Visual programming tool system based on artificial intelligence and Internet of Things

Abstract

The invention discloses a visual programming tool system based on artificial intelligence and Internet of Things, and relates to the technical field of programming tools. According to the method, when a data processing link is entered, the collected data volume of the Internet of Things equipment and the state information of each processing node are comprehensively evaluated, an optimal processing node is selected for data processing of each Internet of Things equipment, a pre-constructed calculation task is used for running on each node, and a calculation capability score is determined according to the completion time; the network bandwidth is calculated according to the minimum value of the network access bandwidth and the network exit bandwidth of each node, and a bandwidth evaluation value is obtained; according to the method, the number of tasks to be processed and the preset interval matching load additional coefficient are counted, the load evaluation index is calculated by combining the CPU and the memory utilization rate, the calculation capability score, the bandwidth evaluation value and the load evaluation index of each node are subjected to normalization processing, a formula is substituted to calculate the comprehensive evaluation index, and the defects in the aspect of data scheduling in the prior art are overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of programming tools, and particularly to a visual programming tool system based on artificial intelligence and the Internet of Things. Background Art

[0002] With the rapid development of technology, the integrated application of artificial intelligence and the Internet of Things is penetrating into various fields at an unprecedented speed. In the field of smart home, indoor environmental data is collected through Internet of Things devices, and intelligent regulation is achieved with the help of artificial intelligence algorithms to improve living comfort. In industrial production, the combination of the two helps enterprises build smart factories, optimize production processes, improve production efficiency, and reduce costs.

[0003] However, the existing visual programming tool system based on artificial intelligence and the Internet of Things still has the following deficiencies in the actual application process:

[0004] In terms of data scheduling, it is unable to intelligently allocate tasks according to the data volume collected by devices and the status of processing nodes, often resulting in some nodes being overloaded while some nodes are idle, reducing the overall system performance.

[0005] In addition, in terms of data security, the encryption protection of the data collected by Internet of Things devices and user programming logic is insufficient, making it difficult to cope with the increasingly severe network security threats and unable to ensure the confidentiality, integrity, and availability of data.

[0006] In summary, the existing programming tools are difficult to meet the diverse and complex needs of the integrated application of artificial intelligence and the Internet of Things. Therefore, a visual programming tool system based on artificial intelligence and the Internet of Things is introduced. Summary of the Invention

[0007] The purpose of the present invention is to solve the problems pointed out in the background art, and to provide a visual programming tool system based on artificial intelligence and the Internet of Things.

[0008] The purpose of the present invention can be achieved through the following technical solutions: A visual programming tool system based on artificial intelligence and the Internet of Things, comprising:

[0009] A programming tool module: including an initialization unit, an Internet of Things unit, an artificial intelligence unit, a logic unit, and an execution unit;

[0010] The initialization unit is used to connect Internet of Things devices to the corresponding network and load the pre-constructed visual programming interface at the same time;

[0011] The Internet of Things unit is used to trigger device scanning signaling, search for available Internet of Things devices in the network; identify the scanned devices, classify them according to the type and model of the devices, and register the device information in the database;

[0012] The artificial intelligence unit is used to pre-integrate various artificial intelligence algorithms;

[0013] The logic unit is used for the user to select the required programming element nodes in the node library of the visual programming interface, including sensor nodes, artificial intelligence algorithm nodes, and controller nodes, drag them to the programming area, and use connection lines to connect the nodes after dragging to form a complete programming logic;

[0014] The execution unit is used to automatically generate corresponding code according to the programming logic created by the user in the visual programming interface, deploy the generated code to the target Internet of Things device. After the target Internet of Things device receives the deployed code, it starts to execute the program, and performs data collection, processing, and control operations on the Internet of Things device according to the programming logic;

[0015] The data scheduling module: After collecting data from the Internet of Things devices and entering the data processing link, comprehensively evaluate the collected data volume of the Internet of Things devices and the status information of each processing node. Based on the results of the comprehensive evaluation, select the best processing node for the data processing of each Internet of Things device. After the selection is completed, fill the selected processing nodes corresponding to each Internet of Things device into the pre-constructed chart template to generate a visual node chart; the status information includes network bandwidth, computing power score, and load condition.

[0016] As a preferred embodiment of the present invention, the comprehensive evaluation of the collected data volume of the Internet of Things devices and the status information of each processing node is specifically as follows:

[0017] S1: Evaluate the data volume collected by the Internet of Things devices, set the threshold volume corresponding to the data volume. If the data volume of a certain Internet of Things device is lower than the threshold volume, directly execute step S2; if the data volume of a certain Internet of Things device is higher than the threshold volume, execute step S3.

[0018] As a preferred embodiment of the present invention, the specific process of executing step S2 is as follows:

[0019] Use the pre-constructed computing tasks to run on each node and record their completion times; set the intervals where each group of durations corresponding to the completion times are located, and each interval where the group of durations is located corresponds to a computing power score. Match the completion times of each node with the intervals where each group of durations are located to determine the computing power score a1 of each node;

[0020] Calculate the network bandwidth with the minimum value of the incoming and outgoing network bandwidths of each node, and use the calculated network bandwidth as the bandwidth evaluation value a2 of each node;

[0021] Count the number of tasks to be processed at the current time point for each node, preset the intervals where the groups of quantities corresponding to the number of tasks to be processed are located, and each interval where the quantity is located corresponds to a load addition coefficient; match the number of tasks to be processed at the current time point for each node with the intervals where the groups of quantities are located to determine the load addition coefficient for each node at the current time point.

[0022] Obtain the CPU usage rate and memory usage rate of each node at the current time point, multiply them by the corresponding set weight coefficients and then sum, and multiply the sum value by the load addition coefficient to obtain the load evaluation index a3 for each node.

[0023] Normalize the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point and then substitute them into the formula Perform weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; a1 及格 、a2 参考 、a3 阈值 Are respectively the preset passing computing power score, passing bandwidth, and load threshold index; λ1, λ2, λ3 are the influence weight factors of the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 in step S2.

[0024] As a preferred implementation manner of the present invention, the specific process of executing step S3 is:

[0025] If the data volume of a certain Internet of Things device is higher than the threshold volume, normalize the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point and then substitute them into the formula Perform weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; μ1, μ2, μ3 are the influence weight factors of the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 in step S3.

[0026] As a preferred implementation manner of the present invention, based on the comprehensive evaluation result, select the best processing node for the data processing of each Internet of Things device, specifically:

[0027] According to the comprehensive evaluation index GTR of each node calculated in steps S2 and S3, select the node with the GTR of the evaluation index as the processing node.

[0028] As a preferred implementation manner of the present invention, it further includes:

[0029] The security algorithm module is used to encrypt and transmit the data collected by the Internet of Things device and the programming logic of the user using a preset encryption algorithm.

[0030] As a preferred embodiment of the present invention, the data collected by the Internet of Things device and the user's programming logic are encrypted and transmitted by using a preset encryption algorithm, specifically as follows:

[0031] M1: Taking the time point required for the transmission of the Internet of Things device data as the key acquisition time point, randomly extracting X groups of time numbers from within the key acquisition time point as the time key sequence;

[0032] M2: Extracting the model corresponding to the Internet of Things device data, and randomly extracting X groups of letters from the model of the Internet of Things device as the number key sequence;

[0033] M3: After randomly shuffling and overlapping the time key sequence and the number key sequence, obtaining the encryption sequence of the Internet of Things device data at the current time point;

[0034] M4: Using a preset conversion rule to convert the letters in the encryption sequence into numbers, that is, setting different letters to correspond to a number respectively; after the conversion is completed, taking two sets of numerical values in sequence from left to right to construct coordinate points, using the left numerical value in the two sets of numerical values as the abscissa and the right numerical value as the abscissa. If the number of numbers in the time sequence is odd, randomly extract a group of numbers for elimination, draw each group of constructed coordinate points in a rectangular coordinate system, and after the drawing is completed, connect each group of coordinate points in the construction order of the coordinate points to obtain a polygon. Taking the obtained polygon as the encryption graph of the data collected by the Internet of Things device and the user's programming logic and performing encryption.

[0035] As a preferred embodiment of the present invention, according to the set update time interval, after reaching the set update time interval, randomly shuffle and adjust the conversion rule, that is, adjust the numbers corresponding to different letters.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] When the Internet of Things device completes data collection and enters the data processing link, the present invention comprehensively evaluates the data collection amount of the Internet of Things device and the status information of each processing node, including network bandwidth, computing power score, and load condition, selects the best processing node for the data processing of each Internet of Things device, runs the pre-constructed computing tasks on each node, and determines the computing power score according to the completion time; calculates the network bandwidth by taking the minimum value of the incoming and outgoing network bandwidths of each node to obtain a bandwidth evaluation value; counts the number of tasks to be processed, preset an interval to match the load additional coefficient, and calculate the load evaluation index in combination with the CPU and memory usage rates. Normalize the computing power score, bandwidth evaluation value, and load evaluation index of each node, substitute them into the formula to calculate the comprehensive evaluation index, and solve the deficiencies in data scheduling of the prior art;

[0038] The present invention evaluates the amount of data collected by Internet of Things devices, sets a threshold amount, and adjusts the calculation process of the comprehensive evaluation index according to the comparison result, so that the calculation result is more in line with the current requirements;

[0039] The present invention takes the time point required for transmitting Internet of Things device data as the key acquisition time point, and randomly extracts X groups of time numbers from it as the time key sequence; extracts the model corresponding to the Internet of Things device data, and randomly extracts X groups of letters from the model as the number key sequence. The generation of these two key sequences is random, increasing the unpredictability of the key;

[0040] The present invention constructs a complex encryption sequence: randomly shuffles and combines the time key sequence and the number key sequence to obtain an encryption sequence, making the encryption sequence more complex, making it difficult for attackers to guess and restore the complete encryption sequence, and using a preset conversion rule to convert the letters in the encryption sequence into numbers, and sequentially taking two sets of numerical values from left to right to construct coordinate points, and drawing a polygon in a rectangular coordinate system as the encryption graph for encryption. This encryption form is different from the traditional method and increases the difficulty of cracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the accompanying drawings.

[0042] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0043] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0044] Please refer to Figure 1 As shown, a visualization programming tool system based on artificial intelligence and the Internet of Things includes a programming tool module, a data scheduling module, and a security algorithm module;

[0045] The programming tool module includes an initialization unit, an Internet of Things unit, an artificial intelligence unit, a logic unit, and an execution unit;

[0046] The initialization unit is used to connect Internet of Things devices (such as sensors, actuators) to the corresponding network, ensure that the devices are normally powered on and in a communicable state; at the same time, load the pre-constructed visualization programming interface; including operations such as database initialization and algorithm library loading;

[0047] The Internet of Things unit is used to trigger device scanning signaling, search for available Internet of Things devices in the network through supported communication protocols (such as MQTT, CoAP); identify the scanned devices, classify them according to the device type and model, and register the device information in the database; for example, if a temperature sensor is scanned, the system will identify it as a temperature sensor type and record its device ID, communication address, etc.

[0048] The artificial intelligence unit is used to pre-integrate various artificial intelligence algorithms; including but not limited to machine learning algorithms (decision tree, support vector machine, etc.) and deep learning algorithms (convolutional neural network, recurrent neural network, etc.), and each algorithm has a corresponding visual parameter configuration interface for the convenience of users to set.

[0049] The logic unit is used for users to select the required programming element nodes in the node library of the visual programming interface, including sensor nodes, artificial intelligence algorithm nodes, and controller nodes, such as sensor nodes (temperature sensor, humidity sensor, etc.), artificial intelligence algorithm nodes (image recognition algorithm, data prediction algorithm, etc.), and controller nodes (motor control, light control, etc.), and drag them to the programming area. After dragging, use connection lines to connect the various nodes to form a complete programming logic.

[0050] For example, connect the temperature sensor node to the data processing algorithm node, and then connect the output result of the algorithm node to the air conditioner control actuator node to realize the function of automatically controlling the air conditioner according to temperature data.

[0051] For each node, users can configure parameters through the property setting interface. For example, for artificial intelligence algorithm nodes, users can set training parameters, input and output formats, etc. For sensor nodes, users can set data acquisition frequency, range, etc.

[0052] The execution unit is used to automatically generate corresponding code according to the programming logic created by the user in the visual programming interface. The code generation process will be optimized according to the requirements of the target device and programming language to ensure that the generated code can run properly on the target device. For example, if the target device is a Python-based development board, the system will generate Python code; Deploy the generated code to the target Internet of Things device. For some devices that support remote deployment, the system can send the code to the device through the network; For some devices that require local programming, the user can download the generated code locally and then use the corresponding programming tool to burn the code into the device; After the target Internet of Things device receives the deployed code, it starts to execute the program and performs data collection, processing, and control operations on the Internet of Things device according to the programming logic; For example, a temperature sensor collects temperature data in real time, transmits the data to an artificial intelligence algorithm node for analysis, and controls the switch and temperature adjustment of the air conditioner according to the analysis results.

[0053] It should be noted that the data collected by the Internet of Things device and the results processed by the artificial intelligence algorithm are stored in real time. Users can query and analyze the data through the visual programming interface, and the system will display the data in the form of charts, reports, etc. to help users understand the operation status and performance of the system;

[0054] The data scheduling module is used to comprehensively evaluate the amount of data collected by the Internet of Things device and the status information of each processing node when entering the data processing link after collecting data from the Internet of Things device. Based on the results of the comprehensive evaluation, it selects the best processing node for the data processing of each Internet of Things device. After the selection is completed, it fills the selected processing nodes corresponding to each Internet of Things device into a pre-constructed chart template to generate a visual node chart; The status information includes network bandwidth, computing power score, and load condition;

[0055] Specifically:

[0056] S1: Evaluate the amount of data collected by the Internet of Things device, set the threshold amount corresponding to the data amount. If the data amount of a certain Internet of Things device is lower than the threshold amount, directly execute step S2. If the data amount of a certain Internet of Things device is higher than the threshold amount, execute step S3;

[0057] S2: Use the pre-constructed computing tasks to run on each node and record their completion times; Such as matrix multiplication and image filtering, etc.; Set the intervals where each group of durations corresponding to the completion times are located. Each interval where the group of durations is located corresponds to a computing power score. Match the completion times of each node with the intervals where each group of durations are located to determine the computing power score a1 of each node; The shorter the completion time, the higher the corresponding computing power score obtained by matching. The computing power score range is set from 1 to 10;

[0058] Calculate the network bandwidth based on the minimum of the incoming and outgoing bandwidths of each node, and use the calculated network bandwidth as the bandwidth evaluation value a2 of each node; for example, if the incoming bandwidth of node E is 100Mbps and the outgoing bandwidth is 200Mbps, then its network bandwidth quantization value is 100Mbps;

[0059] Count the number of tasks to be processed at the current time point for each node, preset the intervals of each group corresponding to the number of tasks to be processed, and each interval of the number of groups corresponds to a load additional coefficient; the value range of the load additional coefficient is set between 1.019 - 1.187. The fewer the number of tasks to be processed, the lower the corresponding load additional coefficient obtained. When the number of tasks to be processed is 0, the value is 1.019; match the number of tasks to be processed at the current time point for each node with the intervals of each group to determine the load additional coefficient of each node at the current time point;

[0060] Obtain the CPU usage rate and memory usage rate of each node at the current time point, multiply them by the corresponding set weight coefficients and then sum, and multiply the sum value by the load additional coefficient to obtain the load evaluation index a3 of each node;

[0061] Normalize the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point and then substitute them into the formula Perform weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; a1 及格 、a2 参考 、a3 阈值 Are respectively the preset passing computing power score, passing bandwidth, and load threshold index; λ1, λ2, λ3 are the influence weight factors of the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 in step S2;

[0062] S3: If the data volume of a certain Internet of Things device is higher than the threshold volume, then normalize the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point and then substitute them into the formula Perform weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; μ1, μ2, μ3 are the influence weight factors of the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 in step S3; at this time, μ1 > μ2, μ3, and the specific values are set by technical personnel;

[0063] For tasks with a large data volume, the weight of computing power is higher;

[0064] S4: According to the comprehensive evaluation index GTR of each node calculated in steps S2 and S3, select the node with the evaluation index GTR as the processing node;

[0065] It should be noted that if there is a situation where the selection of the S2 and S3 processing nodes is repeated, the IoT devices in the S3 step will be preferentially selected. After selection, the selected nodes will be removed, and then the IoT devices in the S2 step will make a selection;

[0066] By carefully evaluating the data volume collected by IoT devices and combining the computing power, network bandwidth, and load conditions of each node, the best processing node is accurately matched for tasks with different data volumes; for tasks with a small data volume, the S2 process is executed, and appropriate nodes can be quickly screened under general data processing requirements; while for tasks with a large data volume, the S3 process is executed, focusing on factors with a higher computing power weight, and nodes that can efficiently process large data volumes are found to avoid task lag or long waiting due to insufficient node processing capabilities;

[0067] Adopting different processing processes and weight settings for tasks with different data volumes enables the system to flexibly adapt to the diverse data processing requirements in the IoT environment; as the number of IoT devices increases and the data type and data volume change continuously, the system can continuously optimize the task allocation strategy by adjusting parameters such as the threshold value and weight factor without large-scale changes to the system architecture, and has good scalability; generate a visual node chart to intuitively display the corresponding relationship between IoT devices and processing nodes;

[0068] The security algorithm module is used to encrypt and transmit the data collected by IoT devices and the programming logic of users using a preset encryption algorithm;

[0069] Specifically:

[0070] M1: Taking the time point required for the transmission of IoT device data as the key acquisition time point, randomly extracting X groups of time numbers from within the key acquisition time point as the time key sequence; where the value range of X is set between 3 and 5 and is a positive integer;

[0071] For example, if the key acquisition time point is 10:23 on May 13, 2023, then randomly extract X groups of time numbers from 20235131023 as the time key sequence;

[0072] M2: Extracting the model corresponding to the IoT device data, and randomly extracting X groups of letters from the model of the IoT device as the number key sequence;

[0073] For example, if the model corresponding to the IoT device data is Temp-S300, then randomly extract X groups of letters from TempS as the number key sequence;

[0074] M3: After randomly shuffling and overlapping the time key sequence and the number key sequence, obtain the encryption sequence of the IoT device data at the current time point;

[0075] M4: Use the preset conversion rules to convert the letters in the encrypted sequence into numbers, that is, set different letters to correspond to a number, and all of them are positive integers; for example, A=5, b=2; after the conversion is completed, two groups of values are taken from left to right to construct coordinate points, and the left value of the two groups of values is used as the horizontal coordinate, and the right value is used as the horizontal coordinate. If the number of numbers in the time series is an odd number, a group of numbers is randomly selected for elimination, and each group of constructed coordinate points is drawn in the rectangular coordinate system. After the drawing is completed, the construction order of the coordinate points is connected to each group of coordinate points to obtain a polygonal figure, and the obtained polygonal figure is used as the encryption figure of the IoT device collection data and the user programming logic and encrypted;

[0076] M5: According to the set update time interval, after reaching the set update time interval, the conversion rules are randomly adjusted, that is, the numbers corresponding to different letters are adjusted;

[0077] The random time key sequence of key generation is generated based on the time point when the IoT device data needs to be transmitted, and the time is constantly changing and random; X groups of time numbers are randomly extracted from the time point numbers, so that the time key sequence generated each time is different;

[0078] The numbered key sequence is generated by randomly extracting letters from the model of the IoT device. Different device models have different letters, and the extraction process is random. The combination of these two key sequences further increases the randomness, making it difficult for attackers to predict and crack the encryption key.

[0079] The complexity of the encryption sequence randomly shuffles and overlaps the time key sequence and the number key sequence, making the encryption sequence more complex and difficult to guess; even if the attacker obtains part of the time or device model information, it is difficult to restore the complete encryption sequence;

[0080] The uniqueness of the encrypted graph converts the encrypted sequence into a polygonal graph for encryption, which is different from the traditional encryption form based on characters or numbers; the construction of the polygonal graph depends on the coordinate points, which are converted from the encrypted sequence, making the encryption form more diverse and unique, and increasing the difficulty of cracking;

[0081] The dynamic update mechanism randomly adjusts the conversion rules according to the set update time interval, that is, changes the numbers corresponding to different letters; this makes the encryption method dynamic. Even if an attacker cracks the encryption rules within a certain period of time, the previous cracking method will become invalid as the conversion rules are updated, further ensuring the long-term security of the data.

[0082] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only the specific embodiments. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A visual programming tool system based on artificial intelligence and the Internet of Things, characterized in that Including: Programming tool module: including an initialization unit, an Internet of Things unit, an artificial intelligence unit, a logic unit, and an execution unit; The initialization unit is used to connect the Internet of Things device to the corresponding network and load the pre-built visual programming interface at the same time; The Internet of Things unit is used to trigger the device scanning signaling to search for available Internet of Things devices in the network; identify the scanned devices, classify them according to the type and model of the devices, and register the device information in the database; The artificial intelligence unit is used to pre-integrate various artificial intelligence algorithms; The logic unit is used for the user to select the required programming element nodes in the node library of the visual programming interface, including sensor nodes, artificial intelligence algorithm nodes, and controller nodes, and drag them to the programming area. After dragging, use connection lines to connect each node to form a complete programming logic; The execution unit is used to automatically generate corresponding code according to the programming logic created by the user in the visual programming interface, deploy the generated code to the target Internet of Things device. After the target Internet of Things device receives the deployed code, it starts to execute the program, and performs data collection, processing, and control operations on the Internet of Things device according to the programming logic; Data scheduling module: After collecting data from the Internet of Things device and entering the data processing link, comprehensively evaluate the amount of collected data of the Internet of Things device and the status information of each processing node. Based on the results of the comprehensive evaluation, select the best processing node for the data processing of each Internet of Things device. After the selection is completed, fill the selected processing nodes corresponding to each Internet of Things device into the pre-built chart template to generate a visual node chart; the status information includes network bandwidth, computing power score, and load condition.

2. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 1, characterized in that The comprehensive evaluation of the amount of collected data of the Internet of Things device and the status information of each processing node is specifically as follows: S1: Evaluate the amount of data collected by the Internet of Things device, set the threshold amount corresponding to the amount of data. If the amount of data of a certain Internet of Things device is lower than the threshold amount, directly execute step S2. If the amount of data of a certain Internet of Things device is higher than the threshold amount, execute step S3.

3. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 2, characterized in that, The specific process of executing step S2 is: Run the pre-built computing tasks on each node and record their completion times; set the intervals where each group of durations corresponding to the completion times are located. Each interval where each group of durations is located corresponds to a computing power score. Match the completion times of each node with the intervals where each group of durations are located to determine the computing power score a1 of each node; Calculate the network bandwidth based on the minimum value of the incoming and outgoing bandwidths of each node, and use the calculated network bandwidth as the bandwidth evaluation value a2 of each node; Count the number of tasks to be processed by each node at the current time point, preset the intervals where each group of quantities corresponding to the number of tasks to be processed are located. Each interval where each group of quantities is located corresponds to a load additional coefficient; match the number of tasks to be processed by each node at the current time point with the intervals where each group of quantities are located to determine the load additional coefficient of each node at the current time point; Obtain the CPU usage rate and memory usage rate of each node at the current time point, multiply them by the corresponding set weight coefficients and then sum them up, and multiply the sum value by the load additional coefficient to obtain the load evaluation index a3 of each node; Normalize the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point, and then substitute them into the formula Perform weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; a1 及格 , a2 参考 , a3 阈值 Are respectively the preset passing computing power score, passing bandwidth, and load threshold index; λ1, λ2, and λ3 are the influence weight factors for calculating the computing power score a1, the bandwidth evaluation value a2, and the load evaluation index a3 in step S2.

4. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 3, characterized in that, The specific process of executing step S3 is as follows: If the data volume of a certain Internet of Things device is higher than the threshold volume, the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 of each node at the current time point are normalized and then substituted into the formula for weighted calculation to obtain the comprehensive evaluation index GTR of each node at the current time point; μ1, μ2, and μ3 are the influence weight factors of the computing power score a1, bandwidth evaluation value a2, and load evaluation index a3 in step S3.

5. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 4, characterized in that, Based on the comprehensive evaluation results, select the best processing node for the data processing of each Internet of Things device. Specifically: According to the comprehensive evaluation index GTR of each node calculated in steps S2 and S3, select the node with the evaluation index GTR as the processing node.

6. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 5, characterized in that, It also includes: The security algorithm module is used to encrypt and transmit the data collected by the Internet of Things device and the user's programming logic using a preset encryption algorithm.

7. The visual programming tool system based on artificial intelligence and the Internet of Things according to claim 6, characterized in that, Encrypt and transmit the data collected by the Internet of Things device and the user's programming logic using a preset encryption algorithm. Specifically: M1: Take the time point required for the transmission of the Internet of Things device data as the key acquisition time point, and randomly extract X sets of time numbers from within the key acquisition time point as the time key sequence; M2: Extract the model corresponding to the Internet of Things device data, and randomly extract X sets of letters from the model of the Internet of Things device as the number key sequence; M3: After randomly shuffling and overlapping the time key sequence and the number key sequence, obtain the encryption sequence of the Internet of Things device data at the current time point; M4: Use the preset conversion rule to convert the letters in the encryption sequence into numbers, that is, set different letters to correspond to a number respectively; after the conversion, take two sets of values in sequence from left to right to construct coordinate points, use the left value in the two sets of values as the abscissa, and the right value as the abscissa. If the number of numbers in the time sequence is odd, randomly extract a set of numbers for elimination, draw each group of constructed coordinate points in the rectangular coordinate system, and after the drawing is completed, connect each group of coordinate points in the construction order of the coordinate points to obtain a polygon. Use the obtained polygon as the encrypted graph of the data collected by the Internet of Things device and the user's programming logic and perform encryption.

8. The visual programming tool system based on artificial intelligence and Internet of Things according to claim 7, characterized in that, According to the set update time interval, randomly shuffle and adjust the conversion rule after reaching the set update time interval, that is, adjust the numbers corresponding to different letters.

Images