Smart park management system and method

Abstract

The invention relates to the technical field of air quality prediction, in particular to a smart park management system and method, and the method comprises the steps: collecting environment parameters and reference data of a park in real time; evaluating the true degree of the environmental parameters at the current moment, determining the number of historical environmental parameters required for calculating the prediction error of the environmental parameters at the current moment, and then calculating the prediction error of the environmental parameters at the current moment; predicting the environmental parameter at the next moment of the current moment by combining the actual value of the environmental parameter at the current moment and the prediction error; and judging whether the prediction result is abnormal or not so as to monitor and early warn air pollution of the park, thereby realizing air quality management of the park. According to the method, interference data can be effectively discriminated, the accuracy of air quality prediction is improved, and reliable decision support is provided for park management.

Description

Technical Field

[0001] The present invention relates to the technical field of air quality prediction. Specifically, it relates to a smart park management system and method. Background Art

[0002] In the context of the rapid development of urbanization and industrialization today, air quality issues have received increasing attention. Among them, (fine particulate matter), as a key indicator of air quality, its pollution status has a profound impact on public health. As a product of the combination of modern management concepts and technologies, smart parks have the responsibility to create a green and healthy environment. Among the many management tasks in smart parks, air quality monitoring is a very important part. Traditional air pollution monitoring methods mostly rely on manual analysis and data processing means, which have the disadvantage of strong lag and are difficult to respond promptly to changes in air quality.

[0003] A prediction algorithm is an algorithm that predicts future data based on the historical change trend of data. Usually, the error between the predicted value and the actual value at the previous moment of the current moment is multiplied by an error factor as the prediction error of the environmental parameters at the current moment, and then the prediction error of the environmental parameters at the current moment and the actual value at the current moment are used to predict the data at the next moment. This prediction algorithm reflects the prediction effect at the previous moment through the prediction error, and continuously adjusts the predicted value at the subsequent moment based on the prediction error, so that the prediction process can continuously adapt to the change trend and fluctuation of the data, and gradually improve the prediction accuracy. For example, the Chinese patent document with the publication number CN118673387B discloses a high-temperature warning method and system for a pulverized coal gasification burner. This method obtains all combined sequences, calculates the preference degree of each combined sequence, and uses the change trend of the combined sequence with the largest preference degree as the historical trend when the prediction algorithm predicts data, and predicts the temperature data at the next moment based on this, so as to realize the temperature warning of the burner.

[0004] However, when using the prediction algorithm in the above technical solution to predict the data of the park, due to the data is vulnerable to signal interference during the acquisition and transmission process, resulting in possible incorrect transmission of the collected data and the appearance of untrue data. If the data at the previous moment is untrue, then the predicted error of the environmental parameters at the current moment calculated is inaccurate, and finally the prediction result at the next moment has a large deviation, and it is impossible to issue a timely and accurate warning of the air pollution situation. Summary of the Invention

[0005] To solve the problem that during the air quality monitoring in the park, due to the data being affected by signal interference and showing untrue situations, which affects the accuracy of the prediction results of the prediction algorithm, resulting in the inability to timely and accurately monitor and warn of air pollution in the park, the present invention proposes an intelligent park management method, including: Collecting the environmental parameters and their reference data of the park in real time; both the environmental parameters and the reference data are data; taking each moment as the current moment and evaluating the authenticity of the environmental parameters at the current moment: , being the authenticity of the environmental parameters at the th current moment, being the suspicious degree of the environmental parameters at the th current moment, and the suspicious degree is used to reflect the magnitude of the influence of noise on the environmental parameters, being the total number of reference data of the environmental parameters at any moment, and being respectively the environmental parameters at the th current moment and the th reference data of this environmental parameter, being the variance of all reference data of the environmental parameters at the th current moment, being a hyperparameter, being the natural exponential function, being the absolute value symbol; Determining the number of historical environmental parameters required when calculating the prediction error of the environmental parameters at the current moment according to the authenticity of the environmental parameters at the current moment, and selecting historical environmental parameters according to the required number of historical environmental parameters to calculate the prediction error of the environmental parameters at the current moment; Through the prediction algorithm, predicting the environmental parameters at the next moment of the current moment according to the prediction error and the actual value of the environmental parameters at the current moment, and monitoring and warning of air pollution in the park according to the prediction results to achieve air quality management in the park.

[0006] The above technical solution not only obtains direct environmental parameters but also collects reference data, which provides a multi-dimensional information source for subsequent evaluation of the reliability of data. Multiple reference data helps to comprehensively and accurately grasp the real air quality situation. Further, by comprehensively considering factors such as the differences between the environmental parameters at the current moment and multiple reference data, the suspicious degree of the environmental parameters at the current moment itself, and the dispersion degree of the reference data, the true reliability degree of the environmental parameters at the current moment is accurately quantified. It can effectively identify the untrue environmental parameters caused by interference, provide a reliable basis for subsequent predictions, and reduce the risk of prediction deviation caused by incorrect data. Further, considering that when calculating the prediction error of the environmental parameters at the current moment using traditional methods, a fixed number of historical environmental parameters are usually used for calculation. However, in the case where the environmental parameters may be untrue, this method will introduce a large error. Here, the number of selected historical environmental parameters is dynamically adjusted according to the true degree of the environmental parameters, and historical environmental parameters are more reasonably used to accurately calculate the prediction error, improving the accuracy and flexibility of the prediction error calculation, so that the subsequent prediction process can more accurately adapt to the actual situation of the data. Further, using the corrected prediction error and the actual value, combined with the mechanism of the HTFE prediction algorithm, the environmental parameters at the next moment of the current moment are predicted, thereby realizing the forward-looking monitoring and early warning of the air quality in the park. It can overcome the problem of inaccurate prediction caused by data interference, more timely and accurately predict the future air quality in the park, and provide reliable decision-making support for the management of the park.

[0007] Preferably, the calculation formula for the prediction error of the environmental parameters at the current moment is:

[0008] In the formula, is the prediction error of the th environmental parameter at the current moment, is the number of historical environmental parameters required when calculating the prediction error of the th environmental parameter at the current moment, and are respectively the predicted value and the actual value of the th historical environmental parameter required when calculating the prediction error of the th environmental parameter at the current moment, is a uniformly preset error factor.

[0009] Compared with simply using a fixed method or a single data point to calculate the prediction error, this calculation method based on multiple historical environmental parameters can more comprehensively reflect the accuracy and stability of historical predictions, and can effectively reduce the impact of individual abnormal historical data or inaccurate historical predictions on the calculation of the current prediction error, thereby improving the reliability of the prediction error.

[0010] Preferably, the suspiciousness of the environmental parameters at the current moment satisfies the following relational expression: ; In the formula, is the suspiciousness of the th environmental parameter at the current moment, and are respectively the predicted value and the actual value of the th environmental parameter at the current moment, is the average value of the local data segment of the th environmental parameter at the current moment, is the variance of all environmental parameters except the th environmental parameter in the local data segment of the th environmental parameter at the current moment, is the absolute value symbol, is the normalization operation.

[0011] The above technical solution helps to more accurately screen out the environmental parameters that may be affected by abnormal factors such as signal interference in the complex park environment monitoring scenario by accurately evaluating the suspiciousness of each environmental parameter at the current moment.

[0012] Preferably, the local data segment of the environmental parameter is determined according to the following method: The th environmental parameter at the current moment and the environmental parameters at moments before the th current moment form the local data segment of the th environmental parameter at the current moment, is a preset quantity.

[0013] Preferably, the number of historical environmental parameters required when calculating the prediction error of the environmental parameter at the current moment satisfies the following relational expression: , is the number of historical environmental parameters required when calculating the prediction error of the th environmental parameter at the current moment, is the authenticity of the th environmental parameter at the current moment, is the initial preset value, is the ceiling symbol, is the natural exponential function.

[0014] The above technical solution dynamically adjusts the number of historical environmental parameters according to the authenticity of the environmental parameters at the current moment, so as to calculate the prediction error of the environmental parameters at the current moment more accurately. When the reliability of the environmental parameters at the current moment is high, the interference of historical environmental parameters is reduced, highlighting the role of the environmental parameters at the current moment; when the data reliability is low, the historical environmental reference data is fully utilized to correct possible deviations. This targeted adjustment makes the calculation of the prediction error more in line with the actual situation and improves the accuracy of the prediction error.

[0015] Preferably, the method for real-time collecting the environmental parameters and their reference data in the park is as follows: An electrochemical sensor A is set in the park, and several electrochemical sensors are arranged around A; The environmental parameters collected by A at each moment are determined as the environmental parameters of the park at each moment, and the environmental parameters collected by several electrochemical sensors around A at the same moment are used as the reference data of the environmental parameters of the park at each moment.

[0016] The above technical solution collects data from multiple sensors and takes sensor A as the core for data collection. Based on these reference data, the authenticity of the data collected by sensor A can be judged more accurately, and the data that may be incorrect can be identified.

[0017] Preferably, the method for monitoring and warning the air pollution in the park according to the prediction result is as follows: The predicted value of the environmental parameters at the next moment of the current moment and the environmental parameters at the previous moments of the environmental parameters at the next moment of the current moment are constructed into a data set, where is a preset value; Based on this data set, the value of the environmental parameters at the next moment of the current moment is calculated. If the value is greater than the preset threshold, the environmental parameters at the next moment of the current moment are abnormal data, and a warning is issued; if the value is not greater than the preset threshold, the environmental parameters at the next moment of the current moment are normal data, and no warning is issued.

[0018] The present invention also provides an intelligent park management system, which includes a memory and a processor. A computer program is stored on the memory, and the processor executes the computer program to implement the steps of any one of the management methods.

[0019] The present invention has the following effects: By collecting the environmental parameters and reference data of the park in real time, the present invention accurately evaluates the authenticity of the environmental parameters by using specific formulas, effectively identifies interference data, and lays a foundation for accurate prediction. At the same time, the number of historical environmental parameters is dynamically adjusted according to the authenticity of the environmental parameters, and the prediction error is calculated, improving the accuracy and flexibility of the prediction error calculation. Finally, combined with the BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown by way of example and not limitation, and the same or corresponding reference numerals indicate the same or corresponding parts, wherein: Figure 1 is a schematic flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] The following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings.

[0023] Referring to Figure 1 , a smart park management method provided by the present invention includes steps S1 - S6: S1: Collect the environmental parameters of the park and their reference data in real time.

[0024] Both the environmental parameters and reference data referred to in the present invention are data. The specific collection methods for environmental parameters and reference data are as follows: Electrochemical sensors are set in the park , and within a range of 1 centimeter around , 4 electrochemical sensors are evenly arranged, and they are made to collect environmental parameters synchronously at the same time. And at each moment, the environmental parameters collected by are determined as the environmental parameters of the park at that moment, and the environmental parameters collected by the other 4 electrochemical sensors are used as the reference data of the environmental parameters of the park at that moment.

[0025] This method constructs a mutually verified data system. Since the environmental parameters at different locations within the park should have a certain correlation in adjacent areas, this setting can utilize the data consistency of multiple sensors to verify the reliability of environmental parameters. When one of the sensors is interfered with and causes abnormal environmental parameters, the data of other sensors can assist in judging the abnormality, providing multi-dimensional references for subsequent data processing. For example, if the difference between the collected environmental parameters and those collected by the other 4 reference sensors is too large, the accuracy of the environmental parameters can be suspected.

[0026] The acquisition frequency of all the above sensors is set to once every five minutes, and the acquisition duration each time is 24 hours. To analyze the true degree of environmental parameters more accurately, data can be collected one day in advance. The main purpose of the environmental parameters collected on this additional day is to assist in judging the reliability of other environmental parameters. By adding the reference data for this day, more dimensional information can be provided for the analysis of the true degree.

[0027] In the current scenario, the method of using traditional prediction algorithms to predict the environmental parameters at the next moment of the current moment is as follows: S11: Judge the historical trend. Specifically, the method is: Taking the th current moment as an example, obtain the environmental parameters of the th current moment and the previous 100 environmental parameters. If these 100 environmental parameters show a monotonically increasing trend, the historical trend of the environmental parameters of the th current moment is a growth type. If these 100 environmental parameters show a monotonically decreasing trend, the historical trend of the environmental parameters of the th current moment is a decay type. Except for monotonically increasing and monotonically decreasing, other situations are irregular types. If the historical trend is a growth type, then , if it is a decay type, then , if the historical trend is an irregular type, then , where are all adjustment parameters of the prediction algorithm.

[0028] Take the average value of these 100 environmental parameters as the predicted value of the environmental parameters of the th current moment, and take the maximum and minimum values of these 100 environmental parameters as the maximum predicted value and the minimum predicted value of the environmental parameters of the th current moment respectively. Then, from and constitute the The prediction range of the environmental parameters at the current moment; S12: Predict the environmental parameters at the next moment of the th current moment. Specifically, it includes the following steps: S121: Calculate the preliminary predicted value of the environmental parameters at the next moment of the th current moment.

[0029]

[0030] In the formula, represents the next moment of the th current moment, is the preliminary predicted value of the environmental parameters at the next moment of the th current moment, is the actual value of the environmental parameters at the th current moment, is the preset error factor, is the prediction error of the environmental parameters at the th current moment.

[0031] Among them, the method for obtaining the prediction error of the environmental parameters at the th current moment is:

[0032] In this formula, is the predicted value of the environmental parameters at the th current moment, is the actual value of the environmental parameters at the th current moment, is the preset error factor.

[0033] S122: Obtain the prediction range of the environmental parameters at the next moment of the th current moment.

[0034] Specifically, it is based on the preliminary predicted value of the environmental parameters at the next moment of the th current moment, the prediction range of the environmental parameters at the th current moment, and the influence factor of the preset prediction error range, to obtain the prediction range of the environmental parameters at the next moment of the ; ; In this formula, is the The lower limit of the prediction range of the environmental parameters at the next moment of the current moment, is the lower limit of the prediction range of the environmental parameters at the next moment of the th current moment. And constitute the prediction range of the environmental parameters at the next moment of the th current moment, (empirical value).

[0035] S123: Adjust the prediction range of the environmental parameters at the next moment of the th current moment.

[0036] To ensure that the final predicted value of the environmental parameters at the next moment of the th current moment can cover the actual value of the environmental parameters at the th current moment , the following settings are made: When is less than , let be equal to , when is greater than , let be equal to ; S124: Calculate the final predicted value of the environmental parameters at the next moment of the th current moment according to the prediction range of the environmental parameters at the next moment of the th current moment.

[0037] Specifically, the calculation formula is: ; In the formula, is the final predicted value of the environmental parameters at the th current moment.

[0038] S13: Keep repeating to achieve real-time monitoring of environmental parameters.

[0039] Specifically, for the next moment of the next moment of the th current moment, predict the environmental parameters at this moment in the same way as in steps S121 - S124. In this way, real-time monitoring of the environmental parameters in the park can be achieved.

[0040] Thus, it can be seen that in the process of predicting the environmental parameters in the park, due to the traditional The prediction algorithm usually multiplies the error between the predicted value and the actual value of the environmental parameters at the previous moment of the current moment by an error factor as the prediction error of the environmental parameters at the current moment, and then uses the prediction error of the environmental parameters at the current moment and the actual value of the environmental parameters at the current moment to predict the environmental parameters at the next moment of the current moment. It can be seen that three moments are involved here, namely the current moment, the previous moment of the current moment, and the next moment of the current moment, and it is the next moment of the current moment that needs to be predicted, and the predicted value of the next moment is determined by the data at the current moment. Therefore, in order to prevent the data at the current moment from being interfering data and causing the predicted value of the next moment to be inaccurate.

[0041] By deeply analyzing the change characteristics of the environmental parameters at the current moment, the present invention can accurately judge the authenticity of the environmental parameters at each current moment (the first moment immediately adjacent to each moment to be predicted), effectively identify and properly handle possible incorrect data. And, based on the authenticity of the environmental parameters at each current moment, dynamically determine the number of historical environmental parameters required when calculating the prediction error of the environmental parameters at the current moment. Selecting historical data based on the calculated number can accurately calculate the prediction error of the environmental parameters at the current moment. In this way, when applying the prediction algorithm, it can effectively avoid the interference of untrue data on the prediction result, effectively avoid error accumulation, improve the accuracy of the prediction result, and ensure the real-time monitoring and early warning ability of the air pollution situation in the park.

[0042] S2: Evaluate the suspicious degree of the environmental parameters at the current moment.

[0043] This step aims to analyze the numerical values of the environmental parameters in the park to determine the suspicious degree of the environmental parameters at the current moment. The greater the suspicious degree of the environmental parameters at the current moment, the less real the environmental parameters at the current moment are, and the more likely it is to affect the accuracy of the prediction result of the environmental parameters at the next moment.

[0044] Because untrue environmental parameters often show prominent and abnormal numerical performances, analyzing the suspicious degree of the environmental parameters at the current moment aims to determine the degree to which the environmental parameters at the current moment are affected by noise.

[0045] Specifically, when analyzing the suspicious degree of the environmental parameters at the current moment, the following rules are followed: If the environmental parameters at the current moment have a greater difference compared with the surrounding environmental parameters, it indicates that the suspicious degree of the environmental parameters at the current moment is greater. If the difference between the predicted value and the actual value of the environmental parameters at the current moment is greater, it indicates that the suspicious degree of the environmental parameters at the current moment is greater. If the numerical performances of the surrounding environmental parameters of the environmental parameters at the current moment are more unified, it indicates that the suspicious degree of the environmental parameters at the current moment is greater.

[0046] To implement the above logic, in this step, a local data segment is set for the environmental parameters at each current moment to analyze the suspiciousness of the environmental parameters at each current moment.

[0047] In one embodiment, the environmental parameters at the th current moment, and the environmental parameters at the moments before the th current moment are used to form the local data segment of the environmental parameters at the th current moment. is a preset quantity, and the empirical value is 100.

[0048] This step is mainly to evaluate and screen the quality of the environmental parameters during the prediction process of the environmental parameters, find out the environmental parameter data that may be greatly affected by noise, and then more reasonably optimize the calculation of the prediction error in the subsequent prediction process to improve the accuracy of the prediction.

[0049] In one embodiment, the suspiciousness of the environmental parameters at the current moment satisfies the following relational expression:

[0050] In the formula, is the suspiciousness of the environmental parameters at the th current moment. and are respectively the predicted value and the actual value of the environmental parameters at the th current moment. is the average value of the local data segment of the environmental parameters at the th current moment (the average value of the actual values of all environmental parameters within the local data segment). is the variance of all environmental parameters except the environmental parameters at the th current moment within the local data segment of the environmental parameters at the th current moment. is the absolute value symbol. is the normalization operation.

[0051] In this formula, quantifies the difference between the actual value of the environmental parameters at the th current moment and the average value of its local data segment. The larger this value is, the greater the suspiciousness of the environmental parameters at the th current moment. quantifies the numerical difference between the actual value and the predicted value of the environmental parameters at the th current moment. The larger this value is, the greater the difference between the actual value and the predicted value of the environmental parameters at the The greater the difference between the actual value of the environmental parameters at a current moment and the average value of its local data segment, the higher the credibility, which can also indicate that the greater the possibility of untrue numerical changes in the environmental parameters at the current moment. Then, the higher the degree of suspicion of the environmental parameters at the current moment. The variance quantifies the numerical difference of the environmental parameters other than the environmental parameters at the current moment within the local data of the environmental parameters at the current moment. The smaller this value is, the more uniform the numerical performance of the other environmental parameters is. Then, it can be explained that the greater the numerical difference between the actual value and the predicted value of the environmental parameters at the current moment is more likely to be due to the influence of untrue numerical values (rather than the real changes of environmental parameters). It can be explained that the higher the credibility of the possibility of untrue numerical changes in the environmental parameters at the current moment, that is, the higher the degree of suspicion of the environmental parameters at the current moment. current moment.

[0052] Based on the accurate assessment of the degree of suspicion of environmental parameters, the prediction error calculation can be more reasonably optimized in the subsequent prediction process. For environmental parameters with a high degree of suspicion, corresponding measures can be taken to reduce their negative impact on the prediction results; for reliable environmental parameters, their role in prediction can be better exerted. This effectively improves the accuracy of prediction, reduces the prediction error caused by data quality problems, and enhances the reliability and stability of the entire prediction model.

[0053] S3: Optimize the degree of suspicion of the environmental parameters at the current moment based on the similarity between the environmental parameters at the current moment and the reference data to obtain the authenticity degree of the environmental parameters at the current moment.

[0054] Through the previous steps, the degree of suspicion of the environmental parameters at each current moment is obtained. This index is analyzed based on the environmental parameters collected by the electrochemical sensors in the park However, it cannot distinguish between untrue environmental parameters and abnormally large environmental parameters actually generated in the park. That is, only analyzing the degree of suspicion of the environmental parameters at each current moment cannot exclude the influence of abnormally large environmental parameters actually generated in the park.

[0055] The differences between the untrue environmental parameters and the abnormally large environmental parameters actually generated in the park in the corresponding scenarios of this solution are as follows: when the abnormally large environmental parameters are actually generated in the park, at this time, whether it is the electrochemical sensor A or the 4 electrochemical sensors around it, abnormally large environmental parameters will be collected. However, the untrue environmental parameters do not have this characteristic, and they will only appear in the environmental parameters collected by sensor A.

[0056] Therefore, in this step, the similarity between the environmental parameters collected by the electrochemical sensor A and the environmental parameters collected by the 4 electrochemical sensors around it will be analyzed, and the suspicious degree of the environmental parameters at each current moment will be optimized to obtain the authenticity degree of the environmental parameters at each current moment.

[0057] When analyzing the authenticity degree of the environmental parameters at each current moment, if the similarity between the environmental parameters at each current moment and the reference data is weaker, it can be explained that the authenticity degree of the environmental parameters at this current moment is smaller, and vice versa.

[0058] However, when analyzing the similarity between the environmental parameters at each current moment and the reference data, there may be a situation where the environmental parameters at a certain current moment and the reference data are both affected by signal interference during the transmission process. At this time, the environmental parameters at this current moment and the reference data have a high similarity, and the environmental parameters at this current moment are more likely to be recognized as real environmental parameters, affecting the accuracy of subsequent data prediction. Therefore, after analyzing the suspicious degree of the environmental parameters at each current moment in this solution, the similarity between the environmental parameters at each current moment and the reference data is further analyzed to optimize the suspicious degree of the environmental parameters at this current moment and obtain the authenticity degree of the environmental parameters at this current moment, which can well solve this problem.

[0059] In one embodiment, the calculation formula for the authenticity degree of the environmental parameters at the current moment is:

[0060] In this formula, is the authenticity degree of the environmental parameters at the th current moment, is the suspicious degree of the environmental parameters at the th current moment. The suspicious degree is used to reflect the magnitude of the influence of noise on the environmental parameters. is the total number of reference data of the environmental parameters at any moment. and are respectively the environmental parameters at the th current moment and the th reference data of this environmental parameter. is the The variance of all reference data of the environmental parameters at the current moment, is a hyperparameter, and the empirical value is taken as , and its existence is to prevent the situation where it is 0 from occurring.

[0061] In this formula, reflects the suspiciousness of the environmental parameters at the -th current moment. The smaller this value is, the smaller the possibility that the environmental parameters at the -th current moment are untrue data caused by signal interference, and the greater the authenticity of the environmental parameters at the -th current moment.

[0062] In this formula, reflects the difference between the environmental parameters at the -th current moment and the -th reference data. The smaller this value is, the greater the similarity between the environmental parameters at the -th current moment and the reference data, and the greater the authenticity of the environmental parameters at the -th current moment. reflects the numerical difference situation of all reference data of the environmental parameters at the -th current moment. The smaller this value is, the smaller the numerical difference of all reference data of the environmental parameters at the -th current moment, and the smaller the possibility that there are untrue environmental parameters in all reference data of the environmental parameters at the -th current moment. Then, the stronger the reliability of all reference data of the environmental parameters at the -th current moment, the greater the reference value, which can further indicate that the difference between the environmental parameters at the -th current moment and its -th reference data is smaller with higher credibility, that is, the similarity between the environmental parameters at the -th current moment and the reference data is greater with higher credibility. Then, the greater the authenticity of the environmental parameters at the -th current moment.

[0063] In this formula, comprehensively quantifies the similarity between the environmental parameters at the -th current moment and the reference data. The larger this value is, the smaller the possibility that the environmental parameters at the -th current moment are untrue data caused by signal interference, and the greater the authenticity of the environmental parameters at the -th current moment.

[0064] Through this formula, the suspicious degree of the environmental parameters at the current moment can be effectively optimized, so as to accurately obtain the true degree of the environmental parameters at the current moment, which helps to improve the accuracy of subsequent data prediction and also solve the misjudgment problem caused by signal interference during the transmission process.

[0065] S4: Select historical environmental data according to the true degree of the environmental parameters at the current moment to calculate the prediction error of the environmental parameters at the current moment.

[0066] Through the previous step analysis, the true degree of the environmental parameters at each current moment is obtained. In this step, the number of historical environmental parameters used to calculate the prediction error of the environmental parameters at each current moment will be determined based on this index, and then the prediction error of the environmental parameters at each current moment will be calculated.

[0067] In one embodiment, the number of historical environmental parameters required to calculate the prediction error of the environmental parameters at the current moment satisfies the following relational formula:

[0068] In this formula, is the number of historical environmental parameters required to calculate the prediction error of the environmental parameters at the th current moment, is the true degree of the environmental parameters at the th current moment, is the initial preset value, that is, the number of historical environmental parameters required to calculate the prediction error of the environmental parameters at any moment, The empirical value of is is the ceiling symbol. Since even if the true degree of the environmental parameters at a certain current moment is particularly large, at least one historical environmental parameter is required to calculate its predicted value, is the natural exponential function.

[0069] From this formula, it can be seen that when determining the number of historical environmental parameters used to calculate the prediction error of the environmental parameters at each current moment, the smaller the true degree of the environmental parameters at each current moment, the lower the authenticity of the environmental parameters at this current moment, and the more likely it belongs to untrue environmental parameters. Then, when calculating the prediction error of the environmental parameters at this current moment, more historical environmental parameter data is required to eliminate and smooth the environmental parameters at the current moment as much as possible, so as to obtain a more accurate prediction error of the environmental parameters at the current moment.

[0070] Further, select historical environmental parameters before the environmental parameters at the th current moment for subsequent calculation of the The prediction error of the environmental parameters at the current moment.

[0071] In one embodiment, the calculation formula for the prediction error of the environmental parameters at the current moment is:

[0072] In this formula, is the prediction error of the th environmental parameter at the current moment, is the number of historical environmental parameters required when calculating the prediction error of the th environmental parameter at the current moment, and are respectively the predicted value and the actual value of the th historical environmental parameter required when calculating the prediction error of the th environmental parameter at the current moment, is a uniformly preset error factor, , and this value is an empirical value.

[0073] This formula is obtained by summing and averaging the absolute values of the differences between the predicted values of historical environmental parameters and the actual values , and can effectively measure the deviation degree between the predicted value and the actual value. Among them, is dynamically determined according to the authenticity of the environmental parameters at the current moment, so that when facing environmental parameters with different authenticity, the amount of historical data used to calculate the prediction error is more reasonable. Secondly, multiplying by the uniformly preset error factor can appropriately adjust the overall prediction error to adapt to the error evaluation criteria in a specific scenario.

[0074] Generally speaking, this formula comprehensively considers the dynamic selection of historical environmental parameters and the uniformly preset error factor, and can accurately reflect the error magnitude between the predicted value and the actual value of the environmental parameters at the current moment, providing a reliable quantitative index for subsequent analysis, decision-making, and model optimization based on the prediction error.

[0075] S5: Through The prediction algorithm predicts the environmental parameters at the next moment of the current moment based on the prediction error and the actual value of the environmental parameters at the current moment.

[0076] The present invention mainly optimizes the method for obtaining the prediction error of the environmental parameters at each current moment in the process of predicting the environmental parameters at the next moment of the current moment by the traditional prediction algorithm to improve the prediction accuracy, and the remaining steps are the same as those of the traditional The prediction process of the prediction algorithm remains the same.

[0077] Specifically, the prediction error of the environmental parameters at the th current moment is accurately calculated according to steps S2 - S4. Then, the preliminary predicted value of the environmental parameters at the next moment of the th current moment is calculated:

[0078] In the formula, is the preliminary predicted value of the environmental parameters at the next moment of the th current moment, is the actual value of the environmental parameters at the th current moment, is the preset error factor, is the prediction error of the environmental parameters at the th current moment.

[0079] Then, the final predicted value at the th current moment is obtained in the same way as in steps S122 - S124.

[0080] In summary, the core of the present invention lies in improving the traditional prediction algorithm, especially in obtaining the prediction error of the environmental parameters at each current moment. Through this optimization, the entire prediction algorithm becomes more adaptable and accurate. In practical applications, the environmental parameters at different moments may be affected by various complex factors, and the accurate acquisition of the prediction error is crucial for improving the overall prediction reliability. This solution fills the gap in this aspect of the traditional prediction algorithm.

[0081] S6: Monitor and give early warning of the air pollution in the park according to the prediction results to achieve air quality management in the park.

[0082] In one embodiment, the present invention adopts a more detailed and comprehensive method to evaluate the air pollution situation in the park.

[0083] First, for each current moment, the predicted value of the environmental parameters at the next moment and the environmental parameters at M moments before the next moment are integrated to construct a specific data set, where M is set to 100 (empirical value). Subsequently, based on the constructed data set, the Z-Score value of the environmental parameters at the next moment is further calculated. The Z-Score value is a standardized statistic that can clearly indicate the deviation degree between the data point and the mean value of the data set. Next, according to the calculated Z-Score value, a judgment is made: If the Z-Score value of the environmental parameters at the next moment is greater than the pre-set threshold, this means that there is a significant deviation in the environmental parameters at the next moment compared with the historical data, which is very likely to represent an abnormal air pollution situation. At this time, the system issues a warning signal. On the contrary, if the Z-Score value of the environmental parameters at the next moment is not greater than the pre-set threshold, it indicates that the environmental parameters at the next moment are within the reasonable range of historical data fluctuations, that is, the environmental parameters at the next moment belong to normal data, and the system will not issue a warning, indicating that the current air pollution situation in the park is relatively stable.

[0084] Through such a rigorous and scientific monitoring and warning mechanism, the prediction results can be fully utilized to timely capture the abnormal changes in the air pollution in the park, providing strong support for the air quality management in the park.

[0085] In addition, the present invention also provides a smart park management system, the management system includes a memory and a processor, a computer program is stored on the memory, and the processor executes the computer program to implement the steps of the management method.

[0086] In the description of this specification, the meanings of "a plurality of" and "several" are at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0087] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art will think of many changes, alterations and alternative ways without departing from the spirit and idea of the present invention. It should be understood that various alternative solutions to the embodiments of the present invention described herein can be adopted in the process of practicing the present invention.

Claims

1. A smart park management method, characterized in that: include: Collect the park's environmental parameters and reference data in real time; The environmental parameters and reference data are Data; take each moment as the current moment and evaluate the authenticity of the environmental parameters at the current moment: , For the The authenticity of the environmental parameters at the current moment, For the The suspicious degree of the environmental parameters at the current moment is used to reflect the extent to which the environmental parameters are affected by noise. is the total number of reference data of environmental parameters at any time, and Respectively The current environmental parameter and the Reference data, For the The variance of all reference data of the environmental parameters at the current moment, is a hyperparameter, is the natural exponential function, is the absolute value symbol; Determine the number of historical environmental parameters required when calculating the prediction error of the environmental parameters at the current moment according to the authenticity of the environmental parameters at the current moment, and select historical environmental parameters according to the required number of historical environmental parameters to calculate the prediction error of the environmental parameters at the current moment; pass The prediction algorithm predicts the environmental parameters at the next moment based on the prediction error and actual value of the environmental parameters at the current moment, and monitors and warns of air pollution in the park based on the prediction results to achieve air quality management of the park.

2. The smart park management method according to claim 1, characterized in that: The calculation formula for the prediction error of the environmental parameters at the current moment is: , where For the The prediction error of the environmental parameters at the current moment, To calculate the The number of historical environmental parameters required to calculate the prediction error of the current environmental parameters, and In the calculation of The prediction error of the environmental parameters at the current moment is required The predicted and actual values ​​of historical environmental parameters, is a uniformly preset error factor.

3. The smart park management method according to claim 1, characterized in that: The suspicious degree of the environmental parameters at the current moment satisfies the following relationship: , where For the The suspiciousness of the environmental parameters at the current moment, and Respectively The predicted and actual values ​​of the environmental parameters at the current moment, For the The average value of the local data segment of the environmental parameters at the current moment, For the The local data segment of the environmental parameters at the current moment is The variance of all environmental parameters except the current environmental parameters, is the absolute value symbol, It is a normalization operation.

4. The smart park management method according to claim 3, characterized in that: The local data segment of the environmental parameters is determined in the following way: The current environmental parameters, and before the current moment The environmental parameters at each moment constitute the A local data segment of the current environmental parameters, The preset quantity.

5. The smart park management method according to claim 1, characterized in that: The number of historical environmental parameters required to calculate the prediction error of the current environmental parameters satisfies the following relationship: , To calculate the The number of historical environmental parameters required to calculate the prediction error of the current environmental parameters, For the The authenticity of the environmental parameters at the current moment, is the initial default value, is the round-up symbol, is a natural exponential function.

6. The smart park management method according to claim 1, characterized in that: The method for real-time collection of the park's environmental parameters and their reference data is: Installing electrochemical sensors in the park , and in Several electrochemical sensors are arranged around the periphery; Will The environmental parameters collected at each moment are determined as the environmental parameters of the park at each moment. The environmental parameters collected by several electrochemical sensors around the park serve as reference data for the environmental parameters of the park at each moment.

7. The smart park management method according to claim 1, characterized in that: The method of monitoring and early warning the air pollution in the park based on the prediction results is: The predicted value of the environmental parameter at the next moment of the current moment is compared with the previous value of the environmental parameter at the next moment of the current moment. The environmental parameters at each moment are constructed into a data set. is the default value; Calculate the environmental parameters of the next moment based on the data set Value, if If the value is greater than the preset threshold, the environmental parameters at the next moment from the current moment are abnormal data, and an early warning is issued; If the value is not greater than the preset threshold, the environmental parameters at the next moment from the current moment are normal data and no warning is issued.

8. A smart park management system, characterized in that: The management system comprises a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the steps of the management method according to any one of claims 1 to 7.

Images