of sample operation data, Represents the operation time corresponding to the i-th group of sample operation data; a normalization processing module, used to normalize the sample operation data in vector format to obtain the sample operation data in vector format after normalization; an assignment module, used to assign the numerical value corresponding to the sample operation data in vector format after normalization to the influencing factor, so as to obtain multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data.
[0093] Optionally, the training unit includes: a construction module for constructing a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer has 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; an assignment coding module for assigning and coding the network weights and thresholds of the BP neural network structure to obtain an initialized population; a first processing module for processing the initialized population through a WOA algorithm to obtain a target BP neural network structure with target weights and target thresholds; a propagation processing module for performing forward propagation processing on the target BP neural network structure to obtain an output value; an update module for calculating the calculation error of each layer of neurons in the target BP neural network structure starting from the output layer, and updating the parameters of the target BP neural network structure according to the calculation error to obtain an updated target BP neural network structure; a first training module for stopping training when the number of training times reaches a training times threshold or the calculation error is less than an error threshold to obtain a power operation risk assessment model.
[0094] Optionally, the risk assessment processing device for power operations also includes: an updating module, which is used to calculate the individual fitness and update the whale position using the calculation error as the fitness function after training using multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain a power operation risk assessment model, so as to obtain a new whale position; a calculation module, which is used to calculate the individual fitness, individual extreme value and global extreme value of the new whale position, and record the optimal fitness and the new whale position; a second training module, which is used to output the optimal weight and the optimal threshold to the target BP neural network structure for training when it is determined that the iteration termination condition is met.
[0095] Optionally, the determination unit includes: a first determination module, used to determine historical power operations whose similarity with the target power operation is greater than a similarity threshold; an acquisition module, used to acquire power-related data of the historical power operations and compare and analyze the power data of the target power operation to determine a comparison result between the power operation risk coefficient of the historical power operation and the reference power operation risk coefficient of the target power operation; and a second determination module, used to determine that the target power operation is a high-risk power operation when the comparison result indicates that the risk coefficient of the target power operation is higher than the risk coefficient threshold.
[0096] Optionally, the generation unit includes: a third determination module, used to determine high-risk power data of high-risk power operations; a second processing module, used to input the high-risk power data into an optimization model, so as to process the high-risk power data using the optimization model to obtain an optimization strategy, wherein the optimization model is a model obtained by machine training using multiple sets of training data, and each set of the multiple sets of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
[0097] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored program, wherein the program executes any one of the above-mentioned risk assessment processing methods for power operations.
[0098] Optionally, in this embodiment, the computer-readable storage medium may be located in any one of the computer terminals in a computer terminal group in a computer network, or in any one of the communication devices in a communication device group.
[0099] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for executing the following steps: obtaining work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment; inputting the work site data into a power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, the sample work site data including safety risk influencing factors of the work site where the target power operation is located; comparing the power operation risk coefficient with the reference power operation risk coefficient, and determining that the target power operation is a high-risk power operation when the risk coefficient in the target power operation is higher than a risk coefficient threshold based on the comparison result, wherein the reference power operation risk coefficient is a risk coefficient obtained by risk assessment of historical power operations whose similarity to the target power operation is greater than a similarity threshold; generating an optimization strategy for high-risk power operations; and optimizing the high-risk power operation according to the optimization strategy.
[0100] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, obtaining a plurality of sample work site data and a plurality of sample power operation risk coefficients corresponding to the plurality of sample work site data; and training using a plurality of groups of training data including a plurality of sample work site data and a plurality of sample power operation risk coefficients to obtain a power operation risk assessment model.
[0101] Optionally, in this embodiment, the computer-readable storage medium is configured to store a program code for executing the following steps: converting sample operation data corresponding to the sample operation site into a vector format according to the sample risk coefficient of the sample operation site and the sample risk coefficient influencing factor of the sample risk coefficient, to obtain sample operation data in a vector format, wherein the sample operation data in a vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the i-th group of sample operation data, represents the humidity corresponding to the i-th group of sample operation data, represents the number of workers corresponding to the i-th group of sample operation data, represents the climbing height corresponding to the i-th group of sample operation data, represents the environmental visibility corresponding to the i-th group of sample operation data, represents the fatigue level of the operator corresponding to the i-th group of sample operation data, Represents the operation time corresponding to the i-th group of sample operation data; normalizes the sample operation data in vector format to obtain the sample operation data in vector format after normalization; assigns the numerical value corresponding to the sample operation data in vector format after normalization to the influencing factor to obtain multiple sample work site data and multiple sample power operation risk coefficients corresponding to the multiple sample work site data.
[0102] Optionally, in this embodiment, the computer-readable storage medium is configured to store program codes for executing the following steps: constructing a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer has 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; assigning and encoding network weights and thresholds of the BP neural network structure to obtain an initialized population; processing the initialized population through the WOA algorithm to obtain a target BP neural network structure with target weights and target thresholds; performing forward propagation processing on the target BP neural network structure to obtain an output value; calculating the calculation error of each layer of neurons in the target BP neural network structure starting from the output layer, and updating the parameters of the target BP neural network structure according to the calculation error to obtain an updated target BP neural network structure; when the number of training times reaches the training times threshold or the calculation error is less than the error threshold, stopping the training to obtain a power operation risk assessment model.
[0103] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: after training using multiple groups of training data including multiple sample work site data and multiple sample power operation risk coefficients to obtain a power operation risk assessment model, use the calculated error as a fitness function to calculate the individual fitness and update the whale position to obtain a new whale position; calculate the individual fitness, individual extreme value and global extreme value of the new whale position, and record the optimal fitness and the new whale position; when it is determined that the iteration termination condition is met, output the optimal weight and the optimal threshold to the target BP neural network structure for training.
[0104] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining historical power operations whose similarity to the target power operation is greater than a similarity threshold; obtaining power-related data of the historical power operations and performing comparative analysis with the power data of the target power operation to determine a comparison result between the power operation risk coefficient of the historical power operation and the reference power operation risk coefficient of the target power operation; and determining that the target power operation is a high-risk power operation when it is determined that the comparison result indicates that the risk coefficient in the target power operation is higher than the risk coefficient threshold.
[0105] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining high-risk power data of high-risk power operations; inputting the high-risk power data into an optimization model to process the high-risk power data using the optimization model to obtain an optimization strategy, wherein the optimization model is a model obtained by machine training using multiple sets of training data, and each set of the multiple sets of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
[0106] According to another aspect of an embodiment of the present invention, a processor is further provided, and the processor is used to run a program, wherein when the program is run, any one of the above-mentioned risk assessment processing methods for power operations is executed.
[0107] According to another aspect of an embodiment of the present invention, there is also provided a computer program product, including computer instructions, which, when executed by a processor, execute any one of the above-mentioned risk assessment processing methods for power operations.
[0108] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0109] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0110] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0111] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0112] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0113] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
[0114] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A risk assessment method for power operations, characterized in that: include: Acquiring work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment; The work site data is input into an electric power operation risk assessment model, so as to process the work site data using the electric power operation risk assessment model to obtain an electric power operation risk coefficient corresponding to the work site data, wherein the electric power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data includes: sample work site data and a sample electric power operation risk coefficient corresponding to the sample work site data, and the sample work site data includes safety risk influencing factors of the work site where the target electric power operation is located; The risk coefficient of the electric power operation is compared with the risk coefficient of a reference electric power operation, and when it is determined based on the comparison result that the risk coefficient of the target electric power operation is higher than a risk coefficient threshold, the target electric power operation is determined to be a high-risk electric power operation, wherein the risk coefficient of the reference electric power operation is a risk coefficient obtained by risk assessment of historical electric power operations whose similarity to the target electric power operation is greater than a similarity threshold; generating an optimization strategy for the high-risk power operation; The high-risk power operations are optimized according to the optimization strategy.
2. The risk assessment and processing method for power operations according to claim 1 is characterized in that: Before inputting the work site data into the power operation risk assessment model to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, the method further includes: Acquire a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data; The power operation risk assessment model is obtained by training using the multiple sets of training data including the multiple sample work site data and the multiple sample power operation risk coefficients.
3. The risk assessment and processing method for power operations according to claim 2 is characterized in that: Acquiring a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data comprises: The sample operation data corresponding to the sample work site is converted into a vector format according to the sample risk coefficient of the sample work site and the sample risk coefficient influencing factor of the sample risk coefficient to obtain the sample operation data in the vector format, wherein the sample operation data in the vector format is: Y i represents the risk coefficient of the i-th group of sample operation data, represents the temperature corresponding to the sample operation data of the i-th group, represents the humidity corresponding to the sample operation data of the i-th group, represents the number of workers corresponding to the sample operation data of the i-th group, represents the climbing height corresponding to the sample operation data of the i-th group, represents the environmental visibility corresponding to the sample operation data of the i-th group, represents the fatigue level of the operator corresponding to the sample operation data of the i-th group, represents the operation time corresponding to the sample operation data of the i-th group; Normalizing the sample operation data in the vector format to obtain the normalized sample operation data in the vector format; The numerical values corresponding to the normalized sample operation data in vector format are assigned to the influencing factors to obtain a plurality of the sample work site data and a plurality of the sample power operation risk coefficients corresponding to the plurality of the sample work site data.
4. The risk assessment and processing method for power operations according to claim 2 is characterized in that: The power operation risk assessment model is obtained by training using the multiple sets of training data including the multiple sample work site data and the multiple sample power operation risk coefficients, including: Constructing a BP neural network structure, wherein the network structure of the BP neural network includes: an input layer, a hidden layer, and an output layer, wherein the input layer has 7 nodes, the hidden layer includes 5 nodes, and the output layer includes 1 node; Assigning and encoding the network weights and thresholds of the BP neural network structure to obtain an initialized population; The initialized population is processed by the WOA algorithm to obtain a target BP neural network structure with target weights and target thresholds; Performing forward propagation processing on the target BP neural network structure to obtain an output value; Calculating the calculation error of each layer of neurons in the target BP neural network structure starting from the output layer, and updating the parameters of the target BP neural network structure according to the calculation error to obtain the updated target BP neural network structure; When the number of training times reaches a training times threshold or the calculation error is less than an error threshold, the training is stopped to obtain the power operation risk assessment model.
5. The risk assessment and processing method for power operations according to claim 4 is characterized in that: After the plurality of groups of training data including the plurality of sample work site data and the plurality of sample power operation risk coefficients are used for training to obtain the power operation risk assessment model, the method further includes: Calculating individual fitness using the calculation error as a fitness function and updating the whale position to obtain a new whale position; Calculate the individual fitness, individual extreme value and global extreme value of the new whale position, and record the optimal fitness and the new whale position; When it is determined that the iteration termination condition is met, the optimal weight and the optimal threshold are output to the target BP neural network structure for training.
6. The risk assessment and processing method for power operations according to claim 1, characterized in that: The risk coefficient of the electric power operation is compared with the risk coefficient of the reference electric power operation, and when it is determined based on the comparison result that the risk coefficient of the target electric power operation is higher than a risk coefficient threshold, the target electric power operation is determined to be a high-risk electric power operation, including: Determining the historical power operation whose similarity to the target power operation is greater than the similarity threshold; Obtaining the power-related data of the historical power operation and the power data of the target power operation for comparative analysis to determine the comparison result of the power operation risk coefficient of the historical power operation and the reference power operation risk coefficient of the target power operation; When it is determined that the comparison result indicates that the risk factor of the target power operation is higher than a risk factor threshold, the target power operation is determined to be the high-risk power operation.
7. The risk assessment and processing method for power operations according to claim 1, characterized in that: Generate an optimization strategy for the high-risk power operation, including: Determining high-risk power data for the high-risk power operation; The high-risk power data is input into an optimization model so as to process the high-risk power data using the optimization model to obtain the optimization strategy, wherein the optimization model is a model obtained by machine training using multiple sets of training data, and each of the multiple sets of training data includes sample high-risk power data and a sample optimization strategy corresponding to the sample high-risk power data.
8. A risk assessment and processing device for power operations, characterized in that: include: A first acquisition unit is used to acquire work site data of a target power operation, wherein the target power operation is a power operation that requires risk assessment; a first processing unit, configured to input the work site data into a power operation risk assessment model, so as to process the work site data using the power operation risk assessment model to obtain a power operation risk coefficient corresponding to the work site data, wherein the power operation risk assessment model is a model obtained by machine learning training using multiple sets of training data, each of the multiple sets of training data comprising: sample work site data and a sample power operation risk coefficient corresponding to the sample work site data, wherein the sample work site data comprises safety risk influencing factors of the work site where the target power operation is located; a determination unit, configured to compare the risk coefficient of the electric power operation with a risk coefficient of a reference electric power operation, and determine, based on the comparison result, that when the risk coefficient of the target electric power operation is higher than a risk coefficient threshold, to determine that the target electric power operation is a high-risk electric power operation, wherein the risk coefficient of the reference electric power operation is a risk coefficient obtained by risk assessment of historical electric power operations whose similarity to the target electric power operation is greater than a similarity threshold; A generating unit, used for generating an optimization strategy for the high-risk power operation; The second processing unit is used to optimize the high-risk power operation according to the optimization strategy.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the risk assessment processing method for electric power operations according to any one of claims 1 to 7.
10. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the risk assessment processing method for power operations described in any one of claims 1 to 7 is performed.
Citation Information
Cited By
Electric power operation risk early warning method and system based on knowledge enhancement and multi-modal fusion
CN121836365A
