Ozone removal control system based on artificial intelligence

Abstract

The invention discloses an ozone removal control system based on artificial intelligence, and relates to the technical field of ozone removal, and the system comprises the steps: carrying out the real-time monitoring of a target region, obtaining regional ozone monitoring information and regional meteorological monitoring information, and generating an ozone removal decision; generating an ozone removal quality prediction result based on the ozone removal decision; judging whether the ozone removal quality prediction result meets a prediction constraint or not; and adjusting the ozone removal decision, generating an ozone removal decision adjustment result, and performing ozone removal control. According to the invention, the technical problem of low ozone removal effect caused by lack of accuracy in ozone removal decision in the prior art is solved, and the technical effects of realizing accurate monitoring and decision of ozone in the target area and improving the removal effect and reliability are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of ozone removal, and in particular to an ozone removal control system based on artificial intelligence. Background Art

[0002] With the acceleration of industrialization and urbanization, ozone pollution in the atmosphere has become increasingly serious, posing a huge threat to human health and the ecological environment. In the past, the means of monitoring regional ozone were relatively simple, and it was difficult to obtain comprehensive and accurate ozone concentration and meteorological information, which made the formulation of removal decisions lack sufficient data support. At the same time, the generation of removal decisions is mostly based on experience or simple models, which cannot fully consider the comprehensive impact of various complex factors, resulting in unstable removal effects and low efficiency.

[0003] The existing technology has the technical problem of lack of accuracy in ozone removal decision-making, resulting in poor ozone removal effect. Summary of the invention

[0004] The present application provides an ozone removal control system based on artificial intelligence, which is used to solve the technical problem in the prior art that ozone removal decisions lack accuracy, resulting in poor ozone removal effects.

[0005] In view of the above problems, the present application provides an ozone removal control system based on artificial intelligence.

[0006] The present application provides an ozone removal control system based on artificial intelligence, the system comprising: an ozone removal quality prediction module, the ozone removal quality prediction module is used to generate an ozone removal quality prediction result based on the ozone removal decision; an ozone removal quality prediction judgment module, the ozone removal quality prediction judgment module is used to judge whether the ozone removal quality prediction result meets the predetermined ozone removal quality prediction constraint; an ozone removal adjustment module, the ozone removal adjustment module is used to adjust the ozone removal decision according to the predetermined ozone removal quality prediction constraint if the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, and generate an ozone removal decision adjustment result; an ozone removal control module, the ozone removal control module is used to perform ozone removal control on the target area based on the ozone removal decision adjustment result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The regional monitoring module is used to monitor the target area in real time and obtain regional ozone monitoring information and regional meteorological monitoring information; the ozone removal decision module generates an ozone removal decision based on the regional ozone monitoring information and the regional meteorological monitoring information; the ozone removal quality prediction module generates an ozone removal quality prediction result based on the ozone removal decision; the ozone removal quality prediction judgment module is used to judge whether the ozone removal quality prediction result meets the predetermined ozone removal quality prediction constraint; the ozone removal adjustment module is used to adjust the ozone removal decision according to the predetermined ozone removal quality prediction constraint if the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, and generate an ozone removal decision adjustment result; the ozone removal control module performs ozone removal control on the target area based on the ozone removal decision adjustment result. The technical effect of realizing accurate monitoring and decision-making of ozone in the target area and improving the removal effect and reliability is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in 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 only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A structural schematic diagram of an ozone removal control system based on artificial intelligence is provided for an embodiment of the present application.

[0010] Figure 2 A schematic diagram of the execution structure of an ozone removal decision module of an ozone removal control system based on artificial intelligence is provided for an embodiment of the present application.

[0011] Explanation of the reference numerals: regional monitoring module 10 , ozone removal decision module 20 , ozone removal quality prediction module 30 , ozone removal quality prediction and judgment module 40 , ozone removal adjustment module 50 , ozone removal control module 60 . DETAILED DESCRIPTION

[0012] The present application provides an ozone removal control system based on artificial intelligence, which is used to solve the technical problem that the ozone removal decision in the prior art lacks accuracy, resulting in poor ozone removal effect. The technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. Obviously, the described embodiment is only a part of the embodiment of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] Embodiment 1, as Figure 1 As shown, the present application provides an ozone removal control system based on artificial intelligence, the system comprising: The regional monitoring module 10 is used to monitor the target area in real time and obtain regional ozone monitoring information and regional meteorological monitoring information.

[0014] Specifically, the regional monitoring module 10 is the data acquisition part of the whole system, which is equipped with a series of high-precision monitoring instruments and sensors, distributed in key positions of the target area. These monitoring instruments include but are not limited to ozone concentration detectors, which can accurately measure the content of ozone in the air and generate detailed regional ozone monitoring information, covering ozone concentration data in different time periods and different locations. At the same time, it is also equipped with meteorological monitoring equipment, such as thermometers, hygrometers, anemometers, etc., which are used to collect meteorological data in the target area, such as temperature, humidity, wind speed, wind direction, etc., to form regional meteorological monitoring information. These monitoring devices collect data at a high frequency to ensure that the changes in ozone concentration and meteorological conditions in the target area can be captured in real time. The collected data will be quickly transmitted and sorted to provide timely and accurate basic information for subsequent analysis and decision-making.

[0015] The ozone removal decision module 20 is used to generate an ozone removal decision based on the regional ozone monitoring information and the regional meteorological monitoring information.

[0016] Specifically, the ozone removal decision module 20 receives regional ozone monitoring information and regional meteorological monitoring information, and uses BP neural network technology to generate ozone removal decisions. BP neural network is a powerful machine learning algorithm that can automatically learn complex patterns and relationships from input data. In this module, regional ozone monitoring information (such as ozone concentration values ​​at different locations and times, concentration change trends, etc.) and regional meteorological monitoring information (temperature, humidity, wind speed, wind direction, etc.) are used as input layer data of the BP neural network. The network contains multiple hidden layers, and the neurons in each hidden layer are connected to the neurons in the previous and next layers through weights. These weights are constantly adjusted during the training process to optimize the network's fitting ability and prediction accuracy for the input data. The BP neural network is trained through a large amount of historical monitoring data. During the training process, the network learns the intrinsic relationship between different regional ozone and meteorological conditions and effective ozone removal strategies. When new regional ozone monitoring information and regional meteorological monitoring information are input into the trained network, the BP neural network will undergo complex calculations and signal transmission, and finally generate corresponding ozone removal decisions at the output layer. Through analysis and decision generation based on BP neural network, efficient and reasonable ozone removal decisions can be made more accurately based on real-time complex environmental information.

[0017] The ozone removal quality prediction module 30 is used to generate an ozone removal quality prediction result based on the ozone removal decision.

[0018] Specifically, the ozone removal quality prediction module 30 plays a key role in predicting the future ozone removal effect. First, the ozone removal rate prediction model is activated. This model is a tool specifically used to predict the ozone removal rate. It is built based on a large amount of historical data and related algorithms. Based on the given ozone removal decision, the ozone removal rate prediction model is used for calculation and analysis to generate a predicted ozone removal rate. This process takes into account the various factors involved in the decision, such as the impact of equipment type, operating parameters, and processing area on the ozone removal rate. Then, based on the relevant information in the ozone removal decision, energy consumption is predicted to obtain a predicted ozone removal energy consumption coefficient, which can reflect the energy consumption level required to implement the removal decision. Then, based on the ozone removal decision, the ozone removal efficiency is predicted to generate a predicted ozone removal efficiency, that is, the amount of ozone that can be removed per unit time. Finally, a comprehensive evaluation is obtained by combining the above-mentioned predicted results of ozone removal rate, predicted ozone removal energy consumption coefficient and predicted ozone removal efficiency. This is the ozone removal quality prediction result, which can provide a reliable basis for subsequent decision-making evaluation and adjustment to ensure the ultimate realization of the ideal ozone removal quality.

[0019] The ozone removal quality prediction and judgment module 40 is used to judge whether the ozone removal quality prediction result meets the predetermined ozone removal quality prediction constraint.

[0020] Specifically, the ozone removal quality prediction and judgment module 40 is a part specifically responsible for evaluating the ozone removal quality prediction results. Its main function is to compare and judge the obtained ozone removal quality prediction results with a series of pre-set constraints. The predetermined ozone removal quality prediction constraints include ozone removal rate constraints, ozone removal energy consumption constraints and ozone removal efficiency constraints. The ozone removal rate constraint stipulates the proportion of ozone removal that must be achieved under certain conditions. For example, it is required that the ozone removal rate must not be less than 80% in a specific area and time. The ozone removal energy consumption constraint limits the energy consumption in the ozone removal process. For example, it is stipulated that the energy consumption cannot exceed a certain value for each unit amount of ozone removed. The ozone removal efficiency constraint sets the amount of ozone removal that needs to be completed in a unit time. For example, at least a certain proportion of ozone must be removed per hour. The ozone removal quality prediction and judgment module 40 will carefully check the indicators such as ozone removal rate, energy consumption and efficiency in the prediction results to see if they meet these predetermined constraints. If all parameters in the prediction result are within the allowable range of the preset constraints, the module will determine that the prediction result meets the requirements; conversely, if any one or more parameters exceed the constraint range, the module will determine that the prediction result does not meet the preset constraints. Through this precise judgment, it is possible to timely discover the situation where the ozone removal effect is not up to standard, providing a clear basis for subsequent adjustments and optimizations.

[0021] The ozone removal adjustment module 50 is used to adjust the ozone removal decision according to the predetermined ozone removal quality prediction constraint if the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, and generate an ozone removal decision adjustment result.

[0022] Specifically, when the ozone removal quality prediction judgment module 40 determines that the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, the module is activated. It first deeply analyzes the specific aspects and degree of non-compliance with the constraint, for example, if the prediction result shows that the ozone removal speed is too slow and the desired concentration reduction level cannot be achieved within the specified time, or the concentration distribution after removal is uneven. Then, according to the requirements and goals of the predetermined ozone removal quality prediction constraint, the previously generated ozone removal decision is adjusted. This includes increasing the operating power of the ozone removal equipment, extending the working time of the equipment, adjusting the layout of the equipment to improve the uniformity of the removal effect, or replacing a more efficient type of removal equipment. When making adjustments, the module will comprehensively consider multiple factors, such as the performance limit of the equipment, energy consumption, cost-effectiveness, etc., to ensure that the generated ozone removal decision adjustment result can not only meet the constraint requirements, but also has practical operability and economy. For example, if the original decision is to use low-power equipment to run for 2 hours, the adjustment module will adjust it to use medium-power equipment to run for 3 hours, or increase the number of the same type of equipment at the same time. Through this intelligent regulation and optimization, the aim is to make ozone removal work more effectively to achieve the expected quality standards.

[0023] The ozone removal control module 60 is used to perform ozone removal control on the target area based on the ozone removal decision adjustment result.

[0024] Specifically, after receiving the ozone removal decision adjustment result generated by the ozone removal adjustment module 50, the module is responsible for converting these decisions into specific control instructions and accurately sending them to the ozone removal equipment in the target area. It establishes an effective communication connection with each ozone removal device, which can be a wired connection or a wireless communication method, such as Bluetooth, Wi-Fi, etc. The control module generates a corresponding control signal based on the detailed information about the device type, operating parameters, start time, etc. in the adjustment result. For example, for a specific ozone decomposition device, the control signal may include a start or shut down instruction, a power adjustment instruction, a running time setting, etc. These control instructions are quickly transmitted to the corresponding device to ensure that the device can operate accurately according to the latest optimized decision. At the same time, the control module will also monitor the operating status and feedback information of the device in real time to confirm whether the device has executed the control instruction normally and whether the actual effect of the removal work meets expectations. If a fault or abnormality is found in the device during operation, the control module can issue an alarm in time and take corresponding emergency measures, such as suspending the operation of some equipment, re-adjusting the control strategy, etc., to ensure the safety, stability and efficiency of ozone removal work. Through this precise and real-time control, effective ozone removal in the target area can be achieved and the predetermined removal quality requirements can be met.

[0025] In one possible implementation, Figure 2 As shown, the ozone removal decision module 20 also includes: Based on the target area, a regional ozone monitoring information sample set, a regional meteorological monitoring information sample set and an ozone removal decision sample set are loaded; the BP neural network is trained with the regional ozone monitoring information sample set and the regional meteorological monitoring information sample set as input information and the ozone removal decision sample set as output information, and an ozone removal decision loss coefficient is obtained after each predetermined number of trainings; if the ozone removal decision loss coefficient is less than an ozone removal decision loss threshold, the ozone removal decision module is generated.

[0026] Specifically, for a specific target area, a relevant set of sample data is extracted from a pre-prepared database or storage location. The regional ozone monitoring information sample set contains detailed data such as ozone concentrations and concentration change trends monitored at different time points and locations in the target area in the past. The regional meteorological monitoring information sample set covers meteorological condition data for the same time period and location, such as temperature, humidity, wind speed, wind direction, etc. The ozone removal decision sample set records the specific removal decisions taken for these different ozone and meteorological conditions in the past, including the type of equipment used, equipment operating parameters, and the implementation time of the removal strategy. By loading these sample sets, a foundation and basis is provided for the subsequent use of these data to train the neural network in order to establish a model that can accurately generate ozone removal decisions.

[0027] In the process of training the BP neural network, the regional ozone monitoring information sample set and the regional meteorological monitoring information sample set are used as input data and passed to the input layer of the neural network. At the same time, the ozone removal decision sample set is set as the desired output result, that is, the target that the network should learn to approach. During the training process, the network calculates and predicts based on the input data, and compares the predicted results with the desired output (ozone removal decision sample set). According to the pre-set number of training times (for example, every 100 times), the ozone removal decision loss coefficient is calculated once. This loss coefficient is usually calculated by a certain loss function (such as mean square error, cross entropy, etc.), which is used to measure the degree of difference between the ozone removal decision predicted by the current network and the actual ozone removal decision sample. If the loss coefficient is small, it means that the network's prediction result is closer to the actual sample and the network training effect is good; on the contrary, if the loss coefficient is large, it means that the network's prediction still has a large deviation and needs to be further trained and optimized.

[0028] When the calculated ozone removal decision loss coefficient is less than the pre-set ozone removal decision loss threshold, it means that the trained BP neural network has reached a certain precision and accuracy requirement in predicting ozone removal decisions. This loss threshold is a predetermined standard value used to determine whether the training effect of the neural network is good enough. If the loss coefficient is less than the threshold, it means that the neural network has learned sufficiently effective patterns and rules, and can more accurately generate reasonable ozone removal decisions based on the input regional ozone monitoring information and regional meteorological monitoring information. At this point, the trained neural network and its related parameters, configurations, etc. can be encapsulated to form an ozone removal decision module. When this module receives new regional ozone and meteorological monitoring information in the future, it can quickly and accurately output the corresponding ozone removal decision for actual ozone removal work.

[0029] In a possible implementation, the ozone removal quality prediction module 30 further includes: Activate the ozone removal rate prediction model of the ozone removal quality prediction module; based on the ozone removal decision, generate a predicted ozone removal rate according to the ozone removal rate prediction model; predict the ozone removal energy consumption according to the ozone removal decision to obtain a predicted ozone removal energy consumption coefficient; predict the ozone removal efficiency according to the ozone removal decision to generate a predicted ozone removal efficiency; obtain the ozone removal quality prediction result according to the predicted ozone removal rate, the predicted ozone removal energy consumption coefficient and the predicted ozone removal efficiency.

[0030] Specifically, in the process of predicting the ozone removal quality, the prediction model specifically used to calculate the ozone removal rate in the ozone removal quality prediction module is started. The steps to build an ozone removal rate prediction model are as follows: widely collect a large amount of data including regional ozone monitoring information (such as ozone concentrations at different locations and times), regional meteorological monitoring information (such as temperature, humidity, wind speed, etc.), the ozone removal decisions adopted (such as equipment type, operating parameters, etc.) and the corresponding actual ozone removal rate; preprocess and extract features from the collected data, for example, standardize or normalize numerical features such as ozone concentration, temperature, humidity, wind speed, etc. so that the model can learn better, and perform independent encoding conversion for category features such as equipment type; build a multi-layer perceptron model, and design a multi-layer perceptron containing an input layer, two hidden layers, and an output layer. The number of neurons in the input layer is the same as the number of extracted features. The first hidden layer can be set to 128 neurons, and the second hidden layer can be set to 64 neurons, the activation function can choose the ReLU function to introduce nonlinearity, and the output layer sets a neuron to output the predicted ozone removal rate; use the prepared training data set to train the model, and adjust the weights and biases of the model through the back propagation algorithm to minimize the loss function (such as mean square error) between the predicted value and the actual value; use an independent test data set to evaluate the trained model, and calculate the root mean square error (RMSE), mean absolute percentage error (MAPE) and other indicators to measure the performance of the model; if the model performance does not meet the requirements, the model can be optimized by experimenting with different hyperparameters, such as the number of hidden layer neurons, learning rate, number of layers, etc.; use new data that has not been used in training and evaluation to verify the finalized model to ensure its reliability and generalization ability on different data. When it is necessary to activate this ozone removal rate prediction model based on the multi-layer perceptron, the system will load the model structure and trained weight parameters, and be ready to receive new input data for ozone removal rate prediction.

[0031] When a specific ozone removal decision is determined, the relevant information contained therein, such as the type of removal equipment used, equipment operating parameters, removal time, etc., is input into the built and trained ozone removal rate prediction model. The model will process and calculate based on these input decision parameters, and finally output a value, which is the predicted ozone removal rate, indicating the proportion of ozone that is expected to be removed to the initial amount of ozone under the given ozone removal decision conditions.

[0032] After obtaining the ozone removal decision, it is necessary to estimate the energy consumption generated by executing the decision. First, the ozone removal equipment involved in the decision will be analyzed in detail to clarify the power of the equipment and the expected operating time. Power is usually measured in watts, and the operating time is measured in hours. Multiply the power by the operating time to get the total energy consumed during the operation of the equipment. Next, the energy efficiency coefficient of the equipment should be considered. This coefficient represents the amount of ozone that can be removed per unit of power per unit of time. At the same time, the environmental correction factor needs to be introduced because environmental conditions (such as temperature, humidity, etc.) will affect energy consumption and removal effects. Then, multiply the power of the equipment by the operating time, and then multiply it by the energy efficiency coefficient and the environmental correction factor to get the expected amount of ozone removed. Finally, divide the total energy consumed by the operation of the equipment by the expected amount of ozone removed, and the result is the predicted ozone removal energy consumption coefficient.

[0033] Once an ozone removal decision is made, its removal efficiency needs to be predicted. First, analyze the key factors involved in the decision, such as the performance parameters, working mode, and operating time of the removal equipment. Then, consider the initial ozone concentration, area size, spatial distribution, and other conditions of the target area. At the same time, it is also necessary to comprehensively consider the impact of meteorological conditions such as temperature, humidity, and wind speed on ozone diffusion and removal. Next, based on this information, calculate the amount of ozone that is expected to be removed within a given time, compare the expected amount of ozone removed with the initial amount of ozone, and then divide it by the time spent on removal to obtain the amount of ozone removed per unit time. This amount of ozone removed per unit time is the predicted ozone removal efficiency, which reflects the speed and effectiveness of the removal work under a specific ozone removal decision.

[0034] The predicted ozone removal rate reflects the proportion of the amount of ozone that is expected to be removed to the initial amount, and is a direct indicator for measuring the removal effect. A higher removal rate means that the ozone concentration can be reduced more effectively. The predicted ozone removal energy consumption coefficient reflects the energy consumed to achieve ozone removal. A lower energy consumption coefficient means that less energy is consumed when a certain removal effect is achieved, which has better energy utilization efficiency. The predicted ozone removal efficiency reflects the removal effect per unit time. A higher removal efficiency means that the ideal removal effect can be achieved in a shorter time. These three aspects are combined to evaluate the overall quality of ozone removal. If the predicted ozone removal rate is high, the predicted ozone removal energy consumption coefficient is low, and the predicted ozone removal efficiency is high, then the ozone removal quality prediction result will be very good, indicating that the removal decision performs well in terms of effect, energy consumption and time efficiency. By comprehensively considering these three key indicators, the pros and cons of ozone removal decisions can be comprehensively and accurately evaluated, providing strong support for subsequent decision adjustments and optimizations.

[0035] In a possible implementation, the ozone removal quality prediction module 30 further includes: Perform ozone removal energy consumption retrieval according to the ozone removal decision to obtain multiple sample ozone removal energy consumption coefficients; perform deviation value evaluation on the multiple sample ozone removal energy consumption coefficients to generate an ozone removal energy consumption deviation evaluation value; determine whether the ozone removal energy consumption deviation evaluation value is less than an ozone removal energy consumption deviation evaluation threshold; if the ozone removal energy consumption deviation evaluation value is less than the ozone removal energy consumption deviation evaluation threshold, perform centralized value evaluation on the multiple sample ozone removal energy consumption coefficients to generate the predicted ozone removal energy consumption coefficient.

[0036] Specifically, when a specific ozone removal decision is determined, it will be searched and matched in a pre-established database or data storage system based on key information in the decision, such as the type of removal equipment used, equipment operating parameters, characteristics of the removal area, etc. This database stores a large number of past ozone removal cases and their corresponding energy consumption coefficient data. By comparing the characteristics of the current decision with the records in the database, multiple similar or related samples are found, and the energy consumption coefficients corresponding to these similar samples are extracted to form a collection of multiple sample ozone removal energy consumption coefficients.

[0037] After obtaining multiple sample ozone removal energy consumption coefficients, it is necessary to evaluate the degree of dispersion and abnormal conditions of these coefficients, which involves the process of deviation evaluation. Calculate the difference between each sample coefficient and the mean, and perform statistical analysis on these differences to calculate the standard deviation of these differences. The larger the standard deviation, the greater the dispersion of the sample coefficient and the larger the deviation value; the smaller the standard deviation, the relatively concentrated sample coefficient and the smaller the deviation value. Based on the above calculations and analysis, a specific value will eventually be generated, namely the ozone removal energy consumption deviation evaluation value. This value intuitively reflects the degree of dispersion of the ozone removal energy consumption coefficient of this group of samples and the degree of deviation from the normal range.

[0038] After obtaining the ozone removal energy consumption deviation evaluation value, compare this value with the pre-set ozone removal energy consumption deviation evaluation threshold. For example, assuming that the calculated ozone removal energy consumption deviation evaluation value is 1.2, and the set ozone removal energy consumption deviation evaluation threshold is 1.5, then it can be considered that the discrete degree of these sample ozone removal energy consumption coefficients is relatively small and within an acceptable range, which may indicate that the consistency of the data is good and there are no excessive abnormal fluctuations. On the contrary, if the ozone removal energy consumption deviation evaluation value is greater than or equal to the threshold, it means that the data is more discrete, there are some abnormal values, or the distribution of the data is not concentrated enough, and further analysis and processing are required. Through this comparison and judgment, the stability and reliability of the sample data can be determined, providing a basis for subsequent processing and decision-making.

[0039] When it is determined that the ozone removal energy consumption deviation evaluation value is less than the set threshold, it means that the distribution of these sample energy consumption coefficients is relatively concentrated and stable. Next, these sample energy consumption coefficients will be evaluated for the centralized value. Calculate the average value, add all the sample energy consumption coefficients, and then divide them by the number of samples. The average value is used as the predicted ozone removal energy consumption coefficient. Through these centralized value evaluation methods, the final result generated is the predicted ozone removal energy consumption coefficient, which is used to predict and evaluate the energy consumption of ozone removal and improve the prediction accuracy.

[0040] In a possible implementation, the ozone removal control module 60 further includes: If the ozone removal quality prediction result satisfies the predetermined ozone removal quality prediction constraint, the ozone removal control module performs ozone removal control on the target area based on the ozone removal decision.

[0041] Specifically, the predetermined ozone removal quality prediction constraints are a series of standards and conditions set in advance. These standards are formulated based on various factors, such as environmental protection requirements, economic cost considerations, and technical feasibility. For example, it is stipulated that the predicted ozone removal rate must reach more than 85% to ensure that the ozone concentration is sufficiently effectively reduced; the predicted ozone removal energy consumption coefficient shall not be higher than 10 joules per unit ozone removal to control energy costs; the predicted ozone removal efficiency shall not be less than 25% per hour to ensure that the removal task is completed within a reasonable time. When the obtained ozone removal quality prediction results meet these predetermined constraints in all key indicators, it means that the current ozone removal decision is theoretically feasible and meets the requirements. Next, the ozone removal control module will come into play. This module is the control center that actually performs the ozone removal operation. It will send instructions to related equipment and systems based on the specific details of the previously determined ozone removal decision, such as the type of equipment used, equipment operating parameters, operation time and sequence. Through such a process, if everything goes well, the ozone in the target area can be effectively removed according to the predetermined decision under the premise of meeting the quality prediction constraints.

[0042] In a possible implementation, the ozone removal adjustment module 50 further includes: Perform ozone removal control feature backtracking according to the regional ozone monitoring information and the regional meteorological monitoring information to obtain an ozone removal control adjustment constraint space; adjust the ozone removal decision according to the ozone removal control adjustment constraint space to obtain a first adjusted ozone removal decision; perform multi-feature prediction on the first adjusted ozone removal decision based on the ozone removal quality prediction module to obtain an ozone removal quality prediction result of the first adjusted decision; determine whether the ozone removal quality prediction result of the first adjusted decision meets the predetermined ozone removal quality prediction constraint; if the ozone removal quality prediction result of the first adjusted decision meets the predetermined ozone removal quality prediction constraint, add the first adjusted ozone removal decision to the ozone removal decision adjustment result.

[0043] Specifically, regional ozone monitoring information provides detailed data on the specific values, changing trends, and distribution of ozone concentrations at different times and locations. Regional meteorological monitoring information includes real-time and historical data on meteorological elements such as temperature, humidity, air pressure, wind speed, and wind direction. Through in-depth analysis and retrospection of these large amounts of monitoring data, we can understand the ozone removal control measures and their effects taken in the past under different combinations of ozone concentrations and meteorological conditions. For example, we analyze the removal effect achieved by using a certain intensity of removal equipment operating mode and duration under specific meteorological conditions of high ozone concentration, high temperature, and low wind speed. By comprehensively considering the control measures and effects under various different conditions, a series of restrictions and ranges can be determined. For example, the upper and lower limits of the equipment operating power, the shortest and longest limits of the operating time, etc. are determined under a specific ozone concentration range and meteorological condition combination. These restrictions and ranges together constitute the ozone removal control adjustment constraint space, providing a feasible adjustment framework for subsequent optimization of ozone removal decisions.

[0044] When the ozone removal control adjustment constraint space is determined, this space provides clear rules and boundaries for the adjustment of ozone removal decisions. The original ozone removal decision was based on preliminary analysis and settings, but after considering the constraint space determined by the actual regional ozone monitoring information and regional meteorological monitoring information, it needs to be optimized and improved. For example, the constraint space stipulates that under the current ozone concentration and meteorological conditions, the operating power of the equipment cannot exceed 800 watts, while the equipment power is set to 1000 watts in the original ozone removal decision. Then, during the adjustment, the equipment power will be reduced to within 800 watts and adjusted to 750 watts. Alternatively, the constraint space indicates that under certain conditions, the duration of the removal operation should not be less than 2 hours, while the original decision only plans 1.5 hours. At this time, the time of the removal operation will be extended to 2 hours or longer, such as 2.5 hours. By comprehensively considering the various restrictions and requirements in the constraint space, the equipment parameters, operating time, and equipment combination used in the original ozone removal decision are modified and improved, thereby obtaining the first adjustment ozone removal decision.

[0045] The ozone removal quality prediction module is a system specifically used to evaluate and predict the effect of ozone removal. When the first adjustment ozone removal decision is made, the detailed information of this decision, including the type of equipment used, equipment operating parameters, expected operating time, etc., is input into the ozone removal quality prediction module. This module will comprehensively consider multiple features for prediction. It will predict the ozone removal rate that can be achieved after the implementation of this decision, that is, the proportion of the expected amount of ozone that can be removed to the initial amount of ozone; it will also predict the energy consumption of the entire removal process and calculate the energy consumed per unit of ozone removal; at the same time, it will also predict the removal efficiency, that is, the amount of ozone that can be removed per unit time. Through the prediction and calculation of these multiple features, a comprehensive prediction result is finally obtained. This result is the ozone removal quality prediction result of the first adjustment decision, which comprehensively reflects the expected performance of the decision in terms of ozone removal effect, energy consumption and efficiency.

[0046] The predetermined ozone removal quality prediction constraints are a series of standards and conditions set in advance to measure the pros and cons and feasibility of ozone removal decisions. After obtaining the ozone removal quality prediction results of the first adjustment decision, it is necessary to compare them with these predetermined constraints one by one. For example, the predetermined constraints include that the ozone removal rate must reach more than 80%, the energy consumption coefficient must not exceed 8 joules / unit of ozone, and the removal efficiency must not be less than 25% per hour. If the ozone removal rate is 85%, the energy consumption coefficient is 7 joules / unit of ozone, and the removal efficiency is 30% per hour in the prediction results of the first adjustment decision, then it is considered that this prediction result meets the predetermined constraints. Through such detailed comparison and judgment, it can be determined whether the first adjustment decision has met the expected requirements in terms of quality, thereby providing a basis for subsequent decision-making choices and further adjustments.

[0047] If it is determined that the predicted results of the first adjustment decision in terms of ozone removal quality meet the various requirements and standards set in advance (i.e., the predetermined ozone removal quality prediction constraints), then the first adjustment ozone removal decision that is considered feasible after optimization and adjustment will be included in the set of ozone removal decision adjustment results. When the ozone removal operation is actually performed later, this decision becomes an effective solution that can be selected or further referenced and optimized, providing a wealth of options for the final selection of the most appropriate ozone removal solution.

[0048] In a possible implementation, the ozone removal adjustment module 50 further includes: An ozone removal control feature association search is performed based on the regional ozone monitoring information and the regional meteorological monitoring information to obtain an associated ozone removal control scheme set; an ozone removal control trigger interval is identified based on the associated ozone removal control scheme set to obtain a multi-dimensional ozone removal control trigger space; ozone removal control constraint information of the ozone removal control module is obtained; and the multi-dimensional ozone removal control trigger space is corrected and optimized based on the ozone removal control constraint information to generate the ozone removal control adjustment constraint space.

[0049] Specifically, the regional ozone monitoring information includes the real-time data, historical data, concentration distribution and other details of the ozone concentration in the region, while the regional meteorological monitoring information covers the relevant data of various meteorological elements such as temperature, humidity, air pressure, wind speed, wind direction, etc. When performing an associated search, the currently monitored ozone concentration, concentration change trend and specific meteorological conditions (such as the current temperature, humidity, wind speed and direction, etc.) are used as key features for the search. Then, a matching or similar historical record is searched in a pre-established database or data warehouse, which contains ozone removal control plans taken under similar ozone and meteorological conditions. For example, if the currently monitored ozone concentration in the region is high, the temperature is moderate, the humidity is high, and the wind speed is low, the system will search the database for past cases with similar ozone concentration, temperature, humidity and wind speed conditions. The ozone removal control plans corresponding to these similar cases are extracted and gathered into a set, which is the associated ozone removal control plan set.

[0050] When the associated ozone removal control scheme set is obtained, the various control indicators involved in each scheme will be analyzed in detail. These control indicators may include but are not limited to the following aspects: equipment operating parameters, such as power level, frequency speed; treatment time, such as the number of hours of work; the amount and concentration of chemical agents used; the range of specific meteorological conditions, such as the upper and lower limits of temperature, the range of humidity, the range of wind speed; and the size and spatial location of the treatment area. By analyzing and comparing the many schemes in the scheme set one by one, the effective range of each control indicator, that is, the trigger interval, is determined. For example, equipment power between 400 watts and 1000 watts can effectively remove ozone; the treatment time between 1.5 hours and 4 hours is the best effect; the temperature may be between 18°C ​​and 32°C, which is conducive to the removal operation. Combining the trigger intervals of these different control indicators, a space with multiple dimensions is constructed, which is the multi-dimensional ozone removal control trigger space.

[0051] The ozone removal control module will have a series of inherent limitations and requirements, which are called ozone removal control constraint information. These constraint information comes from multiple aspects. On the one hand, it comes from the performance and technical specifications of the hardware equipment used by the ozone removal control module. For example, some equipment has maximum power limit, maximum continuous operation time limit, and adaptability to specific environmental conditions. On the other hand, it is based on the constraints set by safety standards and specifications. For example, in order to prevent the equipment from overheating and causing safety hazards, the upper limit of the operating temperature is stipulated; or in order to ensure the safety of operators, restrictions are set on certain operating parameters. In addition, it is also subject to economic cost constraints. For example, in order to control energy consumption costs, there are certain requirements for the energy consumption of the equipment. The way to obtain this constraint information is usually to consult the technical manual of the equipment, the relevant safety specification documents, or read from the preset parameters of the control system.

[0052] After obtaining the ozone removal control constraint information, it is applied to the previously obtained multidimensional ozone removal control trigger space for correction and optimization. For example, the trigger interval of the device power in the multidimensional ozone removal control trigger space was originally 300 watts-1500 watts, but the ozone removal control constraint information indicates that the device power cannot exceed 1000 watts, then it is necessary to correct this power trigger interval to 300 watts-1000 watts. By adjusting and improving the trigger intervals of each dimension in the multidimensional ozone removal control trigger space based on constraint information, removing the parts that do not meet the constraints, so that all trigger intervals are within the constraint range, the space finally generated after correction and optimization is the ozone removal control adjustment constraint space. Through such a correction and optimization process, it is ensured that the generated ozone removal control adjustment constraint space can not only meet the effective ozone removal control requirements, but also fully meet the various constraints of the ozone removal control module, thereby improving the efficiency and effect of ozone removal.

[0053] In a possible implementation, the ozone removal adjustment module 50 further includes: The ozone removal decision is adjusted according to the ozone removal control adjustment constraint space to obtain a second adjusted ozone removal decision; a multi-feature prediction is performed on the second adjusted ozone removal decision based on the ozone removal quality prediction module to obtain an ozone removal quality prediction result of the second adjusted decision; it is determined whether the ozone removal quality prediction result of the second adjusted decision satisfies the predetermined ozone removal quality prediction constraint; if the ozone removal quality prediction result of the second adjusted decision satisfies the predetermined ozone removal quality prediction constraint, the second adjusted ozone removal decision is added to the ozone removal decision adjustment result.

[0054] Specifically, after the ozone removal control adjustment constraint space is determined, the current ozone removal decision is compared with it. For example, the constraint space stipulates the upper and lower limits of the equipment operating power, the range of operating time, the size limit of the treatment area, etc. In the original ozone removal decision, the equipment operating power is too high or too low, the operating time is too long or too short, and the treatment area does not meet the requirements. During the adjustment, these non-compliant parts will be modified according to the specific restrictions in the constraint space. By comprehensively considering the various conditions and restrictions in the constraint space, the key parameters and operating details in the ozone removal decision are revised and optimized, and finally the second adjusted ozone removal decision is obtained.

[0055] The second adjusted ozone removal decision is input into the ozone removal quality prediction module, which will estimate the implementation effect of this decision from multiple feature aspects and obtain the ozone removal quality prediction result of the second adjusted decision.

[0056] First, the predetermined ozone removal quality prediction constraints are a series of standards and conditions set in advance to measure the quality of ozone removal. After obtaining the ozone removal quality prediction results of the second adjustment decision, it is necessary to compare and evaluate each indicator in this result with the predetermined constraints one by one. Through such a detailed and comprehensive comparison and judgment, it can be determined whether the ozone removal quality prediction results of the second adjustment decision meet the expected quality standards, thereby providing an important basis for subsequent decision selection and execution.

[0057] After determining that the ozone removal quality prediction result of the second adjustment decision meets the various quality requirements and standards set in advance (i.e., the predetermined ozone removal quality prediction constraints), the second adjustment ozone removal decision that is considered valid will be included in the ozone removal decision adjustment result. The purpose of this is to concentrate the decisions that meet the requirements, so as to facilitate the subsequent selection of the most appropriate plan to actually perform the ozone removal operation.

[0058] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0060] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An ozone removal control system based on artificial intelligence, characterized in that: The system comprises: A regional monitoring module, which is used to monitor the target area in real time and obtain regional ozone monitoring information and regional meteorological monitoring information; An ozone removal decision module, the ozone removal decision module is used to generate an ozone removal decision based on the regional ozone monitoring information and the regional meteorological monitoring information; an ozone removal quality prediction module, the ozone removal quality prediction module being used to generate an ozone removal quality prediction result based on the ozone removal decision; an ozone removal quality prediction and judgment module, the ozone removal quality prediction and judgment module being used to judge whether the ozone removal quality prediction result satisfies a predetermined ozone removal quality prediction constraint; an ozone removal adjustment module, the ozone removal adjustment module being used to adjust the ozone removal decision according to the predetermined ozone removal quality prediction constraint if the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, and generate an ozone removal decision adjustment result; An ozone removal control module is used to perform ozone removal control on the target area based on the ozone removal decision adjustment result.

2. The system according to claim 1, characterized in that The steps of constructing the ozone removal decision module include: Based on the target area, loading a regional ozone monitoring information sample set, a regional meteorological monitoring information sample set and an ozone removal decision sample set; Taking the regional ozone monitoring information sample set and the regional meteorological monitoring information sample set as input information and the ozone removal decision sample set as output information, the BP neural network is trained, and the ozone removal decision loss coefficient is obtained after each predetermined number of trainings; If the ozone removal decision loss coefficient is less than the ozone removal decision loss threshold, the ozone removal decision module is generated.

3. The system according to claim 1, characterized in that The ozone removal quality prediction module is used to generate an ozone removal quality prediction result based on the ozone removal decision, including: activating an ozone removal rate prediction model of the ozone removal quality prediction module; Based on the ozone removal decision, generating a predicted ozone removal rate according to the ozone removal rate prediction model; Predicting ozone removal energy consumption according to the ozone removal decision to obtain a predicted ozone removal energy consumption coefficient; Predicting ozone removal efficiency according to the ozone removal decision to generate predicted ozone removal efficiency; The ozone removal quality prediction result is obtained according to the predicted ozone removal rate, the predicted ozone removal energy consumption coefficient and the predicted ozone removal efficiency.

4. The system according to claim 3, characterized in that The ozone removal energy consumption is predicted according to the ozone removal decision to obtain a predicted ozone removal energy consumption coefficient, including: Perform ozone removal energy consumption retrieval according to the ozone removal decision to obtain multiple sample ozone removal energy consumption coefficients; Performing deviation evaluation on the ozone removal energy consumption coefficients of the plurality of samples to generate an ozone removal energy consumption deviation evaluation value; Determining whether the ozone removal energy consumption deviation evaluation value is less than the ozone removal energy consumption deviation evaluation threshold; If the ozone removal energy consumption deviation evaluation value is less than the ozone removal energy consumption deviation evaluation threshold, a centralized value evaluation is performed on the multiple sample ozone removal energy consumption coefficients to generate the predicted ozone removal energy consumption coefficient.

5. The system according to claim 1, wherein: If the ozone removal quality prediction result satisfies the predetermined ozone removal quality prediction constraint, the ozone removal control module performs ozone removal control on the target area based on the ozone removal decision.

6. The system according to claim 1, wherein: The ozone removal adjustment module is used to adjust the ozone removal decision according to the predetermined ozone removal quality prediction constraint if the ozone removal quality prediction result does not meet the predetermined ozone removal quality prediction constraint, and generate an ozone removal decision adjustment result, including: Perform ozone removal control feature backtracking based on the regional ozone monitoring information and the regional meteorological monitoring information to obtain an ozone removal control adjustment constraint space; Adjusting the ozone removal decision according to the ozone removal control adjustment constraint space to obtain a first adjusted ozone removal decision; Performing multi-feature prediction on the first adjusted ozone removal decision based on the ozone removal quality prediction module to obtain an ozone removal quality prediction result of the first adjusted decision; determining whether the ozone removal quality prediction result of the first adjustment decision satisfies the predetermined ozone removal quality prediction constraint; If the ozone removal quality prediction result of the first adjustment decision satisfies the predetermined ozone removal quality prediction constraint, the first adjusted ozone removal decision is added to the ozone removal decision adjustment result.

7. The system according to claim 6, characterized in that Ozone removal control feature backtracking is performed according to the regional ozone monitoring information and the regional meteorological monitoring information to obtain ozone removal control adjustment constraint space, including: Perform ozone removal control feature association retrieval based on the regional ozone monitoring information and the regional meteorological monitoring information to obtain an associated ozone removal control scheme set; Identifying an ozone removal control triggering interval according to the associated ozone removal control scheme set to obtain a multi-dimensional ozone removal control triggering space; Obtaining ozone removal control constraint information of the ozone removal control module; The multi-dimensional ozone removal control trigger space is corrected and optimized according to the ozone removal control constraint information to generate the ozone removal control adjustment constraint space.

8. The system according to claim 6, characterized in that Determining whether the ozone removal quality prediction result of the first adjustment decision satisfies the predetermined ozone removal quality prediction constraint includes: adjusting the ozone removal decision according to the ozone removal control adjustment constraint space to obtain a second adjusted ozone removal decision; Performing multi-feature prediction on the second adjusted ozone removal decision based on the ozone removal quality prediction module to obtain a second adjusted decision ozone removal quality prediction result; determining whether the ozone removal quality prediction result of the second adjustment decision satisfies the predetermined ozone removal quality prediction constraint; If the ozone removal quality prediction result of the second adjustment decision satisfies the predetermined ozone removal quality prediction constraint, the second adjusted ozone removal decision is added to the ozone removal decision adjustment result.

Images