Multi-type sensor scheduling method and system based on petrochemical scenarios

Abstract

The present application relates to the technical field of sensor access scheduling, and discloses a multi-type sensor scheduling method and system based on a petrochemical scenario. The method includes: deploying multiple types of sensors at key locations in the petrochemical scenario to collect monitoring data and remaining power in real time; periodically calculating the risk deviation of the monitoring data collected by the sensor, the proportion of the remaining power of the sensor, and the correlation score between the sensor and the operation plan, and determining the priority of the sensor based on the risk deviation, the proportion of the remaining power, and the correlation score; performing cross-language call adaptation and communication protocol adaptation on different types of sensors based on preset adapters, processing the monitoring data collected by sensors of different priorities differently, and sending high-priority monitoring data first; analyzing specific situations and adjusting the sampling frequency. The present invention can accurately obtain monitoring data, reduce energy consumption, reduce the waste of network and storage resources, and improve data transmission reliability and processing efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of sensor access scheduling, and in particular to a multi-type sensor scheduling method and system based on a petrochemical scenario. Background Art

[0002] In the petrochemical industry, multiple types of sensors are deployed at key locations within the production process to monitor various parameters during production. However, existing technologies typically use a fixed sampling frequency for sensors, which can lead to data redundancy, excessive network and storage resource usage, increased sensor energy consumption, and shortened sensor lifespans. Furthermore, different types of sensors may utilize different communication protocols and programming languages, complicating the unified processing and transmission of data. Therefore, a scheduling method and system are needed that can dynamically adjust sensor sampling frequencies based on actual conditions and effectively process data from different sensor types.

[0003] A similar prior art Chinese patent application with publication number CN119201458A provides a multi-task sensor scheduling method and system, including: a rotating agent collects all tasks to obtain a first task list, and formulates a sensor scheduling plan based on the sensor requirements of all tasks therein; other agents collect all tasks to obtain a second task list, and if the second task list contains more tasks than the first task list, they vote against it; otherwise, they vote in favor; the rotating agent then counts whether all votes are in favor, and if so, performs sensor scheduling based on the sensor scheduling plan, and selects a new rotating agent to enter the next rotation cycle; otherwise, the agent that voted against it publishes the second task list, and selects a new rotating agent to enter the next rotation cycle.

[0004] A similar prior art includes a Chinese patent application with publication number CN118410315A, which provides a fossil information data processing system and method based on multi-dimensional analysis, including: integrating and calculating the element existence identifiers corresponding to all registration items in the early identification processing to obtain a reliable matching identifier of the overall dimension, performing data analysis on the reliable matching identifier to determine whether the sample fossil to be verified is successfully matched and providing a prompt; integrating and calculating data of different dimensions on the audit registration items to which the recognition abnormality status identifier obtained in the early automatic identification processing belongs to obtain a corresponding recognition abnormality impact value, performing data analysis on the recognition abnormality impact value to determine whether the corresponding audit registration item needs recognition training optimization.

[0005] However, the above two technical solutions do not consider the problem of data redundancy caused by using a fixed sampling frequency when using sensors to collect data. Therefore, the present invention provides a multi-type sensor scheduling method and system based on a petrochemical scenario. Summary of the Invention

[0006] This application provides a multi-type sensor scheduling method and system based on petrochemical scenarios, which is used to dynamically adjust the frequency of sensor data collection, reduce sensor energy consumption, delay sensor usage time, reduce network and storage resource waste, and improve data transmission reliability and processing efficiency.

[0007] In a first aspect, the present application provides a multi-type sensor scheduling method based on a petrochemical scenario, the method comprising:

[0008] Step S1: deploy various types of sensors at multiple key locations in the petrochemical scene to collect monitoring data and sensor remaining power in real time;

[0009] Step S2: Periodically calculate the risk deviation of the monitoring data collected by the sensor, calculate the remaining battery percentage of the sensor, calculate the correlation score between the sensor and the operation plan, determine the priority of the sensor based on the risk deviation, remaining battery percentage, and correlation score, and assign different sampling frequencies to different priorities;

[0010] Step S3: Perform cross-language call adaptation and communication protocol adaptation for different types of sensors based on preset adapters. According to the priority of the sensors, the monitoring data collected by sensors with different priorities are processed differently, and the monitoring data with high priority is sent first.

[0011] Step S4: Analyze whether it is in a specific situation. If so, modify the sampling frequency of the sensor. The specific situations include abnormal events, key nodes of the operation plan, sensor power below the preset minimum threshold, and data transmission delay.

[0012] In conjunction with the first aspect, in a first implementation of the first aspect of the present application, calculating the risk deviation of monitoring data collected by the sensor includes:

[0013] Obtain the safety threshold corresponding to the monitoring data, obtain the currently collected real-time monitoring data, divide the absolute value of the result obtained by subtracting the safety threshold from the real-time monitoring data by the safety threshold to obtain a first value, use the prediction model to predict the predicted monitoring data of the monitoring data in the future, divide the absolute value of the result obtained by subtracting the safety threshold from the predicted monitoring data by the safety threshold to obtain a second value, set corresponding weight values ​​for the first value and the second value, multiply the first value and the second value by the corresponding weight values ​​respectively, and then add the obtained result value as the risk deviation.

[0014] In combination with the first aspect, in a second implementation of the first aspect of the present application, calculating the correlation score between the sensor and the operation plan includes:

[0015] For each sensor, obtain the basic correlation value of the current sensor relative to each operation plan, obtain the current operation plan, obtain the duration of the current operation plan and the theoretical maximum duration, divide the duration by the maximum duration and multiply it by the corresponding first correction coefficient and the basic correlation value to obtain a first correction value, obtain the correlation coefficient of each sensor and other sensors, calculate the average of all correlation coefficients for each sensor, multiply the average of the correlation coefficients by the corresponding second correction coefficient and the basic correlation value to obtain a second correction value, and add the basic correlation value, the first correction value, and the second correction value to obtain a correlation score.

[0016] In combination with the first aspect, in a third implementation of the first aspect of the present application, determining the priority of the sensor based on the risk deviation, the remaining power ratio, and the correlation score includes:

[0017] The risk deviation, the remaining power ratio and the correlation score are called three data parameters. The three scores of each data parameter relative to other data parameters are obtained. For each data parameter, the sum of all the medians of the three scores is calculated as the first result value. The three scores are divided by the first result value respectively to obtain three ratios. The average of the three ratios is calculated. The average of the three averages is used as the weight value of the corresponding data parameter. The priority of the sensor is determined based on the data parameters and the corresponding weight values.

[0018] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, determining the priority of the sensor based on the data parameter and the corresponding weight value includes:

[0019] Multiple different priorities are preset. For each priority, a corresponding priority score is set for each data parameter based on the numerical value of each data parameter. The weight value and the priority score corresponding to each priority are multiplied and then added to obtain the priority score of the corresponding priority. All priority scores are obtained, and the priority corresponding to the maximum priority score is used as the priority of the corresponding sensor.

[0020] In combination with the first aspect, in a fifth implementation of the first aspect of the present application, monitoring data collected by sensors of different priorities are processed differently, including:

[0021] The monitoring data is divided into multiple data blocks, and the corresponding sensor priority identifier is added to each data block. The corresponding verification data is generated based on the data block, and the network status is obtained in real time. When the network status is good, all data blocks, priority identifiers and verification data are sent to the data receiving end at the same time. When the network status is not good, the data blocks corresponding to the high priority are sent first, and then the data blocks corresponding to the low priority are sent.

[0022] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, monitoring data collected by sensors of different priorities are processed differently, including:

[0023] After receiving the data block, the data receiving end stores the data blocks corresponding to different priority identifiers in different memory areas respectively, and combines the data blocks in the memory areas in descending order of priority to restore them to the original monitoring data.

[0024] In combination with the first aspect, in a seventh implementation of the first aspect of the present application, before combining and restoring the data blocks, the method includes:

[0025] The data receiving end performs error checking on the received data blocks. If an error occurs, it records a request to resend the corresponding data block, compares the erroneous data block with the correct data block, and records the corresponding error bits. After several comparisons, the number of errors for each error bit is counted. The error bit with a number of errors greater than a preset first threshold is called the first error bit, and the erroneous data block is corrected based on the first error bit.

[0026] In combination with the first aspect, in an eighth implementation manner of the first aspect of the present application, correcting an erroneous data block based on the first error bit includes:

[0027] After determining the erroneous data block again, the first error bit corresponding to the erroneous data block is converted to obtain a converted data block, and the converted data block is checked again based on the verification data. If it is correct, there is no need to resend the data block. If it is wrong, a request is made to resend the corresponding data block.

[0028] In a second aspect, the present application provides a multi-type sensor scheduling system based on a petrochemical scenario, the system comprising:

[0029] The collection module deploys various types of sensors in the petrochemical scene, installing the sensors at multiple key locations in the petrochemical scene to collect monitoring data and remaining battery power of the sensors in real time;

[0030] The calculation module periodically calculates the risk deviation of the monitoring data collected by the sensor, the remaining battery percentage of the sensor, and the correlation score between the sensor and the operation plan. It determines the priority of the sensor based on the risk deviation, remaining battery percentage, and correlation score, and assigns different sampling frequencies to different priorities.

[0031] The transmission module processes the monitoring data collected by sensors of different priorities differently according to the priority of the sensors, and sends the monitoring data with high priority first;

[0032] The adjustment module analyzes the monitoring data to determine whether a specific situation exists. If so, it modifies the sampling frequency of the sensor. Specific situations include abnormal events, key nodes in the operation plan, sensor power below the preset minimum threshold, repeated sampling, and data transmission delays.

[0033] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0034] In the technical solution provided by the present application, the priority of the sensor is determined by comprehensively considering the risk deviation, the remaining power ratio and the correlation score, and the sampling frequency is dynamically allocated accordingly, which can ensure the accurate acquisition of petrochemical scene monitoring data while avoiding data redundancy; different sampling frequencies are allocated according to the priority of different sensors, and the sampling frequency of high-priority sensors is higher, and the sampling frequency of low-priority sensors is lower, thereby effectively reducing the overall energy consumption of the sensor and extending the service life of the sensor; avoiding data redundancy caused by fixed sampling frequency, reducing the occupation of network bandwidth and storage space, and improving resource utilization efficiency; adapting to different types of sensors, giving priority to sending high-priority monitoring data, and dynamically adjusting the data sending strategy according to the network status to ensure that key data can be transmitted in a timely and accurate manner; the data receiving end stores data blocks of different priorities in different memory areas, and restores the data in order of priority, which improves the efficiency of data processing, reduces the delay of data processing, and optimizes the resource utilization of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 This is a schematic diagram of an embodiment of a multi-type sensor scheduling method based on a petrochemical scenario in an embodiment of the present application;

[0037] Figure 2 is a schematic diagram of determining priority in an embodiment of the present application;

[0038] Figure 3 This is a schematic diagram of an embodiment of a multi-type sensor scheduling system based on a petrochemical scenario in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The embodiments of the present application provide a multi-type sensor scheduling method and system based on a petrochemical scenario. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application 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 interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "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 that are inherent to these processes, methods, products or devices.

[0040] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of a multi-type sensor scheduling method based on a petrochemical scenario includes:

[0041] Step S1: Deploy various types of sensors at multiple key locations in the petrochemical scene to collect monitoring data and sensor remaining power in real time.

[0042] Specifically, in the petrochemical industry, in order to monitor various parameters in the production process, various types of sensors are deployed at key locations in the petrochemical scene. The key locations include storage tank areas, pipelines, hot work areas, and environmentally sensitive areas such as plant boundaries. They are used to collect monitoring data such as pressure, temperature, flow, and gas during the production process, determine whether there are anomalies based on the monitoring data, prevent dangerous incidents, and collect the remaining power of the sensor.

[0043] Step S2: Periodically calculate the risk deviation of the monitoring data collected by the sensor, calculate the remaining power ratio of the sensor, calculate the correlation score between the sensor and the operation plan, determine the priority of the sensor based on the risk deviation, remaining power ratio and correlation score, and assign different sampling frequencies to different priorities.

[0044] Specifically, in order to reduce energy consumption while ensuring measurement reliability and extend the use time of sensors, the risk deviation, remaining power ratio and correlation score are calculated periodically. The risk deviation refers to the degree of deviation between the monitoring data collected by the sensor and the normal working range. The specific method of calculating the risk deviation will be explained in detail later. The remaining power ratio is the ratio of the remaining power of the sensor to the full power of the sensor. This ratio can intuitively reflect how long the sensor can still work, which is very important for the scheduling of sensors. The correlation score is an indicator of the degree of correlation between the sensor and the operation plan. In the petrochemical scenario, the operation plan includes various stages of the production process, such as raw material input, reaction process and product output. If the sensor is deployed in a location closely related to the current operation plan, such as around the reactor, when the reactor is undergoing a critical chemical reaction. , the correlation scores of these sensors will be very high. Different weights are determined for these three factors. The priority of the sensor is determined based on the weight, risk deviation, remaining power ratio and correlation score. The specific method of determining the priority will be explained in detail later. Different sampling frequencies are assigned to different sensors according to the priority. The sampling frequency of high-priority sensors is higher, for example, sampling once every 30 seconds. The sampling frequency of medium-priority sensors is moderate, such as sampling once every 5 minutes. The sampling frequency of low-priority sensors is lower, such as sampling once every 30 minutes. If the same sampling frequency is set for all sensors, it will cause data redundancy and occupy network and storage resources. The above method is used to dynamically set the sampling frequency for the sensor, while ensuring that the monitoring data of the petrochemical scene can be accurately obtained, reducing energy consumption and reducing the waste of network and storage resources.

[0045] Step S3: Perform cross-language call adaptation and communication protocol adaptation on different types of sensors based on preset adapters. According to the priority of the sensors, the monitoring data collected by sensors of different priorities are processed differently, and high-priority monitoring data is sent first.

[0046] Specifically, since different types of sensors may use different communication protocols and programming languages, it is necessary to design a preset adapter. For example, some temperature sensors may use the Modbus protocol for communication, while gas composition sensors may use the ZigBee protocol. The adapter can convert the data of these different protocols into a unified format, such as JSON format, to facilitate subsequent data processing. For cross-language call adaptation, if the sensor firmware is written in C language and the data processing system is developed based on Python language, the adapter can realize data interaction through Python extension modules of C language (such as Cython), and process the collected monitoring data according to the priority of the sensor. For high-priority sensor data, real-time data cleaning and preliminary analysis can be performed. During the data transmission process, high-priority monitoring data is sent first, and communication bandwidth is allocated first for transmission to ensure that critical data can be sent in a timely manner.

[0047] Step S4: Analyze whether it is in a specific situation. If so, modify the sampling frequency of the sensor. The specific situations include abnormal events, key nodes of the operation plan, sensor power below the preset minimum threshold, and data transmission delay.

[0048] Specifically, if an abnormal event is detected through analysis, such as a gas composition sensor detecting a sharp increase in combustible gas concentration exceeding 50% of the explosion limit, the sensor's sampling frequency will be immediately modified. For sensors involved in this abnormal event, the sampling frequency can be increased to multiple samples per second to more accurately monitor the changing trend of gas concentration. When the operation plan reaches a critical node, such as the raw material switching stage in the oil refining process, the sampling frequency of sensors related to this stage (such as raw material flow sensors and temperature sensors) will be increased. For example, the sampling frequency of the raw material flow sensor can be increased from once per minute to once per second to ensure timely monitoring of flow changes during the raw material switching process and avoid production accidents caused by unstable flow. If the sensor battery level falls below a preset minimum threshold (such as 10%), the system will reduce the sensor's sampling frequency to conserve power and issue an alarm to notify maintenance personnel to replace the battery or charge the sensor in a timely manner. For example, for a wireless temperature sensor, when the battery level falls below 10%, the sampling frequency is reduced from once per hour to once every two hours. In the event of data transmission delays, the system will adjust the sampling frequency based on the degree of delay. If the delay time is short (such as less than 1 second), the original sampling frequency can be temporarily maintained; if the delay time is long (such as more than 5 seconds), the sampling frequency should be appropriately reduced to avoid data backlog. At the same time, the data transmission path can be optimized or the communication bandwidth can be increased to solve the transmission delay problem.

[0049] In a specific embodiment, calculating the risk deviation of monitoring data collected by the sensor includes the following steps:

[0050] Obtain the safety threshold corresponding to the monitoring data, obtain the currently collected real-time monitoring data, divide the absolute value of the result obtained by subtracting the safety threshold from the real-time monitoring data by the safety threshold to obtain a first value, use the prediction model to predict the predicted monitoring data of the monitoring data in the future, divide the absolute value of the result obtained by subtracting the safety threshold from the predicted monitoring data by the safety threshold to obtain a second value, set corresponding weight values ​​for the first value and the second value, multiply the first value and the second value by the corresponding weight values ​​respectively, and then add the obtained result value as the risk deviation.

[0051] Specifically, safety thresholds are determined based on the safe operating standards of equipment and environments in petrochemical scenarios. For example, for a pressure sensor on an oil storage tank, the safety threshold is set based on the tank's design pressure and safe operating procedures. Assuming the design pressure of an oil storage tank is 2 MPa, the safety threshold is set to 1.8 MPa to account for a safety margin. Real-time monitoring data is the data collected by the sensor at the current moment. The first value R1 is calculated using the formula R1=|Pr-Pt| / Pt, where Pr is the real-time monitoring data and Pt is the safety threshold. The first value represents the degree of deviation of the real-time monitoring data from the safety threshold. The prediction model is built based on historical data and machine learning algorithms to predict changes in monitoring data in the future, such as predicting monitoring data 10 minutes in the future. The second value R2 is calculated using the formula R2=|Pp-Pt| / Pt, where Pp is the predicted monitoring data. The second value represents the degree of deviation of the real-time predicted monitoring data from the safety threshold. Weights can be set based on actual needs and experience. For example, if you pay more attention to current real-time data, you can set a higher weight for the first value; if you pay more attention to future trends, you can set a higher weight for the second value. Assuming that the weight of the first value is 0.6 and the weight of the second value is 0.4, then the calculation formula for the risk deviation D is D=0.6×R1+0.4×R2.

[0052] The above method can more comprehensively evaluate the risk deviation of sensor monitoring data by comprehensively considering real-time monitoring data and predicted data. This method not only focuses on the current actual situation, but also considers possible future change trends, and can provide a more accurate risk assessment.

[0053] In a specific embodiment, calculating the correlation score between the sensor and the operation plan includes the following steps:

[0054] For each sensor, obtain the basic correlation value of the current sensor relative to each operation plan, obtain the current operation plan, obtain the duration of the current operation plan and the theoretical maximum duration, divide the duration by the maximum duration and multiply it by the corresponding first correction coefficient and the basic correlation value to obtain a first correction value, obtain the correlation coefficient of each sensor and other sensors, calculate the average of all correlation coefficients for each sensor, multiply the average of the correlation coefficients by the corresponding second correction coefficient and the basic correlation value to obtain a second correction value, and add the basic correlation value, the first correction value, and the second correction value to obtain a correlation score.

[0055] Specifically, the basic correlation value refers to the degree of correlation between the sensor and each operation plan under normal circumstances. It reflects the importance of the sensor in different operation plans. For example, in the oil refining process, a temperature sensor is installed at the inlet of a key reactor. The basic correlation value for the raw material preheating operation plan may be very high because it directly monitors the temperature of the raw materials. The basic correlation value can be determined based on the expert system. The expert system can assign basic correlation values ​​to sensors at different locations based on the process flow knowledge of the petrochemical industry. For example, the process engineer can set the basic correlation value of the reactor inlet temperature sensor to the raw material preheating operation plan to 0.8 based on experience. The first correction value can adjust the basic correlation value based on the ratio of the duration of the current operation plan to the theoretical maximum duration. The first correction coefficient can be set according to the type and importance of the operation plan. For example, for the key raw material preheating operation plan, the first correction coefficient can be set to 1.2, while for the auxiliary equipment cleaning operation plan, the correction coefficient can be set to 0.8. Assuming the basic correlation value is 0.8, the duration of the current operation plan is 5 hours, the theoretical maximum duration is 10 hours, and the first correction coefficient is 1.2, then the calculation formula of the first correction value C1 is C1=0.8×(5 / 10)×1 .2=0.48. The second correction value can adjust the correlation score based on the correlation coefficient between the sensor and other sensors. The correlation coefficient reflects the mutual dependence between sensors. For example, a temperature sensor and a flow sensor may jointly monitor the operating status of a device, and the correlation coefficient between them may be high. The correlation coefficient of each sensor with other sensors is obtained. The correlation coefficient can be determined based on data correlation analysis. For example, the correlation coefficient is determined by calculating the Pearson correlation coefficient between the sensor data. Assuming that the correlation coefficient of the data of two sensors is 0.7, the correlation coefficient can be set to 0.7. The second correction factor is a coefficient used to adjust the average value of the correlation coefficient. The correlation is adjusted according to the synergy between sensors. This coefficient can be set according to the type of sensor and the monitoring parameters, assuming it is 1.1. For each sensor, the average correlation coefficient between it and all other sensors is calculated. Assuming the average correlation coefficient is 0.5, the second correction value C2 of the sensor is C2=0.5×1.1×0.8=0.44. The basic correlation value, the first correction value, and the second correction value are added to obtain the correlation score between the sensor and the operation plan.

[0056] The above method comprehensively considers the basic correlation values, the execution progress of the operation plan, and the synergy between sensors, and can more accurately evaluate the correlation between sensors and operation plans. This method not only considers the direct relationship between sensors and operation plans, but also considers the dynamic changes of operation plans and the mutual influence between sensors. It can better adapt to complex and changing petrochemical scenarios and provide a more comprehensive correlation assessment.

[0057] In a specific embodiment, determining the priority of a sensor based on the risk deviation, the remaining power ratio, and the correlation score includes the following steps:

[0058] The risk deviation, the remaining power ratio and the correlation score are called three data parameters. The three scores of each data parameter relative to other data parameters are obtained. For each data parameter, the sum of all the medians of the three scores is calculated as the first result value. The three scores are divided by the first result value respectively to obtain three ratios. The average of the three ratios is calculated. The average of the three averages is used as the weight value of the corresponding data parameter. The priority of the sensor is determined based on the data parameters and the corresponding weight values.

[0059] Specifically, for each data parameter, obtain its three scores relative to other data parameters. These scores can be determined by expert scoring, historical data analysis, or machine learning models. For example, suppose we have three sensors A, B, and C. We need to score each data parameter of sensor A relative to the corresponding data parameters of sensors B and C. The three data parameters are represented by a, b, and c respectively. The three scores of a relative to a are (1, 1, 1). Suppose the three scores of a to b are (1, 2, 3), and the three scores of a to c are (0.5, 1, 1.5). The maximum value of the three scores represents the maximum importance of the priority judgment relative to the other data parameters, the minimum value of the three scores represents the minimum importance of the priority judgment relative to the other data parameters, and the middle value of the three scores is the maximum value of the priority judgment relative to the other data parameters. Represents the most likely value of the importance of other data parameters to the priority judgment. The first result value of a is 1+2+1=5. Divide the three groups of scores by 5 respectively, and get (1 / 5, 1 / 5, 1 / 5), (1 / 5, 2 / 5, 3 / 5), (1 / 10, 1 / 5, 3 / 10). Calculate the average of the three group ratios to get 1 / 5, 2 / 5, and 1 / 5. The average of the three averages is 4 / 15=0.27. Take 0.27 as the weight value of the corresponding data parameter. Use the same method as above to calculate the weight values ​​of other data parameters. Subsequently, determine the priority of the sensor based on the data parameters and the corresponding weight values.

[0060] This method more accurately assigns weights to each data parameter by comprehensively considering the relative importance of risk deviation, remaining battery percentage, and correlation score. This method considers not only the absolute value of each parameter but also the relative relationship between them, providing a more reasonable weight distribution.

[0061] In a specific embodiment, determining the priority of the sensor based on the data parameters and the corresponding weight values ​​specifically includes the following steps:

[0062] Multiple different priorities are preset. For each priority, a corresponding priority score is set for each data parameter based on the numerical value of each data parameter. The weight value and the priority score corresponding to each priority are multiplied and then added to obtain the priority score of the corresponding priority. All priority scores are obtained, and the priority corresponding to the maximum priority score is used as the priority of the corresponding sensor.

[0063] Specifically, multiple priority levels are preset, such as high, medium-high, medium, and low. Each priority level corresponds to a different processing strategy and resource allocation. For example, high-priority sensors receive a higher sampling frequency and prioritized data transmission. For each priority level, a corresponding priority score is assigned to each data parameter. The priority score indicates the likelihood of the corresponding priority being assigned based on only one data parameter. For example, if the risk deviation of three data parameters is 0.3, the remaining battery percentage is 0.8, and the correlation score is 0.5, then the probability of being classified as high priority is 0.3, and the probability of being classified as medium-high priority is 0.0 if only the risk deviation is considered. 5, the probability of being classified as medium priority is 0.6, and the probability of being classified as low priority is 0.7. Considering only the remaining power ratio, the probability of being classified as high priority is 0.1, the probability of being classified as medium-high priority is 0.2, the probability of being classified as medium priority is 0.3, and the probability of being classified as low priority is 0.7. Considering only the correlation score, the probability of being classified as high priority is 0.3, the probability of being classified as medium-high priority is 0.4, the probability of being classified as medium priority is 0.6, and the probability of being classified as low priority is 0.4. Assuming that the weight values ​​of the three data parameters are 0.27, 0.14, and 0.56 respectively, as shown in Figure 2 As shown in the figure, it is a schematic diagram for determining the priority. The weight value and the priority score of each priority are multiplied and then added to obtain the priority score of the corresponding priority. Corresponding to 4 priorities, a total of 4 priority scores are obtained, which are 0.263, 0.387, 0.54 and 0.511 respectively. The largest priority score is 0.54, and the priority corresponding to the sensor is determined to be the middle priority.

[0064] The above method can more accurately assign priorities to sensors by comprehensively considering data parameters and corresponding weight values. This method not only considers the absolute value of each parameter, but also their relative importance, thereby providing a more reasonable priority allocation. It can also dynamically adjust priorities according to different scenarios and needs. For example, in some cases, risk deviation may be more important, while in other cases, the remaining power ratio or correlation score may be more critical. By calculating weight values ​​and priority scores, it can flexibly adapt to these changes. By reasonably allocating priorities, it can also ensure that each sensor can receive appropriate attention in the comprehensive evaluation of multiple parameters.

[0065] In a specific embodiment, monitoring data collected by sensors of different priorities are processed differently, specifically including the following steps:

[0066] The monitoring data is divided into multiple data blocks, and the corresponding sensor priority identifier is added to each data block. The corresponding verification data is generated based on the data block, and the network status is obtained in real time. When the network status is good, all data blocks, priority identifiers and verification data are sent to the data receiving end at the same time. When the network status is not good, the data blocks corresponding to the high priority are sent first, and then the data blocks corresponding to the low priority are sent.

[0067] Specifically, the monitoring data collected by the sensor is divided into multiple data blocks. The size of the data block can be set according to actual needs and network conditions. For example, each data block can contain monitoring data within a certain time interval, such as data per minute as a data block. A priority identifier is added to each data block. The identifier indicates the priority of the sensor to which the data block belongs. It is calculated by the previous method. The priority identifier can be a simple number or label, such as high priority (1), medium-high priority (2), medium priority (3), and low priority (4). The check data is used to verify the integrity and accuracy of the data block. Common check methods include CRC (cyclic redundancy check), MD5 hash, etc. The network status includes parameters such as network bandwidth, delay, and packet loss rate. These parameters can reflect the current network transmission capacity. The network status can be obtained in real time through network status monitoring tools or APIs. If the network status is good (such as sufficient bandwidth, low delay, and low packet loss rate), all data blocks, check data, and priority identifiers can be sent to the data receiving end at the same time. If the network status is not good (such as limited bandwidth, high delay, and high packet loss rate), the data blocks, check data, and corresponding priority identifiers are sent in descending order of priority.

[0068] The above method realizes hierarchical management and priority sorting of data by dividing the monitoring data into multiple data blocks and adding a priority identifier to each data block. This method can better adapt to complex network environments and ensure the priority transmission of critical data. Combined with real-time network status monitoring, it dynamically adjusts the data sending strategy. This method not only improves the reliability of data transmission, but also optimizes the utilization of network resources.

[0069] In a specific embodiment, monitoring data collected by sensors of different priorities are processed differently, specifically including the following steps:

[0070] After receiving the data block, the data receiving end stores the data blocks corresponding to different priority identifiers in different memory areas respectively, and combines the data blocks in the memory areas in descending order of priority to restore them to the original monitoring data.

[0071] Specifically, after the data receiving end receives the data block, by storing data blocks of different priorities in different memory areas, it can quickly locate and process high-priority data, improve the efficiency of data processing, ensure that critical data can be processed first, and reduce data processing delays, while low-priority data blocks can be stored in relatively low-priority memory areas to save system resources. Restoring data in combination according to priority order can ensure data integrity. Even if some low-priority data blocks are lost or delayed, high-priority data can still be processed and restored first, ensuring data availability. By reasonably allocating memory areas, the system's resource utilization can be optimized. High-priority data blocks are stored in high-priority memory areas to ensure that these data can be quickly accessed and processed.

[0072] In a specific embodiment, before combining and restoring the data blocks, the following steps are further performed:

[0073] The data receiving end performs error checking on the received data blocks. If an error occurs, it records a request to resend the corresponding data block, compares the erroneous data block with the correct data block, and records the corresponding error bits. After several comparisons, the number of errors for each error bit is counted. The error bit with a number of errors greater than a preset first threshold is called the first error bit, and the erroneous data block is corrected based on the first error bit.

[0074] Specifically, the data receiving end verifies the data block based on the received data block and the verification data, and requests to resend the corresponding data block if an error occurs. By comparing the erroneous data block with the correct data block, the erroneous dislocation is recorded. After several comparisons, for example, 100 comparisons, the number of errors of each error bit is counted. The error bit with the number of errors greater than a preset first threshold is called the first error bit. The erroneous data block can be corrected subsequently based on the first error bit.

[0075] In a specific embodiment, correcting an erroneous data block based on a first error bit specifically includes the following steps:

[0076] After determining the erroneous data block again, the first error bit corresponding to the erroneous data block is converted to obtain a converted data block, and the converted data block is checked again based on the verification data. If it is correct, there is no need to resend the data block. If it is wrong, a request is made to resend the corresponding data block.

[0077] Specifically, assuming that the data block is 16-bit data, statistics show that the 9th bit has the most errors, accounting for 80%. When judging the data block with errors again, the 9th bit data is converted. For example, if the 9th bit of the received data block is 1, the 9th bit of the converted data block is 0. Then the converted back-end data block is checked again. If it is correct, there is no need to send a request to the sensor to resend the data block. If it is wrong, it means that the number of bits with errors this time may not be the 9th bit, or more than the 9th bit. Therefore, the sensor is requested to resend the corresponding data block.

[0078] The above describes the multi-type sensor scheduling method based on the petrochemical scene in the embodiment of the present application. The following describes the multi-type sensor scheduling system based on the petrochemical scene in the embodiment of the present application. Figure 3 In the embodiment of the present application, an embodiment of a multi-type sensor scheduling system based on a petrochemical scenario includes:

[0079] The collection module is used to deploy various types of sensors at multiple key locations in the petrochemical scene to collect monitoring data and sensor remaining power in real time;

[0080] The calculation module is used to periodically calculate the risk deviation of the monitoring data collected by the sensor, calculate the remaining battery percentage of the sensor, calculate the correlation score between the sensor and the operation plan, determine the priority of the sensor based on the risk deviation, remaining battery percentage and correlation score, and assign different sampling frequencies to different priorities;

[0081] The transmission module is used to perform cross-language call adaptation and communication protocol adaptation for different types of sensors based on preset adapters. According to the priority of the sensors, the monitoring data collected by sensors with different priorities are processed differently, and high-priority monitoring data is sent first.

[0082] The adjustment module analyzes whether it is in a specific situation. If so, it modifies the sampling frequency of the sensor. The specific situations include abnormal events, key nodes of the operation plan, sensor power below the preset minimum threshold, and data transmission delay.

[0083] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0084] 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 application, 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. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0085] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-type sensor scheduling method based on a petrochemical scenario is characterized by: The method comprises: Step S1: deploy various types of sensors at multiple key locations in the petrochemical scene to collect monitoring data and sensor remaining power in real time; Step S2: Periodically calculate the risk deviation of the monitoring data collected by the sensor, calculate the remaining power ratio of the sensor, and calculate the correlation score between the sensor and the operation plan. Determine the priority of the sensor based on the risk deviation, remaining power ratio, and correlation score, and assign different sampling frequencies to different priorities. The risk deviation refers to the degree of deviation of the monitoring data collected by the sensor from the normal operating range, the remaining power ratio is the ratio of the remaining power of the sensor to the full power of the sensor, and the correlation score is an indicator of the degree of correlation between the sensor and the operation plan. Step S3: Perform cross-language call adaptation and communication protocol adaptation for different types of sensors based on preset adapters. According to the priority of the sensors, the monitoring data collected by sensors with different priorities are processed differently, and the monitoring data with high priority is sent first. Step S4: Analyze whether a specific situation exists. If so, modify the sampling frequency of the sensor. Specific situations include abnormal events, key nodes in the operation plan, sensor power below a preset minimum threshold, and data transmission delays. In step S3, the monitoring data collected by sensors of different priorities are processed differently, including: dividing the monitoring data into multiple data blocks, adding a corresponding sensor priority identifier to each data block, generating corresponding verification data based on the data block, obtaining the network status in real time, and when the network status is good, sending all data blocks, priority identifiers and verification data to the data receiving end at the same time; when the network status is not good, sending the data blocks corresponding to the high priority after sending the data blocks corresponding to the low priority.

2. The method according to claim 1, characterized in that Calculate the risk deviation of monitoring data collected by sensors, including: Obtain the safety threshold corresponding to the monitoring data, obtain the currently collected real-time monitoring data, divide the absolute value of the result obtained by subtracting the safety threshold from the real-time monitoring data by the safety threshold to obtain a first value, use the prediction model to predict the predicted monitoring data of the monitoring data in the future, divide the absolute value of the result obtained by subtracting the safety threshold from the predicted monitoring data by the safety threshold to obtain a second value, set corresponding weight values ​​for the first value and the second value, multiply the first value and the second value by the corresponding weight values ​​respectively, and then add the obtained result value as the risk deviation.

3. The method according to claim 1, characterized in that Calculates the relevance score of the sensor to the job plan, including: For each sensor, obtain the basic correlation value of the current sensor relative to each work plan, obtain the current work plan, obtain the duration of the current work plan and the theoretical maximum duration, divide the duration by the maximum duration and multiply it by the corresponding first correction coefficient and the basic correlation value to obtain the first correction value, obtain the correlation coefficient of each sensor and other sensors, calculate the average value of all correlation coefficients for each sensor, multiply the average value of the correlation coefficient by the corresponding second correction coefficient and the basic correlation value to obtain the second correction value, and add the basic correlation value to the first correction value and the second correction value to obtain the correlation score, wherein the first correction coefficient is set according to the type and importance of the work plan, and the second correction coefficient is set according to the type of sensor and monitoring parameters.

4. The method according to claim 1, wherein Sensor priorities are determined based on risk deviation, remaining battery percentage, and relevance scores, including: The risk deviation, the remaining power ratio and the correlation score are called three data parameters. The three scores of each data parameter relative to other data parameters are obtained. For each data parameter, the sum of all the medians of the three scores is calculated as the first result value. The three scores are divided by the first result value respectively to obtain three ratios. The average of the three ratios is calculated. The average of the three averages is used as the weight value of the corresponding data parameter. The priority of the sensor is determined based on the data parameters and the corresponding weight values.

5. The method according to claim 4, characterized in that Determine the priority of the sensor based on the data parameters and the corresponding weight values, including: Multiple different priorities are preset. For each priority, a corresponding priority score is set for each data parameter based on the numerical value of each data parameter. The weight value and the priority score corresponding to each priority are multiplied and then added to obtain the priority score of the corresponding priority. All priority scores are obtained, and the priority corresponding to the maximum priority score is used as the priority of the corresponding sensor.

6. The method according to claim 1, wherein The monitoring data collected by sensors of different priorities are processed differently, including: After receiving the data block, the data receiving end stores the data blocks corresponding to different priority identifiers in different memory areas respectively, and combines the data blocks in the memory areas in descending order of priority to restore them to the original monitoring data.

7. The method according to claim 6, characterized in that Before restoring the data blocks, execute the following command: The data receiving end performs error checking on the received data blocks. If an error occurs, it records a request to resend the corresponding data block, compares the erroneous data block with the correct data block, and records the corresponding error bits. After several comparisons, the number of errors for each error bit is counted. The error bit with a number of errors greater than a preset first threshold is called the first error bit, and the erroneous data block is corrected based on the first error bit.

8. The method according to claim 7, characterized in that Correcting an erroneous data block based on a first error bit, comprising: After determining the erroneous data block again, the first error bit corresponding to the erroneous data block is converted into binary to obtain a converted data block. The converted data block is checked again based on the verification data. If it is correct, there is no need to resend the data block. If it is wrong, a request is made to resend the corresponding data block.

9. A multi-type sensor scheduling system based on a petrochemical scenario, used to implement a multi-type sensor scheduling method based on a petrochemical scenario according to any one of claims 1 to 8, characterized in that: The system comprises: The collection module deploys various types of sensors in the petrochemical scene, installing the sensors at multiple key locations in the petrochemical scene to collect monitoring data and remaining battery power of the sensors in real time; The calculation module periodically calculates the risk deviation of the monitoring data collected by the sensor, the remaining battery percentage of the sensor, and the correlation score between the sensor and the operation plan. The sensor priority is determined based on the risk deviation, remaining battery percentage, and correlation score, and different sampling frequencies are assigned to different priorities. The risk deviation refers to the degree of deviation of the monitoring data collected by the sensor from the normal operating range, the remaining battery percentage is the ratio of the remaining battery of the sensor to the full battery of the sensor, and the correlation score is an indicator of the degree of relevance between the sensor and the operation plan. The transmission module processes the monitoring data collected by sensors of different priorities differently according to the priority of the sensors, and sends the monitoring data with high priority first; The adjustment module analyzes monitoring data to determine if specific conditions exist and, if so, modifies the sensor sampling frequency. Specific conditions include abnormal events, critical nodes in the operation plan, sensor battery levels below a preset minimum threshold, repeated sampling, and data transmission delays. In the transmission module, the monitoring data collected by sensors of different priorities are processed differently, including: dividing the monitoring data into multiple data blocks, adding a corresponding sensor priority identifier to each data block, generating corresponding verification data based on the data block, and obtaining the network status in real time. When the network status is good, all data blocks, priority identifiers and verification data are sent to the data receiving end at the same time. When the network status is not good, the data blocks corresponding to the high priority are sent before the data blocks corresponding to the low priority are sent.

Images