Ozone generator adaptive adjustment method and system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent control, in particular to an ozone generator self-adaptive adjustment method and system based on the Internet of Things. Comprising the following steps: monitoring the operation condition of an ozone generator by using an Internet-of-Things terminal; performing three-dimensional grid segmentation on the interior of the ozone generator, and predicting the abnormal state of the ozone generator by using a deep sequence learning algorithm; the Internet of Things terminal sends an ozone generator abnormal operation state message transmission request to the monitoring center; the monitoring center feeds back the predicted quality of the abnormal state of the ozone generator, and outputs an adjusted decision parameter by using a feedback loop algorithm; and the control center performs self-adaptive adjustment on the ozone generator according to the updated rule setting. The internal state of the ozone generator can be comprehensively monitored, and the running action of equipment can be automatically adjusted, so that the stability, fault tolerance and running efficiency of the ozone generator are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to an ozone generator adaptive adjustment method and system based on the Internet of Things. Background Art

[0002] As a strong oxidant, ozone has the functions of efficient sterilization, disinfection, deodorization, etc. Large ozone generators are composed of single or multiple ozone generating units. Each ozone generating unit adopts DBS glass dielectric discharge technology. A high voltage is applied between the high-voltage electrode and the grounding electrode. The gas is ionized to form plasma, and the oxygen molecules are dissociated and combined to form ozone. The generated ozone can be applied to sewage treatment, waste gas treatment and other fields.

[0003] However, large ozone generators face many challenges during operation. As they operate under high current and voltage, they generate high temperatures. This high temperature environment can easily damage the glass medium inside the ozone generator, greatly shortening the service life of the ozone generator, causing the entire ozone generator to stop operating, affecting the treatment of sewage and waste gas.

[0004] The heavier the ozone generator, the higher the possibility of dielectric damage. As the size of the generator increases, it becomes extremely difficult to ensure the uniformity of the electric field in the discharge area. In large electrode structures, the electrode edge effect will be more obvious, resulting in a greater difference in discharge intensity between the edge area and the center area. This will reduce the efficiency of ozone generation and may cause problems such as local overheating, affecting the normal operation and service life of the ozone generator, thereby limiting the manufacture of ozone generators that reach a certain weight.

[0005] There are obvious deficiencies in the existing technology. On the one hand, the monitoring means are relatively simple. At present, it mainly relies on collecting the temperature information of the reactor and transformer, the temperature of the cooling water, the internal temperature, etc. to monitor the operating status of the ozone generator. It is difficult to judge whether there are other potential problems inside just by the change of temperature, and it is easy to misjudge or miss the judgment. On the other hand, although the existing internal monitoring device can monitor the operation by collecting the visual information inside the ozone generator, it lacks the function of automatic control and adjustment, and often requires manual intervention. There is no effective prediction technology proposed in the existing technical solution, and it is impossible to detect abnormal symptoms in advance and prevent them. In short, although some methods have been developed to solve the problem of overheating of the ozone generator, these methods still have limitations. It is difficult to cope with the complex and changeable situation inside the ozone generator by relying solely on temperature information for monitoring. Although visual information monitoring can provide a certain display of the operation status, it cannot automatically adjust the operation of the equipment and cannot fundamentally solve the problem.

[0006] In summary, there is a need for a system and method that can comprehensively monitor the internal state of the ozone generator, which can not only automatically adjust the operation of the equipment, but also predict abnormal symptoms in advance and prevent them, thereby improving the stability, fault tolerance and operation efficiency of the ozone generator. Summary of the invention

[0007] In order to solve the above-mentioned problems, the present invention provides an ozone generator adaptive adjustment method and system based on the Internet of Things.

[0008] In the first aspect, the present invention provides an ozone generator adaptive adjustment method and system based on the Internet of Things, which adopts the following technical solutions: An adaptive adjustment method for an ozone generator based on the Internet of Things, comprising: Use IoT terminals to monitor the operating status of ozone generators; Perform three-dimensional mesh segmentation inside the ozone generator, extract the increase and decrease trend of temperature change, and use deep sequence learning algorithm to predict the abnormal state of the ozone generator; Based on the temperature change configuration of the three-dimensional grid segmentation, the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center provides feedback on the quality of the predicted abnormal state of the ozone generator, uses the feedback loop algorithm to output the adjusted decision parameters, and updates the rule settings for analyzing the prediction results; The control center makes adaptive adjustments to the ozone generator according to the updated rule settings.

[0009] Furthermore, the three-dimensional grid segmentation includes parsing the temperature measurement data, generating a temperature distribution, identifying each region according to the distribution of the glass medium, generating a first identification constraint, identifying each region in the ozone generator based on the voltage result compensation, generating a second identification constraint, configuring constraint conversion, normalizing the first identification constraint and the second identification constraint using the constraint conversion, randomly distributing cluster centers, performing regional continuous constraint aggregation, setting a minimum aggregation granularity, performing dependency planning on the aggregation results through the minimum aggregation granularity, and realizing three-dimensional grid segmentation according to the dependency planning results.

[0010] Furthermore, the extraction of the increase and decrease trend of temperature change includes configuring standard test data, performing multi-point temperature measurement in the ozone generator under the standard test data, compensating the incoming line voltage result based on the temperature measurement result, and measuring the temperature change rate over time by differentiating the temperature change of each three-dimensional grid segment in the ozone generator within a preset unit time.

[0011] Furthermore, the prediction of the abnormal state of the ozone generator includes collecting temperature change rate data, incoming line voltage data, reactor and transformer temperature data and historical operation status data segmented by three-dimensional grids, arranging the data according to time series, building a model using a long short-term memory network, training the model using the arranged data, and predicting the abnormal state of the ozone generator using the trained model.

[0012] Furthermore, the temperature change configuration of the incoming voltage demand characteristics based on the three-dimensional grid segmentation includes executing network initialization of the temperature control decision network with the incoming voltage demand characteristics, and setting trigger nodes and feedback space. When the increase and decrease trend of the temperature change is input into the initialized decision network, an improved clustering algorithm is used to analyze the abnormal state of the ozone generator, a decision result is generated, and a feedback supervision space is synchronously generated based on the decision result.

[0013] Furthermore, the feedback on the quality of predicting the abnormal state of the ozone generator includes setting an interface to receive historical data for analyzing the prediction results, the historical data including success rate, task execution time and sensor utilization efficiency, forming a real-time performance data stream, using a data analysis algorithm to analyze the real-time performance data stream, calculating an indicator of prediction quality, using a feedback loop algorithm to output adjusted decision parameters, and updating the rule settings of the feedback supervision space.

[0014] Furthermore, the use of the feedback loop algorithm to output the adjusted decision parameters includes determining the target performance index, using the PID algorithm as the feedback loop algorithm, determining the initial values ​​of the proportional coefficient, integral coefficient and differential coefficient of the PID algorithm, obtaining the operating status data of the ozone generator, and calculating the error between the current state and the target performance index, and updating the decision parameters according to the calculated control quantity.

[0015] Furthermore, the target performance indicators include the operating temperature range of the medium pipe in the ozone generator, the fluctuation range of the incoming line voltage, the range of the air supply per unit time and the range of the cooling water pressure.

[0016] Furthermore, the adaptive adjustment includes adjusting the incoming line voltage of different dielectric pipe areas, adjusting the air supply of different dielectric pipe areas, and adjusting the pressure of cooling water in the ozone generator, so as to keep the operation of the ozone generator within a normal range.

[0017] Furthermore, the monitoring center's feedback on the quality of predicting the abnormal state of the ozone generator also includes recording multiple rounds of feedback results, and the multiple rounds of feedback results all have decision-making effect identifiers, and adaptive adjustment effects are optimized based on the multiple rounds of feedback results to generate control optimization results, and a new round of adaptive adjustment of the ozone generator is performed based on the control optimization results.

[0018] In the second aspect, an ozone generator adaptive adjustment system based on the Internet of Things includes: An Internet of Things terminal, used to monitor the operating status of the ozone generator using the Internet of Things terminal; The operating status analysis module is used to perform three-dimensional grid segmentation inside the ozone generator, extract the increase and decrease trend of temperature changes, and use the deep sequence learning algorithm to predict the abnormal state of the ozone generator; A request decision module, configured to configure the incoming line voltage demand characteristics based on the temperature change of the three-dimensional grid segmentation, and the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center is used to predict the quality of abnormal conditions of the ozone generator and provide feedback, output the adjusted decision parameters using the feedback loop algorithm, and update the rule settings for analyzing the prediction results; A control center for adaptive adjustment of the ozone generator according to the updated rule settings.

[0019] In summary, the present invention has the following beneficial technical effects: 1. The present invention proposes an adaptive adjustment method and system for an ozone generator based on the Internet of Things. The operating status of the ozone generator is monitored through an Internet of Things terminal, and multi-point temperature measurement and three-dimensional grid segmentation are introduced. By differentiating the temperature change of each three-dimensional grid segment in the ozone generator within a preset unit time, the temperature change trend of different areas can be accurately grasped, and data support for the subsequent analysis of the prediction results can be provided. At the same time, the success rate, task execution time and sensor utilization efficiency are analyzed in the monitoring center to adjust the decision parameters, thereby avoiding the possible misjudgment or missed judgment that may occur when relying on a single type of data for monitoring, and improving the accuracy and reliability of the adaptive adjustment.

[0020] 2. The system of the present invention has the function of automatically adjusting the operation of the equipment. The control center makes adaptive adjustments to the ozone generator according to the updated rule settings, adjusts the incoming line voltage, gas supply and cooling water pressure of different medium pipe areas, and responds to abnormal conditions in the operation of the equipment in a timely manner. The use of deep sequence learning algorithms and improved clustering algorithms for abnormal state prediction and analysis can detect possible abnormal symptoms of the ozone generator in advance and prevent them, which helps to reduce the damage of high temperature to the glass medium and provides an effective solution for the manufacture and stable operation of large ozone generators.

[0021] 3. The present invention realizes optimized decision-making. The monitoring center performs multiple rounds of feedback on the quality of predicting the abnormal state of the ozone generator, and generates control optimization results by recording the multiple rounds of feedback results and performing adaptive adjustment effect optimization. This enables the system to continuously optimize the decision-making process and improve the accuracy of judging abnormal states. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of an adaptive adjustment method for an ozone generator based on the Internet of Things in Example 1 of the present invention. DETAILED DESCRIPTION

[0023] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0024] Example 1 Reference Figure 1 , an ozone generator adaptive adjustment method and system based on the Internet of Things of this embodiment includes: Use IoT terminals to monitor the operating status of ozone generators; IoT terminal detection mainly relies on sensors installed in the ozone generator. The sensors include multiple temperature sensors, which transmit the monitored temperature data to the IoT terminal via wireless.

[0025] Perform three-dimensional mesh segmentation inside the ozone generator, extract the increase and decrease trend of temperature change, and use deep sequence learning algorithm to predict the abnormal state of the ozone generator; The three-dimensional grid segmentation includes parsing the temperature measurement data, generating temperature distribution, identifying each area according to the distribution of the glass medium, generating a first identification constraint, identifying each area in the ozone generator based on the voltage result compensation, generating a second identification constraint, configuring constraint conversion, normalizing the first identification constraint and the second identification constraint by using the constraint conversion, randomly distributing cluster centers, executing regional continuous constraint aggregation, setting a minimum aggregation granularity, performing dependency planning on the aggregation result by using the minimum aggregation granularity, and realizing three-dimensional grid segmentation according to the dependency planning result.

[0026] First, remove outliers and perform data smoothing on the multi-point temperature measurement data obtained by the temperature sensor to ensure the accuracy and reliability of the data. A temperature distribution model inside the ozone generator is constructed based on the processed temperature data. Because the presence of glass medium may affect the conduction and distribution of temperature, the internal space of the ozone generator is divided according to the distribution of glass medium. The generated first identification constraint reflects the division and restriction of the glass medium on the region, providing a physical structure-based constraint for subsequent grid segmentation. Using the data compensated by the incoming line voltage results, changes in the incoming line voltage may affect the temperature and other characteristics of different regions in the ozone generator. The internal space of the ozone generator is divided again according to the voltage compensation results, and a new identification is generated for each region. The second identification constraint divides the region based on the voltage factor, and together with the first identification constraint, provides a more comprehensive constraint for grid segmentation.

[0027] Through constraint conversion, the first identification constraint and the second identification constraint are normalized so that they can be subsequently analyzed and processed on the same basis. Considering the physical continuity and similarity of adjacent regions, each region is aggregated. If two adjacent regions have similarities in temperature, glass medium distribution, and voltage compensation, they can be aggregated together. Set a minimum aggregation granularity, which determines the minimum size of the aggregated region. If the aggregated region is smaller than the minimum aggregation granularity, further adjustment or merging is required. The aggregation results are dependently planned through the minimum aggregation granularity to determine which regions can be merged together and which regions need to remain independent. Finally, three-dimensional grid segmentation is achieved based on the results of dependency planning, and the internal space of the ozone generator is divided into multiple three-dimensional grid regions with specific attributes and constraints.

[0028] The method for extracting the increase and decrease trend of temperature change includes configuring standard test data, performing multi-point temperature measurement in the ozone generator under the standard test data, compensating the incoming line voltage result based on the temperature measurement result, and measuring the temperature change rate over time by differentiating the temperature change of each three-dimensional grid segment in the ozone generator within a preset unit time.

[0029] Configure standard test data so that subsequent temperature measurement and other analyses have a unified standard. Set up multiple temperature sensors at different locations inside the ozone generator to obtain more comprehensive temperature distribution information. The temperature measurement results may be affected by the incoming line voltage, and fluctuations in the incoming line voltage may cause changes in the temperature inside the ozone generator. By compensating the temperature measurement results for the incoming line voltage results, the interference of the incoming line voltage on the temperature measurement can be eliminated and the accuracy of the temperature data can be improved. By differentiating the temperature change of each three-dimensional grid segment in the ozone generator within a preset unit time, the temperature change rate reflects the speed of temperature change and is one of the important indicators for judging whether the ozone generator is in an abnormal state.

[0030] The prediction of the abnormal state of the ozone generator includes collecting temperature change rate data, incoming line voltage data, reactor and transformer temperature data and historical operation status data divided by three-dimensional grids, arranging the data according to time series, building a model using a long short-term memory network, training the model using the arranged data, and predicting the abnormal state of the ozone generator using the trained model.

[0031] The feedback on the quality of predicting the abnormal state of the ozone generator includes setting an interface to receive historical data for analyzing the prediction results, the historical data including success rate, task execution time and sensor utilization efficiency, forming a real-time performance data stream, using a data analysis algorithm to analyze the real-time performance data stream, calculating an indicator of prediction quality, using a feedback loop algorithm to output adjusted decision parameters, and updating the rule settings of the feedback supervision space.

[0032] The temperature control decision network is a neural network used to determine the incoming line voltage adjustment strategy of the ozone generator. The incoming line voltage demand characteristics are used as the initial input to initialize the temperature control decision network. The incoming line voltage demand characteristics need to be converted into the initial weights and biases of the network so that the network can make preliminary decisions based on the incoming line voltage demand characteristics when it starts running.

[0033] The feedback space is used to store and process the output results of the decision network. During the operation of the decision network, information such as decision results, temperature change trends, and line voltage adjustment will be continuously stored in the feedback space. This information can be used for subsequent analysis and optimization, and is also input into the decision network as a feedback signal to adjust the network weights and biases to improve the accuracy of the decision.

[0034] The improved clustering algorithm can classify the operating status of the ozone generator according to multiple factors such as temperature change trend and incoming line voltage adjustment. The operating status can be divided into normal state, mild abnormal state and severe abnormal state. Through cluster analysis, the type and severity of abnormal state of the ozone generator can be determined more accurately. According to the analysis results of the improved clustering algorithm, the decision result is generated. The decision result includes whether the incoming line voltage needs to be adjusted and the adjustment range. At the same time, the feedback supervision space is synchronously generated based on the decision result. The feedback supervision space is a space used to supervise and adjust the decision results. It can correct and optimize the decision results according to the actual situation.

[0035] Based on the temperature change configuration of the three-dimensional grid segmentation, the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center provides feedback on the quality of the predicted abnormal state of the ozone generator, uses the feedback loop algorithm to output the adjusted decision parameters, and updates the rule settings for analyzing the prediction results; The feedback on the quality of predicting the abnormal state of the ozone generator includes setting an interface to receive historical data for analyzing the prediction results, the historical data including success rate, task execution time and sensor utilization efficiency, forming a real-time performance data stream, using a data analysis algorithm to analyze the real-time performance data stream, calculating an indicator of prediction quality, using a feedback loop algorithm to output adjusted decision parameters, and updating the rule settings of the feedback supervision space.

[0036] The feedback loop algorithm includes determining a target performance index, using a PID algorithm as a feedback loop algorithm, determining initial values ​​of a proportional coefficient, an integral coefficient and a differential coefficient of the PID algorithm, obtaining operating state data of the ozone generator, and calculating an error between a current state and the target performance index, and updating a decision parameter according to the calculated control amount.

[0037] PID algorithm is a classic control algorithm. Its basic principle is to adjust the control quantity according to the error of the system so that the system reaches the target state. In this adaptive adjustment of the ozone generator, the operating status data includes various parameters such as temperature, voltage, flow, etc. After obtaining these data through IoT devices such as sensors, they are compared with the target performance indicators to calculate the error between the target and the actual. This error calculation is the basis for the PID algorithm to adjust the control quantity. According to the calculation formula of the PID algorithm, a control quantity is obtained by combining the proportional, integral and differential coefficients and the calculated error. This control quantity can be the adjustment value of the incoming voltage, the adjustment amount of the cooling water flow, or the change in other control parameters. Then use this control quantity to update the decision parameters, which will be used to control the actual operation of the ozone generator, such as adjusting the set value of the voltage regulator, changing the speed of the cooling water pump, etc., so that the operating state of the ozone generator is closer to the target performance indicator.

[0038] After receiving the message from the IoT terminal, the monitoring center will provide feedback on the quality of the predicted abnormal state of the ozone generator. The monitoring center will evaluate whether the prediction of the abnormal state of the ozone generator is accurate and effective.

[0039] The target performance indicators include the operating temperature range of the medium pipe in the ozone generator, the fluctuation range of the incoming line voltage, the range of the air supply per unit time and the range of the cooling water pressure.

[0040] The monitoring center's feedback on the quality of predicting the abnormal state of the ozone generator also includes recording multiple rounds of feedback results, each of which has a decision effect identifier, and using the multiple rounds of feedback results to perform adaptive adjustment effect optimization, generate control optimization results, and perform a new round of adaptive adjustment of the ozone generator through the control optimization results.

[0041] After each prediction and adjustment of the abnormal state of the ozone generator, the monitoring center records the feedback results of this round. The feedback results include the type of abnormal state predicted, the adjustment measures taken, the operating status of the ozone generator after adjustment, and other information.

[0042] According to the preset evaluation indicators, the decision effect of this round is evaluated and the corresponding decision effect mark is assigned. The decision effect mark is determined by comparing the degree of fit between the predicted abnormal state and the actual abnormal state, and the time indicator for the ozone generator to resume normal operation after adjustment. As time goes by, the monitoring center continues to accumulate multiple rounds of feedback results. These results can be stored in the database for subsequent analysis and processing.

[0043] We retrieve multiple rounds of feedback results from the database for comprehensive analysis to find out the effects of different adjustment strategies under different abnormal conditions, including statistics on the success rate, average recovery time and other indicators of various adjustment strategies under different abnormal conditions.

[0044] When a new abnormal state of the ozone generator is detected, the control center will take corresponding adjustment measures according to the recommendations in the control optimization results. For example, if the temperature of the ozone generator is detected to be too high, and according to the control optimization results, the combination of reducing the incoming line voltage and increasing the air supply is more effective in similar situations, then the control center will take measures such as reducing the incoming line voltage and increasing the cooling water pressure to adjust the operating state of the ozone generator. During the adjustment process, the operating state of the ozone generator is continuously monitored to evaluate the effectiveness of the adjustment measures. If the ozone generator still does not resume normal operation after the adjustment, or new problems arise, the feedback results of multiple rounds can be re-analyzed according to the actual situation to further optimize the adjustment strategy.

[0045] The control center makes adaptive adjustments to the ozone generator according to the updated rule settings.

[0046] The adaptive adjustment includes adjusting the incoming line voltage of different medium pipe areas, adjusting the air supply volume of different medium pipe areas, and adjusting the pressure of cooling water in the ozone generator, so as to keep the operation of the ozone generator within a normal range.

[0047] Example 2 This embodiment provides an ozone generator adaptive adjustment system based on the Internet of Things, including: An Internet of Things terminal, used to monitor the operating status of the ozone generator using the Internet of Things terminal; The operating status analysis module is used to perform three-dimensional grid segmentation inside the ozone generator, extract the increase and decrease trend of temperature changes, and use the deep sequence learning algorithm to predict the abnormal state of the ozone generator; A request decision module, configured to configure the incoming line voltage demand characteristics based on the temperature change of the three-dimensional grid segmentation, and the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center is used to predict the quality of abnormal conditions of the ozone generator and provide feedback, output the adjusted decision parameters using the feedback loop algorithm, and update the rule settings for analyzing the prediction results; A control center for adaptive adjustment of the ozone generator according to the updated rule settings.

[0048] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive adjustment method for an ozone generator based on the Internet of Things, characterized in that: include: Use IoT terminals to monitor the operating status of ozone generators; Perform three-dimensional mesh segmentation inside the ozone generator, extract the increase and decrease trend of temperature change, and use deep sequence learning algorithm to predict the abnormal state of the ozone generator; Based on the temperature change configuration of the three-dimensional grid segmentation, the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center provides feedback on the quality of the predicted abnormal state of the ozone generator, uses the feedback loop algorithm to output the adjusted decision parameters, and updates the rule settings for analyzing the prediction results; The control center makes adaptive adjustments to the ozone generator according to the updated rule settings.

2. According to the method for adaptively adjusting an ozone generator based on the Internet of Things in claim 1, it is characterized in that: The three-dimensional grid segmentation includes parsing the temperature measurement data, generating temperature distribution, identifying each area according to the distribution of the glass medium, generating a first identification constraint, identifying each area in the ozone generator based on the voltage result compensation, generating a second identification constraint, configuring constraint conversion, normalizing the first identification constraint and the second identification constraint by using the constraint conversion, randomly distributing cluster centers, executing regional continuous constraint aggregation, setting a minimum aggregation granularity, performing dependency planning on the aggregation result by using the minimum aggregation granularity, and realizing three-dimensional grid segmentation according to the dependency planning result.

3. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 1 is characterized in that: The method for extracting the increase and decrease trend of temperature change includes configuring standard test data, performing multi-point temperature measurement in the ozone generator under the standard test data, compensating the incoming line voltage result based on the temperature measurement result, and measuring the temperature change rate over time by differentiating the temperature change of each three-dimensional grid segment in the ozone generator within a preset unit time.

4. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 3 is characterized in that: The prediction of the abnormal state of the ozone generator includes collecting temperature change rate data, incoming line voltage data, reactor and transformer temperature data and historical operation status data divided by three-dimensional grids, arranging the data according to time series, building a model using a long short-term memory network, training the model using the arranged data, and predicting the abnormal state of the ozone generator using the trained model.

5. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 4 is characterized in that: The temperature change configuration of the incoming voltage demand characteristics based on the three-dimensional grid segmentation includes executing network initialization of the temperature control decision network with the incoming voltage demand characteristics, and setting trigger nodes and feedback space. When the increase and decrease trend of the temperature change is input into the initialized decision network, an improved clustering algorithm is used to analyze the abnormal state of the ozone generator, a decision result is generated, and a feedback supervision space is synchronously generated based on the decision result.

6. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 1 is characterized in that: The feedback on the quality of predicting the abnormal state of the ozone generator includes setting an interface to receive historical data for analyzing the prediction results, the historical data including success rate, task execution time and sensor utilization efficiency, forming a real-time performance data stream, using a data analysis algorithm to analyze the real-time performance data stream, calculating an indicator of prediction quality, using a feedback loop algorithm to output adjusted decision parameters, and updating the rule settings of the feedback supervision space.

7. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 6 is characterized in that: The method of using the feedback loop algorithm to output the adjusted decision parameters includes determining the target performance index, using the PID algorithm as the feedback loop algorithm, determining the initial values ​​of the proportional coefficient, the integral coefficient and the differential coefficient of the PID algorithm, obtaining the operating status data of the ozone generator, and calculating the error between the current status and the target performance index, and updating the decision parameters according to the calculated control quantity.

8. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 7 is characterized in that: The adaptive adjustment includes adjusting the incoming line voltage of different medium pipe areas, adjusting the air supply volume of different medium pipe areas, and adjusting the pressure of cooling water in the ozone generator, so that the operation of the ozone generator remains within a normal range.

9. The method for adaptively adjusting an ozone generator based on the Internet of Things according to claim 8, characterized in that: The monitoring center's feedback on the quality of predicting the abnormal state of the ozone generator also includes recording multiple rounds of feedback results, each of which has a decision effect identifier, and using the multiple rounds of feedback results to perform adaptive adjustment effect optimization, generate control optimization results, and perform a new round of adaptive adjustment of the ozone generator through the control optimization results.

10. An ozone generator adaptive adjustment system based on the Internet of Things, characterized in that: include: An Internet of Things terminal, used to monitor the operating status of the ozone generator using the Internet of Things terminal; The operating status analysis module is used to perform three-dimensional grid segmentation inside the ozone generator, extract the increase and decrease trend of temperature changes, and use the deep sequence learning algorithm to predict the abnormal state of the ozone generator; A request decision module, configured to configure the incoming line voltage demand characteristics based on the temperature change of the three-dimensional grid segmentation, and the Internet of Things terminal sends an ozone generator abnormal operation status message transmission request to the monitoring center; The monitoring center is used to predict the quality of abnormal conditions of the ozone generator and provide feedback, output the adjusted decision parameters using the feedback loop algorithm, and update the rule settings for analyzing the prediction results; A control center for adaptive adjustment of the ozone generator according to the updated rule settings.

Images