Environment monitoring system based on big data

Abstract

The invention relates to the technical field of environment monitoring, in particular to an environment monitoring system based on big data, which comprises a data acquisition module used for collecting current environment data from various sensors; the data preprocessing module is used for correcting and compensating the environmental data collected by the sensor under the extreme temperature condition; the data analysis module is used for analyzing the preprocessed environment data by adopting an environment prediction model and predicting an environment change trend; and the influence evaluation module is used for evaluating the influence degree of environmental change on the ecological system by utilizing a simulation technology. According to the method, the complex interaction among a plurality of related environmental parameters can be comprehensively analyzed and predicted, so that the multi-dimensional characteristics of the environmental change are comprehensively captured, and the multivariable time sequence analysis method not only enhances the understanding of the overall trend of the environmental change, but also can reveal the internal relation among different environmental parameters.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring technology, and in particular to an environmental monitoring system based on big data. Background Technology

[0002] In current environmental monitoring practices, traditional monitoring systems mainly rely on fixed sensor networks and simple data processing methods. These systems face significant challenges in dealing with rapidly changing environmental conditions and processing large-scale datasets. Although these systems can provide useful monitoring information under specific conditions, they still have limitations in terms of accuracy, flexibility, and predictive ability.

[0003] On the one hand, the spatial coverage limitations of fixed sensor networks mean that monitoring data often cannot fully reflect the environmental conditions of vast or inaccessible areas. In addition, traditional sensors are unstable under extreme environmental conditions, such as extremely high or low temperatures, which can easily lead to data deviations and affect the accuracy of monitoring results.

[0004] On the other hand, traditional data processing methods often lack the ability to process and analyze large-scale, high-dimensional and complex time series data. These methods often fail to effectively capture the seasonal changes, long-term trends and interrelationships between multiple environmental parameters in environmental data, resulting in an inability to accurately predict environmental change trends and assess their potential impact on ecosystems.

[0005] Furthermore, existing environmental monitoring systems often fall short in terms of real-time data and dynamic forecasting capabilities, making it difficult to provide timely and effective decision support for environmental management and emergency response. These limitations indicate that in order to better understand and address increasingly complex environmental challenges, there is an urgent need to develop more advanced, flexible, and intelligent environmental monitoring technologies and methods. Summary of the Invention

[0006] To achieve the above objectives, the present invention provides an environmental monitoring system based on big data.

[0007] An environmental monitoring system based on big data includes:

[0008] The data acquisition module is used to collect current environmental data from multiple sensors;

[0009] The data preprocessing module corrects and compensates the environmental data collected by the sensor under extreme temperature conditions.

[0010] The data analysis module uses an environmental prediction model to analyze the preprocessed environmental data and predict environmental change trends.

[0011] Impact Assessment Module: Utilizes simulation techniques to assess the extent to which environmental changes impact ecosystems.

[0012] Furthermore, the various sensors specifically include:

[0013] Temperature sensors: used to monitor temperature changes in the environment and accurately measure ambient temperature under extreme temperature conditions, including extremely high and low temperature environments;

[0014] Humidity sensor: Used to capture the humidity level in the environment;

[0015] Air quality sensor: used to detect the concentration of pollutants in the air, including PM2.5, PM10, sulfur dioxide, carbon monoxide, and ozone;

[0016] Water quality sensors: used to monitor chemical and biological parameters in water bodies, including pH value, dissolved oxygen, and heavy metal content.

[0017] Furthermore, the data preprocessing module specifically includes:

[0018] Data calibration: The responses of temperature sensors, humidity sensors, air quality sensors, and water quality sensors are pre-calibrated under standard conditions and extreme temperature conditions. The calibration process includes measuring the sensor output at different temperatures (including extremely low and extremely high temperatures) and recording the relationship between the output and the real environmental parameters.

[0019] Temperature dependence modeling: Based on the data calibration results, a temperature dependence model is established for each sensor to describe how the relationship between the sensor output and the actual environmental parameters changes with temperature;

[0020] Real-time data correction: Based on the current ambient temperature and temperature dependence model, the collected environmental data is corrected in real time. The temperature dependence model is used to adjust the environmental data collected by the original sensors to compensate for errors caused by extreme temperature conditions.

[0021] Furthermore, the temperature dependence model is established by selecting a linear model, expressed as: Y = aX + b, where Y is the sensor output, X is the actual environmental parameter (such as temperature), and a and b are model parameters.

[0022] Furthermore, the environmental prediction model employs an improved seasonal ARIMA model, specifically including:

[0023] Data preprocessing and integration: The environmental parameter dataset is transformed into a multivariate time series dataset. Data acquisition time points from different sensors are synchronized to ensure time consistency. The integrated dataset is normalized to convert data of different scales and units into comparable formats. For non-stationary series, logarithmic transformation is considered to stabilize variance and mean.

[0024] Seasonal differencing and trend identification: Seasonal and non-seasonal differencing are performed separately for each environmental parameter to eliminate seasonality and trends. The KPSS test is used to determine the order of differencing to ensure the stationarity of each series. The moving average technique is used to identify and remove long-term trends, preparing for accurate seasonal modeling.

[0025] Model parameter optimization: The seasonal ARIMA model parameters p, d, q, P, D, Q, m are selected using the AIC information criterion. For environmental data, long-term seasonal cycles (such as different months of the year) and short-term periodic changes (such as daily cycles) are considered. The SVAR (Seasonal Vector Autoregression) model is introduced to process multiple related environmental parameters simultaneously, so as to take advantage of the interrelationships between multiple environmental parameters.

[0026] Model Fitting and Validation: A rolling prediction method is used to perform cross-validation of the model to ensure its robustness and generalization ability. White noise is used to test the residuals of the fitted model to ensure that the model has fully captured the information in the data.

[0027] Prediction and Application: Based on the latest data, the model parameters are updated in real time to make dynamic predictions and capture the real-time changes in environmental parameters over time.

[0028] Furthermore, the KPSS test is a statistical test used to detect the stationarity of a time series. The KPSS test assumes that the series is stationary. If the test statistic is greater than the critical value, the null hypothesis is rejected, and the series is considered non-stationary, requiring differencing. The calculation formula is as follows:

[0029] Among them, S t It is part and X i It is the i-th environmental parameter observation in the time series. σ is the mean of the sequence, T is the length of the time series, and σ is the mean of the time series. 2 It is the variance of the sequence.

[0030] Furthermore, let there be multiple environmental parameters n (temperature, humidity, air quality index, water quality index), for each environmental parameter X i (t), where i = 1, 2, ..., n, and time point t, the seasonal ARIMA model is expressed as:

[0031] in, This represents a seasonal differencing operation, where D is the order of seasonal differencing, used to eliminate the effects of seasonal variations. This indicates a non-seasonal differencing operation, where d is the order of the non-seasonal differencing, used to eliminate trends. Φ(P) and Θ(Q) are the polynomials of the seasonal autoregressive and seasonal moving average terms, respectively, with orders P and Q. θ(q) and θ(q) are the polynomials of the non-seasonal autoregressive and non-seasonal moving average terms, respectively, with orders p and q, and ε. t It is a white noise error term.

[0032] Furthermore, the moving average technique is applied to each environmental parameter X. i (t) Calculate the moving average over period g:

[0033] Then subtract this moving average from the original data to remove the trend:

[0034] Furthermore, in the SVAR model, multiple time series X1(t), X2(t), ..., X n X(t) is treated as a vector X(t), and the model is represented as:

[0035] X(t)=A1X(t-1)+…+A p X(tp) + B1X(tm) + … + B P X(t-Pm)+ε(t)

[0036] Where A1,…,A p It is the coefficient matrix of the non-seasonal autoregressive term, B1,…,B P It is the coefficient matrix of the seasonal autoregressive term, where m is the seasonal period. For monthly data, m = 12 when representing the annual period. ε(t) is the error term vector, i.e., multivariate white noise.

[0037] Furthermore, the impact assessment module specifically includes:

[0038] Environmental change trend input: Receives environmental change trend prediction results from the data analysis module, including future prediction values ​​of multiple environmental data;

[0039] Ecosystem model selection: Select the appropriate ecosystem response model based on the ecosystem type of the monitoring area, including forests, wetlands, cities, and farmland;

[0040] Model parameterization: Parameterizing the ecosystem response model;

[0041] Simulation execution: Using predicted environmental data change trends as input, the simulation of the ecosystem response model is executed. During the simulation, the impact of changes in various environmental data on the ecosystem composition and function is considered.

[0042] Impact assessment: Analyze simulation results to assess the potential impact of environmental changes on the ecosystem.

[0043] The beneficial effects of this invention are:

[0044] This invention improves upon traditional seasonal ARIMA models, enabling more accurate capture and prediction of seasonal and non-seasonal trends in single environmental parameters. Improvements include automatically determining the difference order to ensure data stationarity and effectively eliminating long-term trends using moving average techniques. These improvements enhance the model's predictive accuracy and applicability, allowing environmental monitoring systems to provide timely warnings of potential environmental changes and offering more reliable scientific evidence for environmental management and decision-making.

[0045] This invention, by employing the SVAR model, can comprehensively analyze and predict the complex interactions between multiple relevant environmental parameters, thereby fully capturing the multidimensional characteristics of environmental change. This multivariate time series analysis method not only enhances the understanding of the overall trend of environmental change, but also reveals the intrinsic connections between different environmental parameters, providing strong support for the formulation of comprehensive environmental protection measures.

[0046] This invention employs a correction and compensation technique to ensure the accuracy and reliability of environmental data collected under extreme temperature conditions. By accurately calibrating sensor data and modeling its temperature dependence, the system can correct environmental data in real time, compensate for potential errors caused by extreme temperature changes, improve data quality, and lay a solid foundation for subsequent data analysis and environmental impact assessment. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of the system functional modules according to an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of the improved seasonal ARIMA model according to an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0051] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0052] like Figure 1-2 As shown, an environmental monitoring system based on big data includes:

[0053] The data acquisition module is used to collect current environmental data from multiple sensors;

[0054] The data preprocessing module corrects and compensates the environmental data collected by the sensor under extreme temperature conditions.

[0055] The data analysis module uses an environmental prediction model to analyze the preprocessed environmental data and predict environmental change trends.

[0056] Impact Assessment Module: Utilizes simulation techniques to assess the extent to which environmental changes impact ecosystems.

[0057] The various sensors specifically include:

[0058] Temperature sensors: used to monitor temperature changes in the environment and accurately measure ambient temperature under extreme temperature conditions, including extremely high and low temperature environments;

[0059] Humidity sensor: Used to capture the humidity level in the environment;

[0060] Air quality sensor: used to detect the concentration of pollutants in the air, including PM2.5, PM10, sulfur dioxide, carbon monoxide, and ozone;

[0061] Water quality sensors: used to monitor chemical and biological parameters in water bodies, including pH value, dissolved oxygen, and heavy metal content.

[0062] The data preprocessing module specifically includes:

[0063] Data calibration: The responses of temperature sensors, humidity sensors, air quality sensors, and water quality sensors are pre-calibrated under standard conditions and extreme temperature conditions. The calibration process includes measuring the sensor output at different temperatures (including extremely low and extremely high temperatures) and recording the relationship between the output and the real environmental parameters.

[0064] Temperature dependence modeling: Based on the data calibration results, a temperature dependence model is established for each sensor to describe how the relationship between the sensor output and the actual environmental parameters changes with temperature;

[0065] Real-time data correction: Based on the current ambient temperature and temperature dependence model, the collected environmental data is corrected in real time. The temperature dependence model is used to adjust the environmental data collected by the original sensors to compensate for errors caused by extreme temperature conditions.

[0066] The temperature dependence model is established by choosing a linear model, which is expressed as: Y = aX + b, where Y is the sensor output, X is the actual environmental parameter (such as temperature), and a and b are model parameters.

[0067] Parameter estimation: Regression analysis is used to estimate model parameters by minimizing the difference between model predictions and actual measurements.

[0068] Adjust the raw sensor data using a temperature-dependent model:

[0069] Real-time temperature monitoring: The current ambient temperature is monitored in real time within the environmental monitoring system.

[0070] Data Correction: Based on the monitored current temperature, the temperature dependence model established above is used to predict the theoretical output of the sensor at that temperature. Then, the actual measured value is compared with the model prediction, and corrections are made based on the difference. Specifically, if the model predicts that the sensor output at the current temperature is Y... predicted The actual measured value is Y. measured Then the corrected value Y corrected It can be calculated as follows:

[0071] Y corrected =Y measured +(Y true -Y predicted ), where Y true The calibration process uses real environmental parameter values ​​that match the current temperature. The sensor is recalibrated periodically or as needed, and the model parameters are updated to ensure the accuracy and adaptability of the calibration process.

[0072] The environmental prediction model uses an improved seasonal ARIMA model, which includes:

[0073] Data preprocessing and integration: The environmental parameter dataset is transformed into a multivariate time series dataset. Data acquisition time points from different sensors are synchronized to ensure time consistency. The integrated dataset is normalized to convert data of different scales and units into comparable formats. For non-stationary series, logarithmic transformation is considered to stabilize variance and mean.

[0074] Seasonal differencing and trend identification: Seasonal and non-seasonal differencing are performed separately for each environmental parameter to eliminate seasonality and trends. The KPSS test is used to determine the order of differencing to ensure the stationarity of each series. The moving average technique is used to identify and remove long-term trends, preparing for accurate seasonal modeling.

[0075] Model parameter optimization: The seasonal ARIMA model parameters p, d, q, P, D, Q, m are selected using the AIC information criterion. The parameter m is used to specify the period length of the recurrence of seasonal effects in the data. For example, when monthly data shows an annual seasonal pattern, m is usually set to 12. For environmental data, considering long-term seasonal cycles (such as different months of the year) and short-term periodic changes (such as daily cycles), the SVAR (Seasonal Vector Autoregression) model is introduced to handle multiple related environmental parameters simultaneously, so as to take advantage of the interrelationships between multiple environmental parameters.

[0076] The above model parameters are explained as follows:

[0077] p: The order of the non-seasonal autoregressive term. This parameter specifies the number of lags to be considered in the model, used to capture trends and periodic changes in time series data.

[0078] d: The order of the non-seasonal difference term, indicating how many differences are needed for the data to reach a stationary state. It is used to eliminate trend components in the data and make the series stationary.

[0079] q: The order of the non-seasonal moving average term, which specifies the lag period of the moving average term in the model and is used to capture random fluctuations in the time series.

[0080] P: The order of the seasonal autoregressive term, which specifies the number of lag periods for the seasonal autoregressive term in the model, and is used to capture the long-term dependence in the seasonal effect.

[0081] D: The order of the seasonal difference term, indicating the number of times seasonal differences are performed. It is used to eliminate seasonal trends in the series and make the series smooth in terms of seasonal structure.

[0082] Q: The order of the seasonal moving average term specifies the number of periods for the seasonal moving average term, used to capture random fluctuations in seasonal effects.

[0083] Model Fitting and Validation: A rolling prediction method is used to perform cross-validation of the model to ensure its robustness and generalization ability. White noise is used to test the residuals of the fitted model to ensure that the model has fully captured the information in the data.

[0084] Prediction and Application: Based on the latest data, the model parameters are updated in real time to make dynamic predictions and capture the real-time changes in environmental parameters over time.

[0085] The KPSS test is a statistical test used to detect the stationarity of a time series. The KPSS test assumes the series is stationary. If the test statistic is greater than the critical value, the null hypothesis is rejected, and the series is considered non-stationary, requiring differencing. The formula is as follows:

[0086] Among them, S t It is part and X i It is the i-th environmental parameter observation in the time series. σ is the mean of the sequence, T is the length of the time series, and σ is the mean of the time series. 2 It is the variance of the sequence.

[0087] In this invention, the KPSS test can be used to determine whether a time series of environmental parameters (such as temperature and humidity) needs to be differencing to achieve stationarity. If the KPSS test indicates that the series is non-stationary, first-order differencing is performed and the KPSS test is performed again until the series is stationary. Each differencing is equivalent to increasing the differencing order by one.

[0088] Let there be multiple environmental parameters n (temperature, humidity, air quality index, water quality index), for each environmental parameter X i (t), where i = 1, 2, ..., n, and time point t, the seasonal ARIMA model is expressed as:

[0089] in, This represents a seasonal differencing operation, where D is the order of seasonal differencing, used to eliminate the effects of seasonal variations. This indicates a non-seasonal differencing operation, where d is the order of the non-seasonal differencing, used to eliminate trends. Φ(P) and Θ(Q) are the polynomials of the seasonal autoregressive and seasonal moving average terms, respectively, with orders P and Q. θ(q) and θ(q) are the polynomials of the non-seasonal autoregressive and non-seasonal moving average terms, respectively, with orders p and q, and ε. t It is a white noise error term.

[0090] For each environmental parameter X i (t), perform the KPSS test. If the test result shows that the sequence is non-stationary, then perform first-order differencing. The KPSS test is repeated until the sequence becomes stationary, thereby determining the difference order d and D.

[0091] The moving average technique is used to smooth time series and identify trends. By calculating the average of several consecutive values ​​in a time series, the moving average can eliminate short-term fluctuations and reveal long-term trends. The moving average technique applies to each environmental parameter X... i (t) Calculate the moving average over period g:

[0092] Then subtract this moving average from the original data to remove the trend:

[0093] In environmental monitoring data analysis, the moving average technique is used to smooth the time series data of each environmental parameter, thereby identifying long-term environmental change trends. These identified trends can be removed from the original data as a whole, making the remaining time series more suitable for analyzing seasonality and other short-term changes. In the SVAR model, multiple time series X1(t), X2(t), ..., X... n X(t) is treated as a vector X(t), and the model is represented as:

[0094] X(t)=A1X(t-1)+…+A p X(tp) + B1X(tm) + … + B P X(t-Pm)+ε(t)

[0095] Where A1,…,A p It is the coefficient matrix of the non-seasonal autoregressive term, B1,…,B P It is the coefficient matrix of the seasonal autoregressive term, where m is the seasonal period. For monthly data, m = 12 when representing the annual period. ε(t) is the error term vector, i.e., multivariate white noise.

[0096] Within this SVAR model framework, the interactions between different environmental parameters can be considered, capturing their dynamic changes over time. Each coefficient matrix A... i and B i The interaction between environmental parameters under different time lags is described.

[0097] Steps for using SVAR models:

[0098] 1. Model identification: Determine the lag order p and seasonal lag order P in the model. This can be achieved through the information criterion AIC, or by observing the cross correlation function (CCF) between different environmental parameters.

[0099] 2. Parameter estimation: Estimate the coefficient matrix A using the least squares method or the maximum likelihood estimation method. iand B i The parameters in.

[0100] 3. Model Validation: Perform diagnostic tests on the model, including checking the autocorrelation and normality of the residuals. If the model is unsuitable, it may be necessary to adjust the lag order or consider other potential problems in the model.

[0101] 4. Prediction: Using the estimated SVAR model, future changes in environmental parameters are predicted. The predictions can be short-term or long-term, depending on the needs of the monitoring system and the stability of the model.

[0102] The ARIMA model is representative of univariate models, while the SVAR model extends to multivariate models, allowing the analysis of dynamic relationships between multiple related sequences. Both can handle seasonal data, but ARIMA focuses on the seasonal patterns of a single sequence, while SVAR considers seasonality within a multivariate framework, enabling the exploration of seasonal interactions between different sequences. In the context of environmental monitoring, seasonal ARIMA models focus on enhancing the accuracy and sensitivity of predictions for individual environmental parameters, for example, through more sophisticated differencing strategies and seasonal adjustments. The introduction of the SVAR model aims to leverage the correlations between multiple environmental parameters in a broader environmental monitoring system, thereby providing more comprehensive predictions of environmental change.

[0103] The impact assessment module specifically includes:

[0104] Environmental change trend input: Receives environmental change trend prediction results from the data analysis module, including future prediction values ​​of multiple environmental data;

[0105] Ecosystem model selection: Select the appropriate ecosystem response model based on the ecosystem type of the monitoring area, including forests, wetlands, cities, and farmland;

[0106] Model parameterization: Parameterizing the ecosystem response model;

[0107] Simulation execution: Using predicted environmental data change trends as input, the simulation of the ecosystem response model is executed. During the simulation, the impact of changes in various environmental data on the ecosystem composition and function is considered.

[0108] Impact assessment: Analyze simulation results to assess the potential impact of environmental changes on ecosystems. Assessment indicators include ecosystem health status, changes in biodiversity, and losses or gains in ecosystem services.

[0109] The ecosystem response model is as follows:

[0110] 1. The forest ecosystem model adopts GAP Models (Gap Dynamic Models): used to simulate the growth, competition and death processes of different tree species in the forest, as well as the dynamic changes of the forest.

[0111] 2. The wetland ecosystem model adopted is PnET-Wetlands: This is a process model that treats the water, carbon and nitrogen cycles of wetland ecosystems, taking into account the growth and decomposition processes of vegetation.

[0112] 3. The urban ecosystem model adopted is the UFORE Model (Urban Forest Effects Model): used to estimate the impact of urban forests on air quality and assess the absorption and interception of environmental pollutants by urban vegetation cover.

[0113] 4. The farmland ecosystem model uses Crop Growth Models: to simulate crop growth processes, yields, and responses to environmental changes (including climate change).

[0114] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention is limited to these examples; within the framework of the invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.

[0115] This invention is intended to cover all such substitutions, modifications, and variations falling within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An environmental monitoring system based on big data, characterized in that, include: The data acquisition module is used to collect current environmental data from multiple sensors; The data preprocessing module corrects and compensates the environmental data collected by the sensor under extreme temperature conditions. The data analysis module uses an environmental prediction model to analyze the preprocessed environmental data and predict environmental change trends. Impact Assessment Module: Utilizes simulation techniques to assess the extent to which environmental changes impact ecosystems.

2. The environmental monitoring system based on big data according to claim 1, characterized in that, The various sensors specifically include: Temperature sensor: Used to monitor temperature changes in the environment; Humidity sensor: Used to capture the humidity level in the environment; Air quality sensor: used to detect the concentration of pollutants in the air, including PM2.5, PM10, sulfur dioxide, carbon monoxide, and ozone; Water quality sensors: used to monitor chemical and biological parameters in water bodies, including pH value, dissolved oxygen, and heavy metal content.

3. The environmental monitoring system based on big data according to claim 2, characterized in that, The data preprocessing module specifically includes: Data calibration: The responses of temperature sensors, humidity sensors, air quality sensors, and water quality sensors are pre-calibrated under standard conditions and extreme temperature conditions. The calibration process includes measuring the sensor output at different temperatures and recording the relationship between the output and the real environmental parameters. Temperature dependence modeling: Based on the data calibration results, a temperature dependence model is established for each sensor to describe how the relationship between the sensor output and the actual environmental parameters changes with temperature; Real-time data correction: Based on the current ambient temperature and temperature dependence model, the collected environmental data is corrected in real time. The temperature dependence model is used to adjust the environmental data collected by the original sensors to compensate for errors caused by extreme temperature conditions.

4. The environmental monitoring system based on big data according to claim 3, characterized in that, The temperature dependence model is established by selecting a linear model, expressed as: Y = aX + b, where Y is the sensor output, X is the actual environmental parameter (such as temperature), and a and b are model parameters.

5. The environmental monitoring system based on big data according to claim 1, characterized in that, The environmental prediction model employs an improved seasonal ARIMA model, specifically including: Data preprocessing and integration: The environmental parameter dataset is transformed into a multivariate time series dataset. Data acquisition time points from different sensors are synchronized to ensure time consistency. The integrated dataset is normalized to convert data of different scales and units into comparable formats. Seasonal differencing and trend identification: Seasonal and non-seasonal differencing are performed separately for each environmental parameter to eliminate seasonality and trends. The KPSS test is used to determine the order of differencing to ensure the stationarity of each series. The moving average technique is used to identify and remove long-term trends, preparing for accurate seasonal modeling. Model parameter optimization: The seasonal ARIMA model parameters p, d, q, P, D, Q, m are selected using the AIC information criterion. For environmental data, considering long-term seasonal cycles and short-term periodic variations, the SVAR model is introduced to simultaneously process multiple related environmental parameters in order to utilize the interrelationships between multiple environmental parameters. Model Fitting and Validation: A rolling prediction method is used to perform cross-validation of the model to ensure its robustness and generalization ability. White noise is used to test the residuals of the fitted model to ensure that the model has fully captured the information in the data. Prediction and Application: Based on the latest data, the model parameters are updated in real time to make dynamic predictions and capture the real-time changes in environmental parameters over time.

6. The environmental monitoring system based on big data according to claim 5, characterized in that, The formula for calculating the KPSS test is as follows: Among them, S t It is part and X i It is the i-th environmental parameter observation in the time series. σ is the mean of the sequence, T is the length of the time series, and σ is the mean of the time series. 2 It is the variance of the sequence.

7. The environmental monitoring system based on big data according to claim 5, characterized in that, Let there be multiple environmental parameters, n, for each environmental parameter X i (t), where i = 1, 2, ..., n, and time point t, the seasonal ARIMA model is expressed as: in, This represents a seasonal differencing operation, where D is the order of seasonal differencing, used to eliminate the effects of seasonal variations. This indicates a non-seasonal differencing operation, where d is the order of the non-seasonal differencing, used to eliminate trends. Φ(P) and Θ(Q) are the polynomials of the seasonal autoregressive and seasonal moving average terms, respectively, with orders P and Q. θ(q) and θ(q) are the polynomials of the non-seasonal autoregressive and non-seasonal moving average terms, respectively, with orders p and q, and ε. t It is a white noise error term.

8. The environmental monitoring system based on big data according to claim 5, characterized in that, The moving average technique is applied to each environmental parameter X. i (t) Calculate the moving average over period g: Then subtract this moving average from the original data to remove the trend: X i ′(t)=X i (t)-MAX i (t).

9. The environmental monitoring system based on big data according to claim 7, characterized in that, In the SVAR model, multiple time series X1(t), X2(t), ..., X n X(t) is treated as a vector X(t), and the model is represented as: X(t)=A1X(t-1)+…+A p X(t-p)+B1X(t-m)+…+B P X(t-Pm)+ε() Where A1,…,A p It is the coefficient matrix of the non-seasonal autoregressive term, B1,…,B P It is the coefficient matrix of the seasonal autoregressive term, where m is the seasonal period. For monthly data, m = 12 when representing the annual period. ε(t) is the error term vector, i.e., multivariate white noise.

10. The environmental monitoring system based on big data according to claim 1, characterized in that, The impact assessment module specifically includes: Environmental change trend input: Receives environmental change trend prediction results from the data analysis module, including future prediction values ​​of multiple environmental data; Ecosystem model selection: Select the appropriate ecosystem response model based on the ecosystem type of the monitoring area, including forests, wetlands, cities, and farmland; Model parameterization: Parameterizing the ecosystem response model; Simulation execution: Using predicted environmental data change trends as input, the simulation of the ecosystem response model is executed. During the simulation, the impact of changes in various environmental data on the ecosystem composition and function is considered. Impact assessment: Analyze simulation results to assess the potential impact of environmental changes on the ecosystem.

Images