Search

CN-122018589-A - Ecological cabin resource recycling intelligent regulation and control system based on deep learning

CN122018589ACN 122018589 ACN122018589 ACN 122018589ACN-122018589-A

Abstract

The invention discloses an intelligent regulation and control system for recycling resources in an ecological cabin based on deep learning, which relates to the technical field of resource recycling and comprises a state data acquisition module, a time sequence conflict detection module, a reset sequence scheduling module, a pressure transition control module and a dynamic reset execution module, wherein the state data acquisition module acquires pressure values, flow values, valve opening, emission preparation signals and corresponding time records of the ecological cabin resource recycling system before reset to form a reset state record table. The invention realizes multi-loop reset time sequence coordination and conflict elimination based on dynamic regulation and control of time and state data, avoids control dislocation caused by overlapping of discharge operations, sets a pressure transition time period in the reset process and dynamically starts a post-execution loop, keeps the balance of gas-liquid pressure in the cabin, prevents backflushing and impact, and improves the stability and operation reliability of ecological cabin resource circulation.

Inventors

  • YIN QUANXI
  • FANG CHANGJIANG

Assignees

  • 广东伊斐智居科技有限公司

Dates

Publication Date
20260512
Application Date
20260205

Claims (10)

  1. 1. The ecological cabin resource recycling intelligent regulation and control system based on deep learning is characterized by comprising a state data acquisition module, a time sequence conflict detection module, a reset sequence scheduling module, a pressure transition control module and a dynamic reset execution module: the state data acquisition module acquires pressure values, flow values, valve opening degrees, emission preparation signals and corresponding time records of the ecological compartment resource recycling system before resetting to form a reset state record table; The time sequence conflict detection module divides respective discharge time periods of the two loops according to the to-be-discharged time points of the air purification loop and the water regeneration loop in the reset state record table, compares whether the discharge time periods of the two loops are overlapped, generates a time conflict record if the discharge time periods are overlapped, and marks the specific time position of the occurrence of the conflict in the time conflict record; The resetting sequence scheduling module is used for determining the resetting execution sequence of the air purification loop and the water regeneration loop according to the time conflict record, calculating the interval duration between the execution loop and the execution loop, updating the interval duration and the new discharging time point to a resetting state record table, and forming updated resetting arrangement; The pressure transition control module is used for executing reset operation according to the updated reset arrangement, setting a pressure transition time period in the prior execution loop discharging process, enabling the post execution loop to be in a waiting state, and recording pressure change data in the pressure transition time period to a reset state record table; And the dynamic reset execution module determines the time when the pressure transition time period is ended according to the pressure change data recorded in the reset state recording table, starts the post-execution loop to discharge after the pressure transition time period is ended, and supplements the discharge time period of the post-execution loop discharge process to the reset state recording table, thereby forming a continuously updated reset time schedule.
  2. 2. The intelligent regulation and control system for recycling of resources in an ecological compartment based on deep learning as set forth in claim 1, wherein the reset state record table is formed as follows: The state data acquisition module acquires pressure values, flow values, valve opening degrees, emission preparation signals and corresponding time records of the ecological compartment resource recycling system in the operation stage before resetting, and records the acquired operation parameters according to time indexes; The state data acquisition module performs merging processing on the obtained pressure value, flow value, valve opening and emission preparation signal according to time sequence, integrates different loop operation parameters at the same time point into a unified record data row, and establishes a corresponding relation among the parameters; The state data acquisition module extracts time points to be discharged of the air purification loop and the water regeneration loop according to the discharge preparation signals contained in the merged operation data, and binds each time point with the corresponding pressure value, flow value and valve opening to form a time identification set; The state data acquisition module generates a reset state record table from the tidied and marked operation data, the reset state record table takes a time field as an index field, records pressure data, flow data, valve opening and discharge preparation signal fields, and marks the time points to be discharged of the air purification loop and the water regeneration loop.
  3. 3. The intelligent regulation and control system for recycling of resources in an ecological compartment based on deep learning according to claim 2 is characterized in that when a reset state record table is generated, the time points to be discharged of the air purification circuit and the water regeneration circuit are distinguished by different identifiers, a one-to-one correspondence is established in a time field, and meanwhile, the pressure value, the flow value and the valve opening are synchronously recorded according to time indexes, so that the reset state record table is formed.
  4. 4. The deep learning-based intelligent regulation and control system for recycling resources in an ecological compartment according to claim 2, wherein the specific time position step for marking the occurrence of the conflict in the time conflict record is as follows: Extracting continuous operation data before and after activation of the emission preparation signal from the point of time to be emitted of the air purification circuit and the water regeneration circuit by taking a time recording field in the reset state recording table as an index, and determining the respective emission starting time and ending time to form an emission time interval; The time sequence conflict detection module maps the discharge time intervals of the air purification loop and the water regeneration loop to a unified time axis, compares the discharge start time with the discharge end time of the opposite loop, judges whether the two time intervals are overlapped and records a time overlapping interval; the time sequence conflict detection module generates a time conflict record based on the time index of the reset state record table, records the conflict time position and the running state parameters of the two loops, and sets an independent number for each time superposition interval; the time sequence conflict detection module backfills the starting time and the ending time of the time conflict record to the time row corresponding to the reset state record table, forms a labeling field to identify the conflict state and generates a unified discharge time mapping table.
  5. 5. The intelligent regulation and control system for recycling resources in an ecological compartment based on deep learning according to claim 4, wherein when a time conflict record is generated, the pressure value, the flow value and the valve opening of the air purification loop and the water regeneration loop in a time superposition section are synchronously recorded, and the conflict duration is written into the time conflict record, so that the time conflict record simultaneously has time information and operation parameter information.
  6. 6. The intelligent regulation and control system for recycling of resources in an ecological compartment based on deep learning as set forth in claim 4, wherein the updated reset schedule is formed as follows: Sequentially sorting conflict time intervals according to the initial time field in the time conflict record, and extracting operation state parameters of the air purification loop and the water regeneration loop to judge the operation load and the resource occupation condition; The reset sequence scheduling module determines the reset execution sequence of the air purification loop and the water regeneration loop by combining the running state parameters in the conflict interval, and records the determined sequence of the execution loop and the execution loop in a time index form in a time conflict record; The reset sequence scheduling module calculates the interval duration between the execution loop and the execution loop according to the conflict end time in the time conflict record and the discharge start time in the reset state record table, and records the interval duration and the corresponding discharge start time in the reset state record table; The reset sequence scheduling module integrates the first execution loop, the second execution loop and the interval duration data into a recovery bit state record table to generate updated reset arrangement by taking the conflict interval number in the time conflict record as a search index.
  7. 7. The intelligent regulation and control system for recycling of resources in an ecological cabin based on deep learning according to claim 6, wherein when the interval duration between the execution-first loop and the execution-later loop is calculated, the interval duration is consistent with the time period for recovering the pressure in the cabin to a stable state by performing time matching on the difference between the discharge ending time and the starting time by taking the pressure value, the flow value and the valve opening change of the pipe network in the cabin as reference conditions.
  8. 8. The deep learning-based intelligent regulation and control system for recycling resources in an ecological compartment according to claim 6, wherein the step of recording pressure change data in a pressure transition time period to a reset state record table is as follows: Reading the information of the discharge time point and the interval duration of the execution loop from the reset state record table according to the updated reset arrangement, positioning the discharge starting time by taking the time field as an index, determining a pressure reference value before the start of discharge, and starting a reset flow; The pressure transition control module sets a pressure transition time period in the process of executing the loop discharge firstly, wherein the starting point of the pressure transition time period is a discharge starting time point, the end point of the pressure transition time period is a time point when the pressure in the cabin is restored to a stable state, and maintains stable flow in the pressure transition time period for performing the discharge operation, and then the execution loop is kept in a waiting state; The pressure transition control module continuously monitors the pressure change condition in the cabin in a pressure transition time period, and records the pressure value, the valve opening, the flow change and the discharge signal state in a reset state record table to form a pressure change sequence; And the pressure transition control module updates a reset state record table when the pressure transition time period is ended, uniformly writes the starting time, the ending time and the pressure change range into the table, and forms a time hierarchy relation of a first execution loop discharging stage, a pressure transition stage and a later execution loop waiting stage.
  9. 9. The intelligent regulation and control system for recycling of resources in an ecological compartment based on deep learning according to claim 8, wherein the pressure change data in the compartment is continuously recorded by taking a time field as an index in a pressure transition time period, and the pressure value at each time point is synchronously stored with the corresponding flow value, valve opening and discharge signal state to form a time-series pressure change record.
  10. 10. The intelligent regulation and control system for recycling of resources in an ecological compartment based on deep learning according to claim 8, wherein the step of determining the end time of the pressure transition time period according to the pressure change data in the reset state record table, starting the post-execution loop discharge when the pressure transition time period is ended, and complementarily recording the post-execution loop discharge time period to the reset state record table is as follows: The dynamic reset execution module reads pressure field data in a reset state record table when the prior execution loop discharging operation is finished, and extracts continuous pressure values of a pressure transition time period by taking a time field as an index to form a complete pressure change sequence; The dynamic reset execution module compares the pressure value at each moment with a reference pressure value at the end of discharge according to the pressure change trend in the reset state record table, determines the time index for restoring the pressure in the cabin to the stable state and records the end moment in the reset state record table; The dynamic reset execution module activates a discharge unit of the execution circuit according to the ending time recorded in the reset state recording table, and records the discharge starting time, the pressure value, the flow value and the valve opening state of the execution circuit by taking the time field as an index; And the dynamic reset execution module supplements the data of the discharge time period to the reset state record table when the discharge operation of the post-execution loop is completed, and synchronously updates the pressure change data, the flow data and the valve opening information of the post-execution loop to form continuously updated reset time schedule.

Description

Ecological cabin resource recycling intelligent regulation and control system based on deep learning Technical Field The invention relates to the technical field of resource recycling, in particular to an intelligent regulation and control system for resource recycling in an ecological compartment based on deep learning. Background The intelligent regulation and control of the resource recycling in the ecological cabin based on deep learning refers to a comprehensive control method for carrying out real-time monitoring, dynamic prediction and self-adaptive regulation on various resources (including water, air, nutrient substances, energy and waste) in the cabin by utilizing the cooperative work of a deep learning model and a data acquisition control system in a closed or semi-closed ecological environment (such as a space cabin, a biological regeneration cabin, an extreme environment survival cabin and the like). The system collects multidimensional data such as temperature and humidity, oxygen concentration, carbon dioxide content, microbial community change, crop growth state, waste conversion efficiency and the like through a multisource sensing network and a data acquisition control system, builds a dynamic correlation model of material flow and energy flow in a cabin, carries out pattern recognition and trend analysis on the time sequence data through a deep learning algorithm, automatically judges the resource supply and demand state and the circulation efficiency, and then carries out cooperative regulation and control on links such as air purification, water regeneration, nutrient solution proportioning, bioreactor operation, energy distribution and the like according to a model output result and acquisition control system feedback, so that closed loop circulation utilization and resource optimal allocation in an ecological cabin are realized. The intelligent ecological cabin is characterized in that the artificial intelligent driven prediction-feedback mechanism is used for bi-directional linkage with the data acquisition control system to replace the traditional static control logic, so that the ecological cabin has self-learning, self-correcting and self-optimizing capabilities in complex and changeable life support environments, and long-term stable operation and ecological balance of the system are ensured. The prior art has the following defects: In the prior art, in the multi-loop resetting process of the ecological compartment resource recycling system, the air purification loop and the water regeneration loop are often subjected to time sequence scheduling by the same control end, and when the air purification loop and the water regeneration loop trigger a discharge program simultaneously in operation, the problem of phase overlapping of control rhythm is easy to occur. Because the priority judgment of the deep learning control end in a complex environment depends on real-time data learning, once the model cannot correctly distinguish the sequence relation of the discharge time sequence, the situation that the execution instructions overlap or are misplaced easily occurs, so that the gas-liquid discharge path is instantaneously reversed, and a pressure recoil phenomenon is formed. The recoil can enable a pipe network in the cabin to generate short-time high-pressure fluctuation, cause valve body impact, filter unit instability and sealing component damage, and can also cause misjudgment of a sensor and cascading failure of a control system when serious, so that stable operation of an ecological cabin resource circulation system is affected. The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art. Disclosure of Invention The invention aims to provide an intelligent regulation and control system for recycling resources in an ecological cabin based on deep learning so as to solve the problems in the background technology. In order to achieve the purpose, the invention provides the technical scheme that the ecological cabin resource recycling intelligent regulation and control system based on deep learning comprises a state data acquisition module, a time sequence conflict detection module, a reset sequence scheduling module, a pressure transition control module and a dynamic reset execution module: The state data acquisition module acquires a pressure value, a flow value, a valve opening, a discharge preparation signal and a corresponding time record of the ecological compartment resource recycling system before resetting to form a resetting state record table, and marks a time point to be discharged of the air purification loop and the water regeneration loop in the resetting state record table; The time sequence conflict detection module divides respective discharge time periods