Intelligent environment monitoring system based on FPGA

Abstract

The invention relates to a field programmable gate array (FPGA)-based intelligent environment monitoring system in the field of environment monitoring systems, which comprises an FPGA master control module used for processing and analyzing monitoring data of each partition to obtain a partition state and an environment parameter change trend of the partition in the next one hour, and selecting a corresponding processing strategy according to the partition state; the partition management module is used for setting the type of each partition and judging and associating similar partitions; the control module is used for controlling related equipment in each partition to execute a preset processing strategy; the data acquisition module is used for data acquisition of each independent partition; the self-adaptive processing module is used for adjusting the range of the basic threshold value, the scene threshold value and the alarm threshold value; and the abnormity evaluation module is used for evaluating the abnormity probability according to the environment change trend and the threshold range, and the system can associate similar areas by monitoring the environment parameters of each area in real time and comprehensively evaluating the similarity of the areas so as to realize efficient environment monitoring and management.

Description

Technical Field

[0001] The present invention relates to an environment monitoring system, and in particular to an intelligent environment monitoring system based on FPGA applied in the field of environment monitoring systems. Background Art

[0002] Existing smart environmental monitoring systems integrate advanced sensor technology, the Internet of Things (IoT), artificial intelligence (AI), and big data analytics to monitor and manage environmental conditions in real time. These systems can detect multiple parameters such as air quality, temperature, humidity, light intensity, and water quality, and transmit the data to a central processing platform via wireless networks. Using AI algorithms, the system analyzes the collected data, identifies abnormal patterns, predicts potential problems, and automatically adjusts equipment operating status to optimize environmental conditions. Users can remotely monitor environmental conditions through mobile applications or web interfaces and receive instant alerts and recommendations. Smart environmental monitoring systems are widely used in various fields, including industry, agriculture, construction, and homes, aiming to improve energy efficiency, protect human health, and enhance quality of life.

[0003] The specification of Chinese invention patent CN108574737B discloses an automatic agricultural monitoring system and method based on cloud technology and the Zynq platform. Multiple sensors monitor the crop environment, condition the sensor output signals, collect AD data, and send them to the FPGA module via ZigBee wireless. The PC uploads the received information to the cloud server. This invention, based on cloud technology and centered on data analysis, greatly reduces the burden on farmers to manage large areas of crops. Mobile devices can simultaneously monitor the growth status of crops at multiple sites.

[0004] The specification of Chinese invention patent CN105791777B discloses a device for a Gigabit Ethernet multi-channel video acquisition and transmission system based on an FPGA. This invention uses an FPGA chip to encapsulate video data collected by multiple cameras using the Ethernet protocol, adding a CRC checksum, and then transmits it via the Gigabit Ethernet physical chip. The receiving FPGA chip receives the data via the Gigabit Ethernet physical chip, performs data verification and analysis, and displays it via VGA. This invention is applicable to a variety of video transmission and monitoring systems, reducing product design and maintenance costs.

[0005] The environmental monitoring system lacks efficiency and intelligence. The existing system still has shortcomings in efficient resource allocation, dynamic adjustment and abnormal warning response, especially the lack of coordinated management and linkage mechanisms for regions. Summary of the Invention

[0006] In view of the above-mentioned existing technologies, the technical problem to be solved by the present invention is that the existing systems still have deficiencies in efficient resource allocation, dynamic adjustment and abnormal warning response, especially the lack of collaborative management and association mechanism for regions.

[0007] To address the above issues, the present invention provides an FPGA-based intelligent environmental monitoring system, comprising: an FPGA master control module for processing and analyzing monitoring data from each partition to obtain the partition status and the environmental parameter change trend of the partition in the next hour, and select a corresponding processing strategy based on the partition status; the FPGA automatically adjusts the DSP slice allocation ratio based on the data traffic of each area;

[0008] Partition management module, used to set each partition type and determine and associate similar partitions;

[0009] A control module is used to control the execution of preset processing strategies of relevant devices in each partition;

[0010] The data acquisition module is used to collect data from each independent partition. During data acquisition, the acquisition frequency is adjusted accordingly according to the changes in the partition threshold. The partition threshold includes a basic threshold and a scene threshold. The data acquisition module is equipped with multiple parallel data channels, backup parallel channels, and regular data transmission channels.

[0011] Adaptive processing module, used to adjust the range of basic threshold, scene threshold and alarm threshold according to the basic parameters of the area and the changing trend of its environmental parameters in the next hour;

[0012] The anomaly assessment module is used to evaluate the anomaly probability based on the environmental change trend and threshold range.

[0013] In the above-mentioned FPGA-based intelligent environmental monitoring system, by real-time monitoring of environmental parameters in each area and comprehensively evaluating regional similarity and anomaly probability and adaptively adjusting the partitions, the system can associate similar areas to achieve efficient environmental monitoring and management.

[0014] As a further improvement of the present application, the area types include core areas and boundary areas. The monitoring data of multiple core areas are transmitted in real time through parallel data channels, and the monitoring data of the boundary areas are transmitted through regular data transmission channels according to the request queue.

[0015] As a further improvement of the present application, a sensor array and a distribution terminal are provided in each area. The distribution terminal is used to collect detection data in the sensor array and send the collected data to the data acquisition module; a data preprocessing unit is provided in the distribution terminal, and the data preprocessing unit preprocesses the detection data and extracts feature data.

[0016] As a further improvement of the present application, the distributed terminal monitors various environmental parameters in real time and identifies whether each environmental parameter exceeds the alarm threshold. If so, the corresponding processing strategy is directly called, and the data collection frequency is adjusted to the maximum while using the backup parallel data channel; if not, normal monitoring is maintained.

[0017] As another improvement of the present application, it also includes an energy management module, which monitors the energy consumption of each area in real time and then executes a dynamic power supply strategy: the core area is powered first, and the edge area is awakened on demand.

[0018] As another improved supplement of the present application, the environmental parameter distribution state is determined according to the distribution of each sensor in the sensor array and the environmental parameters detected by each sensor;

[0019] At set intervals, the FPGA main control module generates the environmental parameter change trend of each partition for the next hour based on the historical monitoring data of each partition. When the environmental parameter change trend is updated, the partition management module re-determines similar areas.

[0020] As another improvement and supplement to the present application, the data acquisition module extracts monitoring data from a set number of similar areas at set intervals for comparison and verification. If the similarity between the distribution state of the environmental parameters and the historical trend of environmental parameter changes is still greater than the set value, the comparison and verification passes;

[0021] If the similarity is still less than the set value, the change trend of the environmental parameters of each similar partition will be updated immediately, and the similar partition determination will be re-performed.

[0022] As another improvement of the present application, the anomaly assessment module supports multi-level early warning response. When the anomaly probability of a partition is greater than 75%, a first-level alarm is triggered. When the anomaly probability is greater than 30% and less than 75%, a second-level pre-check is started. When the anomaly probability is less than 30%, routine monitoring is maintained. During the second-level pre-check, the data acquisition frequency of the partition will be adjusted to the set frequency and marked. When the marked partition performs data transmission, the idle parallel channel is called for data transmission. When the first-level alarm occurs, the data acquisition frequency of the partition is adjusted to the maximum value, and the spare parallel channel is opened for data transmission of the partition.

[0023] In summary, this solution achieves efficient environmental monitoring and management by monitoring the environmental parameters of each area in real time and comprehensively evaluating the regional similarity and anomaly probability. It then associates similar areas and quickly deploys the adjustments to all similar areas when any area is adjusted. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a system block diagram of the first and second implementation modes of this application;

[0025] Figure 2Schematic diagram of the process of uploading monitoring data for each partition in the first and second implementation modes of this application;

[0026] Figure 3 This is a schematic diagram of the system workflow for the first and second implementation modes of this application;

[0027] Figure 4 This is a schematic diagram of the partitioning workflow of the first and second implementation methods of this application;

[0028] Figure 5 This is a schematic diagram of the similar area determination workflow of the second implementation method of this application. DETAILED DESCRIPTION

[0029] The following describes two implementation methods of the present application in detail with reference to the accompanying drawings.

[0030] The first implementation method:

[0031] Figure 1-4 The figure shows an FPGA-based intelligent environmental monitoring system, including: an FPGA main control module for processing and analyzing monitoring data of each partition to obtain the partition status and the environmental parameter change trend of the partition in the next hour, and select the corresponding processing strategy according to the partition status; the FPGA automatically adjusts the DSP slice allocation ratio according to the data flow of each area;

[0032] The FPGA master control module uses existing technology and works in conjunction with the ARM processor (PS) through programmable logic (PL). Those skilled in the art can select appropriate equipment in the existing technology to set up the FPGA master control module, such as the Xilinx Zynq UltraScale+ MPSoC chip.

[0033] The FPGA master control module is provided with a dynamic resource allocation algorithm, which adjusts the DSP slice allocation ratio according to the data flow of each partition. A person skilled in the art can select a suitable dynamic resource allocation algorithm in the prior art to set it.

[0034] When the FPGA main control module analyzes the partition status, it determines whether the partition is normal or abnormal based on the values of each environmental parameter in the partition monitoring data. The technical personnel in this field can set the abnormal status type corresponding to each environmental parameter as needed. For example, when the comprehensive humidity parameter exceeds the specified threshold of the partition, it is determined to be a humidity abnormal state.

[0035] The environmental parameter change trend of the next hour in each partition is obtained by analyzing the collected monitoring data using an existing prediction model. Persons skilled in the art can use appropriate time series analysis technology and prediction algorithms in the existing technology to predict the environmental parameter change trend; for example, a long short-term memory network is used to process nonlinear time series relationships to predict parameter changes in the next 60 minutes.

[0036] The partition management module is used to set the partition type and determine and associate similar partitions. The similarity of each partition is regularly evaluated based on the distribution status of the partition's environmental parameters and the trend of environmental parameter changes in the next hour. Areas with a similarity greater than a set value (such as 85%) are determined to be similar areas. When evaluating the similarity of partitions, those skilled in the art will set an appropriate regional similarity evaluation algorithm in the prior art.

[0037] For example, by fusing multi-source environmental data (such as geographic information, real-time monitoring parameters, and weather forecasts), features are extracted for each partition's static parameter distribution (e.g., through environmental parameter distribution status and cluster analysis) and future trends (e.g., based on trend prediction models). Static distribution differences are quantified using Euclidean distance and cosine similarity, and trend matching is assessed using dynamic time warping. Finally, similarity is comprehensively calculated using weighted fusion or data-driven models (e.g., PCA weight allocation).

[0038] When any partition is adjusted, the adjustment items are deployed synchronously to all similar partitions; the adjustment items include data collection frequency and processing strategy;

[0039] Area types include core areas and boundary areas. Monitoring data from multiple core areas is transmitted in real time through parallel data channels. That is, when uploading data from core areas, it is quickly uploaded through parallel independent data transmission channels. Monitoring data from boundary areas is transmitted through regular data transmission channels according to request queues. Monitoring data from multiple boundary areas is transmitted sequentially according to requests.

[0040] A control module is used to control the execution of preset processing strategies for relevant devices in each partition; the preset processing strategies include the workflow of relevant devices in each partition under different abnormal conditions;

[0041] The data acquisition module is used to collect data from each independent partition. During data acquisition, the acquisition frequency is adjusted accordingly according to the changes in the partition threshold. The partition threshold includes a basic threshold and a scene threshold. The data acquisition module is equipped with multiple parallel data channels, backup parallel channels, and regular data transmission channels.

[0042] The adaptive processing module is used to adjust the range of basic thresholds, scene thresholds, and alarm thresholds based on the basic parameters of the area and the changing trends of the environmental parameters in the next hour. The basic threshold is the static baseline critical value of the environmental parameter, which is used to determine whether the environment is in a normal state.

[0043] The scenario threshold is a dynamically adjusted critical value that is adjusted based on the regional basic parameters and the changing trends of its environmental parameters in the next hour. The adjustment range is 70-100% of the basic threshold. The alarm threshold indicates the critical value at which the environmental parameters reach an emergency state.

[0044] The adjustment process specifically includes the following: If the trend of regional environmental parameters changes significantly, such as a sudden temperature rise or a sharp drop in air quality (i.e., the temperature or air quality change rate is greater than the set value), the adaptive processing module will respond quickly and dynamically adjust the threshold range to adapt to the new environmental conditions; specific adjustment values, such as: the basic temperature threshold range is adjusted from 20 to 30 degrees Celsius under normal conditions to 18 to 35 degrees Celsius to cope with high temperatures; the basic air quality threshold range is adjusted from 50 to 100 AQI (Air Quality Index) in good conditions to 60 to 150 AQI to better reflect the situation of a sharp drop in air quality. This adaptive adjustment mechanism enables the system to more accurately reflect environmental conditions and improve the accuracy and reliability of monitoring.

[0045] The above threshold adjustment realizes a three-level response: when the parameter exceeds the scene threshold, log recording and data upload are performed; unless the basic threshold is exceeded, the spare parallel channel can be called to upload data and control the relevant equipment in the partition to execute the preset strategy. After the data is uploaded, the trend analysis of the environmental parameters that exceed the basic threshold of the partition is performed; when the alarm threshold is exceeded, the partition controls the relevant equipment to execute the preset strategy and push the alarm information.

[0046] Anomaly assessment module, used to assess anomaly probability based on environmental change trends and threshold ranges;

[0047] The anomaly assessment module supports multi-level early warning response. When the anomaly probability of a partition is greater than 75%, a first-level alarm is triggered. When the anomaly probability is greater than 30% and less than 75%, a second-level pre-check is started. When the anomaly probability is less than 30%, routine monitoring is maintained. During the second-level pre-check, the data collection frequency of the partition will be adjusted to the set frequency and marked. When the marked partition is transmitting data, the idle parallel channel is called for data transmission. When a first-level alarm occurs, the data collection frequency of the partition is adjusted to the maximum value, and a spare parallel channel is opened for data transmission of the partition.

[0048] Each area is equipped with a sensor array and a distribution terminal. The distribution terminal is used to collect detection data from the sensor array and send the collected data to the data acquisition module. A data preprocessing unit is set in the distribution terminal. The data preprocessing unit preprocesses the detection data and extracts feature data. The data preprocessing unit is set by those skilled in the art using appropriate data preprocessing and feature extraction technologies in the prior art. Preprocessing methods include filtering and noise reduction. The extracted features include time domain features, frequency domain features, and spatial features.

[0049] The distributed terminal has the ability to cache data in the event of disconnection: 72 hours of data storage (1TB industrial-grade SSD);

[0050] The distributed terminal monitors various environmental parameters in real time and identifies whether each environmental parameter exceeds the alarm threshold. If so, it directly calls the corresponding processing strategy and adjusts the data collection frequency to the maximum while using the backup parallel data channel. If not, it maintains normal monitoring.

[0051] It also includes an energy management module, which monitors the energy consumption of each area in real time and then implements a dynamic power supply strategy: the core area is powered first, and the edge area is awakened on demand. The energy management module is configured by those skilled in the art based on existing technologies.

[0052] When this embodiment works: the FPGA main control module receives the environmental monitoring data from the data acquisition module and processes and comprehensively analyzes the monitoring data of each partition through the built-in algorithm;

[0053] Comprehensive monitoring data determines the status of each partition, such as whether it is in normal, warning or abnormal state, and then selects the corresponding processing strategy based on the different status;

[0054] The FPGA master control module also has the ability to dynamically allocate resources, automatically adjusting the allocation ratio of DSP slices based on the data traffic of each area to ensure efficient and accurate data processing. This dynamic resource allocation mechanism enables the system to ensure stable monitoring of core partitions.

[0055] The partition management module sets the type of each partition. The monitoring data of multiple core areas are transmitted independently and quickly through parallel channels. The monitoring data can be processed in real time to ensure the timeliness and accuracy of the data; the monitoring data of the boundary areas are transmitted in queues to reduce the system load.

[0056] The control module, based on the analysis results from the FPGA master control module, controls the execution of pre-set processing strategies for the relevant devices within each zone. For example, if a zone is identified as abnormal, the control module immediately activates the relevant devices connected to the distributed terminals according to the set workflow based on the type of abnormality. For example, if the humidity in a zone is abnormal, the ventilation system and air purifier will be activated.

[0057] The data acquisition module is responsible for collecting data from each independent partition. It dynamically adjusts the collection frequency based on changes in partition thresholds. For example, when the basic threshold or scene threshold within a partition approaches the alarm threshold (that is, the difference between the basic threshold or scene threshold and the alarm threshold is less than a set constant), the data acquisition module increases the data collection frequency for that partition, for example, from once per minute to once every ten seconds, to more accurately monitor the dynamic changes in environmental parameters. This dynamic adjustment mechanism enables the system to maintain optimal collection results in different environments.

[0058] The adaptive processing module adjusts the range of basic thresholds, scene thresholds and alarm thresholds based on the basic parameters of the region and the changing trends of its environmental parameters in the next hour. This adaptive adjustment mechanism enables the system to better adapt to environmental changes and improve the accuracy of early warnings.

[0059] The entire system achieves real-time monitoring, analysis, and processing of environmental parameters through the collaborative work of various modules. When an abnormal situation occurs, the system can respond quickly and take appropriate measures to ensure the safety and stability of the environment.

[0060] Second implementation method:

[0061] Figure 2-Figure 5 It is shown that the distribution state of the environmental parameters is determined according to the distribution of each sensor in the sensor array and the environmental parameters detected by each sensor;

[0062] At set intervals, the FPGA master control module generates the environmental parameter change trend of each partition for the next hour based on the historical monitoring data of each partition. When the environmental parameter change trend is updated, the partition management module re-determines similar areas. When re-determining similar areas, partitions with similarity in historical determination records lower than a preset value (e.g., 70%) are excluded from comparison.

[0063] The data acquisition module extracts monitoring data from a set number of similar areas at set intervals for comparison and verification. If the similarity between the distribution status of environmental parameters and the historical trend of environmental parameter changes is still greater than the set value, the comparison and verification is passed;

[0064] If there is any inconsistency, the change trend of the environmental parameters of each similar partition will be updated immediately, and the similar partition determination will be re-performed.

[0065] This embodiment determines the distribution state of environmental parameters in each area based on the detection data of the sensor array; this state not only reflects the actual situation of the current environment, but also serves as the basis for subsequent analysis and prediction.

[0066] At regular intervals, the FPGA master control module uses historical monitoring data and a built-in prediction algorithm to generate the environmental parameter change trend for each partition over the next hour. Whenever the environmental parameter change trend is updated, the partition management module immediately re-evaluates similar areas to ensure that the system can always accurately identify and associate similar areas. (Under normal working procedures, similar area evaluation is performed once an hour.)

[0067] In order to ensure the accuracy and reliability of the data, this embodiment is equipped with a self-checking mechanism. The data acquisition module will also regularly (which can be set to a set interval time or a set number of acquisitions) extract a set number of similar area monitoring data for comparison and verification;

[0068] If the comparison results show that the similarity between the environmental parameter distribution state and the historical environmental parameter change trend is still greater than the set value, then the comparison verification can be considered passed and the system continues to operate normally. However, if the comparison results are inconsistent, the system will immediately update the environmental parameter change trend of each similar partition (that is, each similar partition of the comparison partition) and re-evaluate the similar partitions (compare the partitions with the updated environmental parameter change trend with other similar partitions) to ensure the accuracy and stability of the system.

[0069] The data management and self-checking mechanisms of this implementation make it easy for the system to maintain efficient operation in complex and changing environments, promptly detect and handle similar partition changes, and provide strong support for environmental monitoring and management;

[0070] In summary, this solution achieves real-time monitoring of environmental parameters, efficient data management, and accurate anomaly warnings. Through the dynamic resource allocation and intelligent analysis of the FPGA main control module, the system can ensure stable monitoring of core areas while optimizing data transmission in boundary areas and reducing system load. The data acquisition module dynamically adjusts the acquisition frequency based on changes in partition thresholds, improving the efficiency and accuracy of data acquisition. The adaptive processing module adaptively adjusts the threshold range based on the changing trends of environmental parameters, enhancing the system's early warning capabilities. In addition, the multi-level early warning response mechanism of the anomaly assessment module enables the system to take appropriate treatment measures under different abnormal conditions to ensure environmental safety and stability. Through the collaborative work of various modules, the entire system achieves efficient monitoring and management of environmental parameters, providing strong technical support for environmental protection and sustainable development.

[0071] In view of current actual needs, the protection scope of the above-mentioned implementation mode adopted in this application is not limited to this. Various changes made within the knowledge scope of technical personnel in this field without departing from the concept of this application still fall within the protection scope of the present invention.

Claims

1. An intelligent environment monitoring system based on FPGA, characterized by: include: The FPGA master control module is used to process and analyze the monitoring data of each partition to obtain the partition status and the environmental parameter change trend of the partition in the next hour, and select the corresponding processing strategy based on the partition status; The FPGA automatically adjusts the DSP slice allocation ratio according to the data flow of each area; The partition management module is used to set the partition type and determine and associate similar partitions. It regularly evaluates the similarity of each partition based on the distribution status of the partition's environmental parameters and the trend of environmental parameter changes in the next hour. Areas with similarity greater than the set value are identified as similar areas. When any partition is adjusted, the adjustment is synchronously deployed to all similar partitions. A control module is used to control the execution of preset processing strategies of relevant devices in each partition; A data acquisition module is used to collect data from each independent partition. During data acquisition, the acquisition frequency is adjusted accordingly according to changes in the partition threshold. The partition threshold includes a basic threshold and a scene threshold. The data acquisition module is provided with multiple parallel data channels, a backup parallel channel, and a regular data transmission channel. Adaptive processing module, used to adjust the range of basic threshold, scene threshold and alarm threshold according to the basic parameters of the area and the changing trend of its environmental parameters in the next hour; The anomaly assessment module is used to evaluate the anomaly probability based on the environmental change trend and threshold range.

2. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: The area types include core areas and boundary areas. Monitoring data of multiple core areas are transmitted in parallel and in real time. Monitoring data of the boundary areas are transmitted through a regular data transmission channel according to a request queue.

3. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: A sensor array and a distribution terminal are provided in each of the areas. The distribution terminal is used to collect detection data in the sensor array and send the collected data to the data acquisition module; a data preprocessing unit is provided in the distribution terminal, and the data preprocessing unit preprocesses the monitoring data and extracts feature data.

4. The FPGA-based intelligent environment monitoring system according to claim 3, characterized in that: The distributed terminal monitors various environmental parameters in real time and identifies whether each environmental parameter exceeds the alarm threshold. If so, the corresponding processing strategy is directly called and the data collection frequency is adjusted to the maximum while using the backup parallel data channel; if not, normal monitoring is maintained.

5. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: It also includes an energy management module, which monitors the energy consumption of each area in real time and then implements a dynamic power supply strategy: the core area is powered first, and the edge area is awakened on demand.

6. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: The environmental parameter distribution state is determined according to the distribution of each sensor in the sensor array and the environmental parameters detected by each sensor; At each set time interval, the FPGA main control module generates the environmental parameter change trend of each partition in the next hour based on the historical monitoring data of each partition. When the environmental parameter change trend is updated, the partition management module re-determines the similar areas.

7. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: The data acquisition module extracts monitoring data of a set number of similar areas at set intervals for comparison and verification. If the similarity between the distribution state of the environmental parameters and the historical trend of environmental parameter changes is still greater than the set value, the comparison and verification is passed; If the similarity is still less than the set value, the change trend of the environmental parameters of each similar partition will be updated immediately, and the similar partition determination will be re-performed.

8. The FPGA-based intelligent environment monitoring system according to claim 1, characterized in that: The anomaly assessment module supports multi-level early warning response. When the anomaly probability of a partition is greater than 75%, a first-level alarm is triggered. When the anomaly probability is greater than 30% and less than 75%, a second-level pre-check is started. When the anomaly probability is less than 30%, routine monitoring is maintained. During the second-level pre-check, the data acquisition frequency of the partition will be adjusted to the set frequency and marked. When the marked partition performs data transmission, the idle parallel channel is called for data transmission. When a first-level alarm occurs, the data acquisition frequency of the partition is adjusted to the maximum value, and a spare parallel channel is opened for data transmission of the partition.

Images