An ecological restoration method for a coal mine dump in an arid region

Abstract

The present application relates to the technical field of ecological restoration of coal mines in arid regions, and discloses an ecological restoration method for a coal mine dump in an arid region. A multi-parameter environment perception model containing soil humidity, soil composition, surface temperature, vegetation index and regional slope data is established to determine whether to start ecological restoration. At least one of water, nutrient solution or soil conditioner is selected as a restoration medium, which is quantitatively delivered to the predetermined area through an independent delivery path isolated from natural precipitation. The restoration area is composed of an independently laid delivery pipe network, the water flow direction is dynamically determined according to the regional slope, and the pipe section in the position with the advantage of natural diffusion is preferentially started. The operation time period is determined according to the medium absorption rate during delivery, intermittent delivery is performed on the selected pipe section, and the average flow of each cycle is controlled not to exceed the system storage capacity. When the storage is insufficient, the delivery is suspended and the operation information is recorded, and after the storage is restored, the remaining operation is continued to realize closed-loop restoration control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ecological restoration of coal mines in arid regions, in particular to an ecological restoration method for a coal mine dump in an arid region. BACKGROUND

[0002] In the process of coal mining in arid regions, the ecological environment of the dump area, where mining waste is concentrated, is extremely fragile. Arid regions have little rainfall and high evaporation all year round, and the soil moisture content is generally low. The soil of the dump is mainly composed of rock debris, gravel and a small amount of topsoil stripped by mining, and the soil composition is chaotic, the nutrients are scarce, and the water and fertilizer retention capacity is extremely poor. In addition, the dump terrain is mostly artificial accumulation of slopes, with a surface slope of more than 30°, and the water and soil loss is serious. Natural precipitation is difficult to infiltrate into the soil and be preserved. These factors together make it difficult for vegetation to naturally recover in the dump area, and the ecological system is severely damaged, which has a serious impact on the environment of mining.

[0003] The method for ecological restoration of coal mine dumps in arid regions is mostly focused on simple soil improvement and vegetation planting, and lacks systematic perception and precise control of the complex environment of the dump. Some restoration schemes only intervene by single irrigation or application of soil improver, but due to insufficient understanding of key environmental parameters such as soil humidity, nutrient status and surface slope in different areas, it is impossible to determine the appropriate timing of ecological restoration, often resulting in problems such as waste of resources due to early start of restoration work, or missing the best restoration window period. In terms of restoration medium delivery, the existing technology mostly uses uniform delivery pipelines without considering the influence of slope difference in different areas of the dump on the direction of water flow, which cannot realize directional and quantitative delivery of the restoration medium, resulting in over-supply of the medium in some areas, while some dry or nutrient-deficient areas are difficult to be effectively supplied.

[0004] The existing delivery method is mostly continuous delivery without adjustment according to the absorption rate of the soil to the restoration medium. When the soil absorption capacity is limited, continuous delivery can easily cause the medium to accumulate and lose on the ground, not only reducing the restoration efficiency, but also possibly causing secondary environmental problems. At the same time, the existing scheme lacks dynamic monitoring and closed-loop control of the storage amount of the restoration medium. When the storage is insufficient, the restoration work is often directly interrupted, and the information of the areas where the work has been completed is not recorded, so that the subsequent work cannot be accurately connected after the storage is restored, resulting in discontinuity in the restoration process and further affecting the ecological restoration effect. Overall, the current ecological restoration method for coal mine dumps in arid regions lacks systematicness, precision, stability and intelligent management, and cannot meet the needs of efficient ecological restoration of the dump. SUMMARY

[0005] The present application aims to provide an ecological restoration method for a coal mine dump in an arid region to solve the problems raised in the background.

[0006] To achieve the above object, the application provides an ecological restoration method for a coal mine dump in an arid region, which comprises the following steps:

[0007] A multi-parameter environment perception model containing soil humidity, soil composition, surface temperature, vegetation index and regional slope data is established, and whether the initial condition for starting ecological restoration is reached is determined based on the multi-parameter environment perception model;

[0008] At least one of water, nutrient solution or soil conditioner is selected as a restoration medium, and the restoration medium is quantitatively delivered to a predetermined restoration area through an independent delivery path isolated from natural precipitation;

[0009] The restoration area is formed by an independently laid delivery pipe network in structure, and the water flow direction is dynamically determined according to the current regional slope data, and the pipe section in a position having a natural diffusion advantage is preferentially started to deliver;

[0010] During the delivery process, the target operation time period is determined according to the predicted medium absorption rate, and intermittent delivery is performed on the selected delivery pipe section, and the average flow rate in each operation cycle is controlled to be not more than the current system reserve capacity;

[0011] When it is detected that the reserve is insufficient, the delivery process is suspended and the operation time length and area information that have been completed are recorded, and after the reserve is recovered, the remaining operation process is continued to be performed until the closed-loop restoration control is completed.

[0012] Preferably, the construction process of the multi-parameter environment perception model comprises dynamic coupling calculation of arid region climate data and topographic change trend, soil humidity information is continuously collected by a multi-point sensing device, and an instantaneous water demand prediction value is obtained by using a water migration model and a soil porosity equation for derivation;

[0013] The regional slope information is obtained by a height measurement unit to calculate multi-point height data, and is corrected in combination with actual landform features to form a slope distribution model;

[0014] The vegetation index information is obtained by periodic scanning of a multi-spectral sensor, and the vegetation coverage data is estimated to obtain the water demand in the growth period according to the spectral reflection characteristics;

[0015] The multi-parameter environment perception model is based on a fitting curve of the environmental change law in historical ecological data, and multi-parameter prediction is performed in combination with the current collection result to determine whether the ecological restoration starting critical condition has been reached.

[0016] Preferably, the structural layout mode of the independent delivery path comprises a closed pipe system that is physically completely isolated from natural precipitation, the restoration medium delivered in the internal pipe system is selected from one of water, nutrient solution or soil conditioner, and controllable variable flow is realized by a multi-stage booster pump;

[0017] The conveying path is provided with multiple branch conveying sections, each of which has one-way conveying performance to avoid medium backflow;

[0018] In the control strategy, the recovery medium penetration rate in the current environment per unit time is calculated to match the current water requirement of the soil, and the flow rate and conveying time are dynamically selected;

[0019] The medium flow direction is adjusted in real time according to the regional slope information, and is preferentially guided to the branch section in the gravity flow dominant direction, so as to improve the medium utilization efficiency and reduce the flow peak value per unit time.

[0020] Preferably, the slope direction identification strategy for judging whether the conveying target region is in a natural diffusion dominant position includes vector analysis on the conveying pipe layout direction of the dump in three-dimensional terrain data, and the gravity vector is taken as a reference. The most favorable medium diffusion direction is determined by calculating the included angle between the conveying pipe axis and the gravity vector.

[0021] When the included angle is less than a set angle threshold, it is considered that the conveying section has natural diffusion capacity; if all regions do not meet the condition, it enters a waiting state until the next cycle of environmental parameter update to evaluate whether to activate again.

[0022] Preferably, the penetration modeling algorithm for predicting the medium absorption rate includes linkage calculation of medium flow rate, initial concentration, soil porosity and current vegetation coverage, modeling by using a water migration model and a plant root absorption model, and dynamic adjustment of the predicted operation period by calculating the deviation between the unit time penetration amount and the total water requirement of the region.

[0023] Preferably, the intermittent conveying strategy adopts a dynamic flow control algorithm, which sets the start-stop ratio in the conveying cycle according to the current system reserve capacity, the environmental evaporation rate and the soil saturation degree. The flow ratio is limited by the capacity control threshold and can be automatically adjusted at different operation stages.

[0024] The operation cycle is divided into two stages of activation period and intermittent period. The activation period is executed after the system resource management unit allocates the available capacity, and the intermittent period is adjusted according to the environmental temperature change trend to prevent excessive evaporation of water.

[0025] Preferably, the operation state recording strategy for suspending the conveying process includes synchronous recording of the current operated region, the total amount of medium put in, the average soil humidity change and the time stamp of the current operation time point. The recording information forms an operation breakpoint state structure body and is written into the system data storage area.

[0026] In the reserve recovery, first, the previous breakpoint information is read, and the current required compensation is estimated according to the unfinished work area, the medium quantity put in and the soil water shortage degree, and compared with the medium conveying rate to determine whether the execution ability is available;

[0027] If the continuous execution condition is met, the corresponding conveying pipe section work process is directly recovered, otherwise, the next cycle is delayed.

[0028] Preferably, the system reserve capacity acquisition method comprises real-time comparison of the current area water demand total amount and the water source supply capacity, judgment of whether the conveying operation is allowed to be activated according to the reserve capacity threshold, and execution of the conveying control sub-process only under the condition that the system resource scheduling mechanism sends an available signal.

[0029] Preferably, the completion standard of the closed-loop recovery control is based on the soil humidity recovery rate, the vegetation growth state and the remaining untreated area statistics result to make a joint decision, and the termination threshold is satisfied as the judgment basis.

[0030] The soil humidity is detected by a multi-point sensing device in real time, and compared with the initial drought state to calculate whether the humidity improvement rate is stable and tends to the expected value.

[0031] The vegetation growth state is judged by a multispectral sensor whether the growth standard is reached.

[0032] The untreated area is compared with the vegetation coverage difference before and after the treatment according to the geographic information data.

[0033] When the three indicators all reach the preset threshold, it is automatically determined that the ecological recovery is completed, and the work record is emptied.

[0034] Preferably, the ecological recovery method further comprises:

[0035] The environmental monitoring data are encrypted and transmitted by a key stream, and the key stream is obtained by chaotic nested encryption and random distribution of key stream coefficients.

[0036] The ciphertext monitoring data are transmitted to a central processing system for decryption based on a communication interface.

[0037] The medium conveying instruction is generated according to the decrypted monitoring data.

[0038] Compared with the prior art, the beneficial effects of the present application are:

[0039] The ecological restoration method of the coal mine dump in the arid region can comprehensively and accurately master the environmental conditions of each region of the dump by establishing a multi-parameter environmental perception model, integrating key environmental data such as soil humidity, soil composition, surface temperature, vegetation index and regional slope, so as to scientifically judge whether the initial conditions of ecological restoration are met, avoid the misjudgment of the restoration opportunity caused by incomplete environmental information, make the start of ecological restoration operation more reasonable, and carry out the work when the environmental conditions are most suitable for vegetation growth and soil improvement, thereby laying a good foundation for the subsequent restoration effect.

[0040] In the restoration medium transportation link, at least one of water, nutrient solution or soil conditioner is selected as the restoration medium, and is transported through an independent transportation path isolated from natural precipitation, effectively avoiding the interference of natural precipitation on the concentration and transportation amount of the restoration medium, ensuring the accurate control of the medium amount transported to the predetermined restoration area, and selecting the appropriate restoration medium type according to the actual needs of different regions to realize targeted ecological intervention, such as focusing on water transportation in extremely dry soil areas, focusing on nutrient solution transportation in nutrient-deficient areas, and focusing on soil conditioner transportation in poor soil structure areas, thereby improving the utilization efficiency of the restoration medium.

[0041] The restoration area adopts an independently arranged transportation pipe network in structure, dynamically judges the water flow direction combined with the current regional slope data, and preferentially starts transportation for the pipe section in a position with a natural diffusion advantage, so as to fully utilize the topographic advantage, make the restoration medium efficiently diffuse along the natural water flow direction, cover more areas needing restoration, reduce the invalid consumption of the medium in the transportation process, especially in the dump with large slope difference, avoid the problem that the medium cannot reach some areas due to the influence of slope, and realize the uniform and efficient distribution of the restoration medium.

[0042] During the transportation process, the target operation time period is determined according to the predicted medium absorption rate, and intermittent transportation is performed on the selected transportation pipe section, while the average flow in each operation cycle is controlled to be not more than the current system reserve capacity, so as to make the restoration medium fully contact and be absorbed by the soil, avoid the medium accumulation and loss caused by continuous transportation, improve the absorption utilization rate of the medium, and reasonably control the flow to avoid system failure due to excessive load, thereby ensuring the stable progress of the transportation process.

[0043] When the shortage of reserves is detected, the transportation process is suspended and the completed operation time and regional information are recorded, and after the reserves are recovered, the remaining operation process is continued, forming a complete closed-loop restoration control, ensuring the continuity of the ecological restoration operation, avoiding the problem that the restoration process cannot be connected after interruption due to the problem of medium reserves, and enabling each region of the dump to complete ecological restoration according to the plan, thereby avoiding the problem of incomplete restoration in some areas. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The working principle diagram of the ecological restoration method of the coal mine dump in the arid region according to the present application;

[0045] Figure 2 The working principle diagram of the multi-parameter environment perception model construction process;

[0046] Figure 3 The working principle diagram of the independent conveying path layout and control process. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0048] Please refer to Figure 1 The present application provides an ecological restoration method for a coal mine dump in an arid region, which comprises integrated environment perception, medium conveying and intelligent control to achieve efficient ecological restoration. The method establishes a multi-parameter environment perception model that integrates soil humidity, soil composition, ground temperature, vegetation index and regional slope data to determine whether the initial conditions for starting ecological restoration are met. After the initial conditions are met, the system selects at least one of water, nutrient solution or soil conditioner as the restoration medium, and delivers the restoration medium to the predetermined restoration area through an independent conveying path isolated from natural precipitation. The restoration area is composed of independently laid conveying pipe networks in structure, and the system dynamically determines the water flow direction according to the current regional slope data, and preferentially selects the pipe section in a position with a natural diffusion advantage to start conveying. During the conveying process, the target operation time period is determined according to the predicted medium absorption rate, and intermittent conveying is performed on the selected conveying pipe section, with the average flow rate in each operation cycle controlled to be no more than the current system reserve capacity. If a shortage of reserves is detected, the system suspends the conveying process and records the completed operation time and regional information, and continues to perform the remaining operation after the reserves are restored, until the closed-loop restoration control is completed. The entire implementation process relies on real-time data acquisition and automated control to ensure efficient and sustainable ecological restoration of the coal mine dump in the arid region.

[0049] Embodiment 1: Please refer to Figure 2, the construction of a multi-parameter environment perception model is the initial stage of the ecological restoration method, and its core lies in the dynamic monitoring and intelligent fusion of complex environmental parameters in the arid region; the construction of the model starts with the dynamic coupling calculation of climate data and topographic change trend in the arid region, the climate data come from the meteorological station network deployed in the dump and the surrounding area, these meteorological stations continuously collect indicators such as air temperature, precipitation, wind speed and solar radiation, and the topographic change trend is obtained through periodic three-dimensional scanning, and the direction of geomorphic evolution is predicted by using time series analysis; the collection of soil moisture information depends on the multi-point sensing device arranged at different depths and positions, these sensors use frequency domain reflection technology, which can penetrate the soil medium at high frequency and return dielectric constant readings, so as to accurately invert the volumetric water content, the original data collected are filtered and normalized, and then input into the water migration model in real time.

[0050] The water migration model is derived in combination with the soil porosity equation, the soil porosity is calculated by the particle density of the soil sample obtained by laboratory analysis and the bulk density data measured in the field, the model simulates the movement process of water in the unsaturated zone, and comprehensively considers the interaction of matric potential and gravity potential, so as to derive the instantaneous water demand prediction value of different soil layers; the acquisition of regional slope information relies on the elevation measurement unit, which is usually composed of laser radar or high-precision GPS receiving mechanism carried by unmanned aerial vehicle, through collecting dense point cloud data of ground surface and constructing digital elevation model, using surface fitting algorithm to calculate the slope and slope direction of each point, and then combining with the geomorphic features such as gully and platform obtained by field investigation to carry out artificial correction, finally generating a high-reliability slope distribution model.

[0051] The collection of vegetation index information is realized by multi-spectral sensor, the sensor platform can choose satellite remote sensing or low-altitude unmanned aerial vehicle system, which regularly scans the target area to collect spectral reflection data in visible light, near-infrared and short-wave infrared bands; using these data to calculate normalized vegetation index, enhanced vegetation index and other indicators, these indicators have strong correlation with leaf area index and biomass of vegetation, through the analysis of spectral reflection characteristics, the water demand of vegetation in different growth stages can be estimated; the multi-parameter environment perception model further integrates historical ecological data, including long-term observed soil moisture sequence, vegetation coverage change record and meteorological disaster event library.

[0052] Based on the environmental change rules extracted from historical data, a trend model of each parameter changing with time is established using curve fitting technology. The trend model is trained by machine learning algorithms such as support vector machine or random forest to identify key patterns such as drought stress and soil degradation. Real-time data collected in the present are compared with these patterns, and a multi-parameter prediction algorithm is used to generate a probability distribution of the environmental state in the future. Finally, the system determines whether to start the ecological restoration process according to the preset threshold set of critical conditions. The critical condition is usually set as a comprehensive state where the soil moisture is continuously lower than the wilting coefficient of plants, the vegetation index shows severe degradation, and the slope characteristics indicate the risk of water and soil loss. Only when all parameters meet the trigger conditions at the same time, the system will activate the subsequent delivery control mechanism.

[0053] Embodiment 2: refer to Figure 3 The structure of the independent delivery path constitutes the physical execution basis of the ecological restoration method, and the core of its design is to create a closed pipeline network completely isolated from natural precipitation. The pipeline system is made of high-density polyethylene material, and its burial depth is accurately calculated to avoid the influence of dramatic fluctuations in surface temperature. Meanwhile, all interfaces are sealed by fusion welding technology to completely prevent the intrusion of external precipitation or pollutants. The restoration medium delivered inside the system is dynamically selected according to the instructions output by the multi-parameter environmental perception model. It can be single water, or a suspension of nutrient solution or soil conditioner in a specific ratio. These media are stored in impermeable underground tanks, and their variable flow in the pipeline is realized by controllable pressure generated by multi-stage centrifugal booster pumps. The speed of the pump body is accurately controlled by a variable frequency drive, thereby matching the differentiated needs of flow rate and pressure in different restoration stages.

[0054] The delivery path is preset with multiple branch delivery sections during the network planning stage, and each branch section serves a geomorphic unit or a vegetation patch. Its pipe diameter and length are customized according to the estimated water demand of the region. The key is that a hydraulic one-way valve is installed at the end of each branch section. This valve opens relying on the flow pressure of the medium itself, and automatically closes relying on gravity when the flow weakens, thereby effectively preventing the backflow of the medium during slope changes or pumping intervals. The control strategy of the system continuously runs dynamic calculations. It obtains real-time data of soil porosity, initial moisture content, and medium viscosity, calculates the penetration rate of the restoration medium in the current environment per unit time using fluid dynamics principles, matches this rate with the demand curve output by the soil water demand model, and dynamically adjusts the output frequency of the variable frequency drive to accurately control the flow rate and total delivery time of the medium.

[0055] The flow direction of the medium in the pipeline network is not fixed, but is intelligently regulated according to real-time updated regional slope information; the slope distribution model generates a vector map identifying the gravity potential gradient in the entire dump area, and the control system preferentially guides the medium to the branch conveying section with the smallest included angle with the gravity vector direction; this preferential guidance strategy means that the medium will naturally flow to areas with lower terrain or steeper slopes, which usually have better natural diffusion conditions and higher water use efficiency, while this strategy significantly reduces the peak flow that the system needs to achieve in unit time by avoiding forced pumping to high areas, thereby reducing the pressure on the booster pump and energy system.

[0056] The slope direction recognition strategy for determining whether the delivery target area is in a naturally diffusing advantage position relies on high-precision three-dimensional terrain data; this data is generated by unmanned aerial vehicle laser radar scanning and contains the precise coordinates of millions of ground points; the control system processes these point cloud data to first build a digital elevation model, and then generates a detailed slope aspect map through surface normal vector calculation; the layout axis data of the conveying pipeline has been entered into the system database, and the algorithm performs vector analysis on the spatial orientation of each pipeline to calculate the spatial included angle between its direction vector and the gravity vector (vertically downward direction); when the system identifies that the included angle of a certain pipeline is less than the preset 15-degree threshold, it is determined that the pipeline is in a naturally diffusing advantage position and is marked as a priority activation object; if in the current period, due to soil supersaturation or other environmental factors, all regions do not meet the diffusion condition, the system will not forcibly start delivery, but will place the task in the waiting queue, and after re-evaluating environmental parameters in the next data update period (usually one hour), it will decide whether to activate the delivery operation of the corresponding pipeline section.

[0057] Taking a large coal mine dump in the arid region of Northwest China as an example, the soil in the southern slope of the dump has been severely desertified due to long-term exposure, and this area is planned as the third-stage ecological restoration unit; the independent delivery path layout project was started in early spring, and the construction team first designed a completely closed main pipeline network based on geological exploration data; the main pipeline uses DN200 caliber high-density polyethylene pipe, buried 1.2 meters deep to avoid the effects of freezing and high temperature, and all pipeline interfaces are connected by electric fusion welding process to achieve full-sealed connection; the main pipeline is laid along the ridge line of the dump and extends six branch conveying sections to the north and south slopes; a hydraulic-controlled one-way check valve is installed at the end of each branch section to prevent gravity backflow of the slope medium.

[0058] The selection of the recovery medium is determined according to the soil detection report of the area: the soil organic matter content is less than 0.5%, and the pH value is alkaline, so the mixed solution of water and soil conditioner is used as the conveying medium; the conditioner is a compound formula of humic acid and ammonium phosphate, which is mixed with water in a ratio of 1:100 in the central blending tank and then injected into the main pipeline through a multi-stage centrifugal pump set; the pump set is equipped with a variable frequency control system, which can accurately adjust the medium flow speed within the range of 0.5-2.5 m / s according to the instructions. The core of the control strategy is to dynamically match the slope and flow rate, and a typical operation process is as follows: at 6 am, the environmental perception system updates the slope data of the south area: the slope distribution model shows that the angle between the B3 branch section (azimuth angle SW235°) and the gravity vector is only 8°, and the soil moisture sensor shows that the water content in this area is less than 12%; the system activates the B3 pipe section preferentially, and the variable frequency pump delivers the modified liquid at an initial flow rate of 1.2 m / s, while monitoring the pipeline pressure change and soil permeability data in real time; when the pressure drops suddenly, indicating that the soil is starting to saturate, the system automatically reduces the flow rate to 0.8 m / s, and part of the flow is diverted to the C1 branch section with an angle of 12°.

[0059] The slope direction recognition strategy runs continuously in this process: the elevation model generated by the unmanned aerial vehicle laser scanning shows that there is a natural gully terrain in the service area of the B3 pipe section, and its axis is highly consistent with the direction of gravity; the system recalculates the angle of each pipe section every 10 minutes, and when the sun causes the temperature of the slope to rise at noon, the angle of the D2 pipe section, which is southeast, decreases from 15° to 9°, and the system immediately includes D2 in the priority delivery sequence; it is worth noting that the F4 pipe section on the north slope has an angle greater than 20° due to rock outcrops, and the system marks it as low priority and only allows it to be activated after the work in other areas is completed. The whole delivery process shows adaptive characteristics: when the evaporation speed increases due to the increase in wind speed in the afternoon, the system prolongs the length of the intermittent period in time; when the temperature drops in the early morning of the next day, it compensates for the water loss by increasing the single delivery time of the B3 pipe section; this dynamic adjustment based on real-time environmental feedback significantly improves the utilization rate of the modified liquid compared to traditional flood irrigation, and there is no medium backflow or pipeline overpressure phenomenon.

[0060] Example 3: The permeability modeling algorithm used to predict the absorption rate of the medium constitutes the core decision-making mechanism of the intelligent transportation system. This algorithm continuously integrates real-time data streams from multiple sensors, including the medium flow rate in the pipeline, the initial concentration monitoring value at the tank outlet, remote sensing feedback of soil porosity, and multispectral imaging results of the current vegetation cover. After preprocessing, these heterogeneous data are input into a linked computing framework. This framework first performs physical modeling of the movement of the medium in the soil, using a water transport model to describe the capillary action and gravity-driven process of liquid in porous media, while coupling a plant root absorption model to quantify the rate at which vegetation extracts water. The root absorption model considers the negative pressure gradient generated by root zone density distribution and transpiration. By calculating the dynamic deviation between the infiltration depth per unit time and the total water demand of the region, the system can continuously adjust the predicted operation time period, so that the transportation plan always keeps in sync with the actual absorption capacity of the soil.

[0061] The execution of the intermittent delivery strategy relies on a dynamic flow control algorithm that takes three key parameters as input: the current system reserve capacity (obtained from tank level sensors and supply network pressure gauges), the environmental evaporation rate (calculated using Dalton's formula from temperature, humidity, and wind speed data provided by a weather station), and the soil saturation level (monitored by a network of buried humidity sensors). These parameters are integrated into a control function to set the pump start-up to stop time ratio within each delivery cycle. This flow ratio is strictly limited by the system's preset capacity control threshold to prevent over-extraction of resources. At the same time, the algorithm has adaptive capabilities, allowing automatic adjustment of the start-up and stop ratios at different stages of the operation (such as the initial soaking period and the continuous replenishment period) to adapt to changing environmental conditions.

[0062] The operation cycle is clearly divided into two alternating phases: an activation phase and an intermittent phase. The activation phase is initiated by the system resource management unit, which continuously monitors the real-time available capacity of all storage tanks and pipelines. Based on a priority algorithm, resources are allocated to the most urgently needed transport sections before pumping is executed. The duration of the intermittent phase is not fixed but dynamically adjusted according to the trend of ambient temperature changes. Temperature data is updated every five minutes from the meteorological sensor. When the temperature rises above the preset critical point, the system automatically extends the intermittent phase duration to effectively suppress excessive moisture evaporation loss by reducing the time the medium is exposed to the high-temperature atmosphere.

[0063] The mathematical decisions for the entire transport process rely on a core permeation rate estimation model, which quantifies the actual absorption per unit time using the following relationship:

[0064]

[0065] Where: characters represents the integrated infiltration flux under a specific soil-vegetation complex, whose physical meaning is the medium volume passing through the unit area of soil interface per unit time; character represents a conduction coefficient determined by soil texture, root structure and medium viscosity; character represents the total duration since the start of irrigation. The calculation result of this formula is used to dynamically refresh the predicted operation period, ensuring that the system decision is always based on the accurate estimation of the current infiltration capacity.

[0066] Take the west slope area of a coal mine dump in a northwest arid region as an example, which is selected as the demonstration area for infiltration modeling and intermittent delivery. The soil in this area is mainly sandy loam, and the vegetation coverage is less than 15%. After the system starts, the infiltration modeling algorithm begins to continuously receive multi-source data streams: the electromagnetic flowmeter returns the flow rate data of the modified liquid in the pipeline in real time, the concentration sensor monitors the concentration of humic acid and water in the mixed liquid, the soil porosity probe feeds back the porosity changes at different depths, and the multispectral imager updates the vegetation coverage distribution map every week. These data are input into the linkage computing framework, the water movement model simulates the motion trajectory of the liquid between soil particles according to the porosity data and flow rate information, and the plant root absorption model calculates the potential absorption capacity of the root layer combined with the vegetation coverage data. By comparing the dynamic deviation between the actual infiltration amount and the total regional water demand per unit time, the system predicts that the current operation needs to continue for 85 minutes to reach the target infiltration depth.

[0067] The dynamic flow control algorithm starts operation at the same time, which obtains real-time parameters from three channels: the current storage capacity of 68% returned by the ultrasonic liquid level meter of the water storage tank, the evaporation rate of 2.1 mm / h calculated from the temperature and humidity data transmitted by the weather station, and the soil saturation of 42% at a depth of 20 cm displayed by the underground humidity sensor network. The algorithm substitutes these parameters into the control function to calculate that the current delivery cycle should use a start-stop time ratio of 3:1, i.e., every 9 minutes of pumping needs to be paused for 3 minutes. This flow ratio strictly follows the capacity warning line set by the system (forced hibernation when the storage capacity is less than 20%), and the algorithm reserves adaptive adjustment space. When the evaporation rate climbs to 3.5 mm / h at noon, the system automatically adjusts the start-stop ratio to 2:1 to shorten the exposure time. The operation cycle is clearly divided into two stages: in the activation period, the resource management unit allocates the available capacity to the three delivery sections of the west slope first, and the centrifugal pump delivers the modified liquid at a flow rate of 1.8 m / s, and the pressure sensor monitors the pressure fluctuations at the end of the pipeline network in real time; when entering the intermittent period, all pump groups immediately stop running, but the monitoring system is still working, which dynamically adjusts the waiting time for the next activation according to the ground temperature curve returned by the temperature sensor. When the monitoring system detects that the ground temperature reaches 47℃ at 2 pm, the system extends the intermittent period from the standard 3 minutes to 5 minutes, so that the micro-crust layer formed on the soil surface reduces water evaporation loss.

[0068] The decision of the entire delivery process relies on the continuous estimation of the infiltration rate. Through the analysis of soil texture and root distribution data, the model found that the east side of the area has a small amount of Artemisia grass roots, and its conductivity coefficient is 30% higher than that of the bare area. The system dynamically adjusts the delivery priority of each pipe segment accordingly, and the single activation time of the east pipe segment is extended by 20%, while the bare area adopts a higher frequency short delivery mode. This differentiated scheduling based on real-time infiltration capacity allows the improved liquid to be more fully absorbed and utilized in the vegetation development area. When the surface temperature falls to 28°C as the sun sets, the system detects that the evaporation rate has dropped to 0.9 mm / h, and immediately cancels the intermittent extension strategy to restore the standard cycle. The next day's vegetation scanning data shows that the water use efficiency of the area using dynamic intermittent delivery is significantly higher than that of the traditional uniform irrigation area, and there is no surface runoff or soil compaction phenomenon.

[0069] In the embodiment 4, the job state recording strategy for suspending the delivery process is an important fault tolerance mechanism for the system to cope with resource fluctuations. The core of its design is to take a complete snapshot of the system state at the moment of interruption. The specific implementation process is as follows: when the system reserves monitoring unit detects that the liquid level of the storage tank or the pressure of the pipe network is below the safe operation threshold, the control center will immediately issue a suspension instruction, which triggers the data recording module to start working. The recording module first obtains the accurate boundary coordinates of the current working area from the geographic information subsystem. These coordinates are stored in the form of a polygon vertex sequence, clearly defining the physical range of the actual coverage of this round of work. At the same time, the electromagnetic flowmeter integrated on the delivery main pipeline uploads the cumulative total amount of the medium that has been put in, which is accurate to the degree and is bound with the corresponding area ID. The humidity sensor network buried in the soil of the working area synchronously uploads the average soil humidity change value at the moment of suspension, which is obtained by calculating the average change of each sensor reading relative to the initial value of the work. All the above data are bound with the timestamp generated by the high-precision clock, which is accurate to milliseconds and converted into the coordinated universal time standard format. These captured data points are combined into a structured job breakpoint state record, which contains data in four dimensions of regional spatial information, medium input amount, humidity change index, and time reference. This structured record is written into the non-volatile storage area through the system interface, and the storage area adopts a redundant array design to prevent data loss due to hardware failure, providing a unique and reliable state basis for subsequent job recovery.

[0070] The first action of the recovery process is to read the latest breakpoint record in the storage area when the external water source is replenished or the system reserve is automatically restored above the operating threshold; the system analyzes the record and extracts the "unfinished work area" set, which is obtained by calculating the spatial difference between the total area polygon of the planned work in this round and the worked area polygon; then, the resource assessment module estimates the medium compensation required to complete the remaining work according to the area of the unfinished area, the soil water deficit degree of the last round of work on it (i.e. the difference between the target humidity and the humidity at the time of interruption) and the historical infiltration rate; this estimated amount is immediately compared with the current real-time medium delivery rate of the system, which is a dynamic value depending on the pump station power and the pipe network pressure, if the current rate can safely deliver the required compensation within the allowed time window without triggering the reserve alarm again, the system determines that the execution condition is met.

[0071] The real-time acquisition of system reserve capacity is a continuous background process that continuously collects data on the total water demand of the current area and the water supply capacity through the data bus; the water demand is periodically refreshed by the environmental perception model, and the supply capacity is calculated by combining the tank level sensor, water inlet flow meter and pump station state signal; the system compares the two values in real time and determines whether to activate or resume delivery operation according to the preset reserve capacity threshold (which is usually set to 20% of the total capacity to meet sudden demand); the start-up authority of the entire delivery control sub-process is set to be exclusively controlled by the resource scheduling mechanism, which is an independent monitoring thread, only when it detects that all resource indicators are continuously stable above the safety line and sends a clear available signal, the main control system will send a start command to the execution unit.

[0072] Table 1: Breakpoint state record

[0073] Recorded Field Name Data Content Example Data Source and Explanation Area_ID Zone_B-12 A geospatially defined area code, uniquely identifying a specific work zone within the mine site, as defined by the GIS. Total Media Invested 12580L Cumulative reading from the electromagnetic flow meter, recording the total volume of media delivered to this zone since the start of the current operation up to the point of interruption. Average Change in Soil Moisture +8.6% The average of all soil moisture sensor readings at the point of interruption minus the baseline value before the start of the operation, representing the relative increase in moisture. Interruption Timestamp 2023-10-26T08:17:42.103Z Coordinated Universal Time generated by a high-precision clock, recording the precise time when the interruption command was issued. Incomplete Zone Geometry Description POLYGON((102.1352,35.1234,102.1365,35.1241,102.1371,35.1228,102.1352,35.1234)) The boundary of the unprocessed zone calculated through spatial computation, represented as a sequence of geographic coordinates. System Reserve Status 22% The percentage of the effective capacity of the storage tank relative to the total capacity as monitored by the system at the point of interruption.

[0074] Referring to Table 1, if the resource comparison result shows that the current capacity is insufficient to safely complete the remaining work, the system will not immediately retry, but will lower the task priority and delay to the next preset work period, and send a pending alarm to the monitoring system; the whole process ensures that ecological restoration work can be accurately resumed after interruption in resource-limited arid environments, avoiding waste and repeated work, and achieving resilient management of the ecological restoration process of the dump.

[0075] The completion criteria of the closed-loop recovery control are established on the basis of multi-source sensor data fusion and intelligent analysis, and the decision logic comprehensively considers three core indicators: soil moisture recovery rate, vegetation growth status, and statistical results of remaining untreated areas. The monitoring of soil moisture relies on a multi-point sensor network pre-buried underground. These sensors are arranged in a grid layout to cover the entire target area. They collect volumetric moisture content data at different depths every thirty minutes. After filtering and removing outliers, the raw data are compared with the initial drought state baseline value recorded in the system database. The initial state is defined as the average humidity value for the seven days before the start of ecological restoration work. The system does not only focus on the absolute increase in humidity, but also calculates a recovery rate over time. This rate is obtained by linearly fitting the humidity trend over the past seventy-two hours and comparing it with the expected rate curve provided by the plant growth model. Only when the actual rate consistently and stably falls within the expected value confidence interval and the fluctuation amplitude is less than the preset tolerance is this indicator considered to meet the standard.

[0076] The evaluation of vegetation growth status is achieved through a multispectral sensor mounted on a fixed observation tower or a drone. This sensor performs a full coverage scan of the target area every week, collecting spectral information in the visible and near-infrared bands. After atmospheric correction and radiometric calibration, the collected data are used to calculate the normalized vegetation index and the enhanced vegetation index. The system sets a growth standard based on the phenological characteristics of typical pioneer species in arid areas. This standard is not a single threshold, but a dynamically changing range. For example, it allows lower vegetation indices at the early stage of recovery, but requires the index value to consistently rise and eventually stabilize at a higher level in the later stage. The judgment logic checks whether the vegetation index obtained from the last three consecutive scans is higher than the minimum requirement of this dynamic standard and whether the trend has not significantly decreased, thereby confirming whether the vegetation is growing healthily and reaching a self-sustaining state.

[0077] The statistics of the remaining untreated area depend on high-resolution geographic information data, which is derived from periodic aerial remote sensing or satellite mapping; the system will compare the latest acquired orthophoto with the base map before the start of the operation at the pixel level, identify the treated area (obvious improvement in vegetation coverage) and the untreated area (still bare or low coverage) through image segmentation and classification algorithms; the statistical results are presented in the form of area percentage, combined with slope and aspect data, to analyze whether the untreated area is distributed in ecologically sensitive or engineering difficult special sections; the standard requirement of this index is that the total area of the untreated area is below the termination threshold set by the system, and its distribution no longer constitutes obvious ecological risk or soil erosion hazard. When the data of the above three indicators are independently evaluated to meet their respective preset thresholds, the central processing system will automatically trigger the ecological restoration completion judgment process; this process first freezes all ongoing data recording tasks, then generates a final operation report detailing the key parameters throughout the process from start to completion, and finally sends a command to the control system to clear all operation records and temporary cache data in this cycle, and resets the system state to standby mode, waiting for the next possible start signal.

[0078] The entire ecological restoration method also includes a secure communication layer that ensures the secure transmission of environmental monitoring data through encryption technology; the encryption process uses a key stream generation algorithm based on a chaotic system to generate a pseudo-random sequence through a nonlinear dynamic system, and performs nested operations with randomly distributed key stream coefficients to generate a one-time pad encryption key; the encrypted monitoring data is transmitted to the central processing system through a physically isolated dedicated communication interface, and the central system holds the corresponding decryption key to decrypt and restore the data; the decrypted pure monitoring data is sent to the analysis module to finally generate medium delivery instructions to drive the pump station and valve.

[0079] It should be noted that, in this text, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0080] Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for ecological restoration of a coal mine dump in an arid region, characterized in that, The ecological restoration method comprises: A multi-parameter environment perception model containing soil humidity, soil composition, surface temperature, vegetation index and regional slope data is established, and whether the initial condition for starting ecological restoration is reached is determined based on the multi-parameter environment perception model; At least one of water, nutrient solution or soil conditioner is selected as a restoration medium, and the restoration medium is quantitatively delivered to a predetermined restoration area through an independent delivery path isolated from natural precipitation; The restoration area is formed by a delivery pipe network independently arranged in structure, and the water flow direction is dynamically determined according to the current regional slope data, and the pipe section in the position of natural diffusion advantage is preferentially started to deliver. During the delivery process, the target operation time period is determined according to the predicted medium absorption rate, and intermittent delivery is performed on the selected delivery pipe section, and the average flow rate in each operation cycle is controlled to be not more than the current system reserve capacity. When the reserve is insufficient, the delivery process is suspended and the completed operation time and area information are recorded, and after the reserve is recovered, the remaining operation process is continued until the closed-loop restoration control is completed. The construction process of the multi-parameter environment perception model comprises dynamic coupling calculation of arid region climate data and topographic change trend, continuous collection of soil humidity information by a multi-point sensing device, derivation using a water migration model and a soil porosity equation to obtain a predicted value of instantaneous water requirement; The regional slope information is obtained by a height measurement unit to calculate multi-point height data, and is corrected in combination with actual landform characteristics to form a slope distribution model; Vegetation index information is obtained by periodic scanning of a multi-spectral sensor, and vegetation coverage data is estimated according to spectral reflectance characteristics to estimate water requirement in the growth period; The multi-parameter environment perception model fits a curve based on the environmental change law in historical ecological data, and performs multi-parameter prediction in combination with current collection results to determine whether the critical condition for starting ecological restoration is reached.

2. The ecological restoration method of the arid coal mine spoil site according to claim 1, characterized in that, The structural arrangement of the independent delivery path comprises a closed pipe system completely isolated from natural precipitation in a physical sense, the restoration medium delivered in the closed pipe system is selected from water, nutrient solution or soil conditioner, and controllable variable flow is realized by a multi-stage booster pump; The delivery path is provided with a plurality of branch delivery sections, each branch section has one-way delivery performance to avoid medium backflow; In the control strategy, the penetration rate of the restoration medium in the current environment per unit time is matched with the current water requirement of the soil to dynamically select the flow rate and delivery time; The medium flow direction is adjusted in real time according to the regional slope information, and the branch section in the gravity flow advantage direction is preferentially guided to improve the medium utilization efficiency and reduce the flow peak value per unit time.

3. The ecological restoration method of the arid coal mine spoil site according to claim 2, characterized in that, The slope direction identification strategy for determining whether the delivery target area is in the position of natural diffusion advantage comprises vector analysis on the delivery pipe arrangement direction of the dump in three-dimensional terrain data, taking the gravity vector as a reference, calculating the included angle between the delivery pipe axis and the gravity vector to determine the most favorable medium diffusion direction; When the included angle is less than a set angle threshold, it is considered that the delivery section has natural diffusion capacity; if all regions do not meet the condition, a waiting state is entered, and whether to activate is evaluated again after the environmental parameter is updated in the next cycle.

4. The ecological restoration method of the arid coal mine spoil site according to claim 3, characterized in that, The infiltration modeling algorithm for predicting the medium absorption rate includes a linkage calculation of the medium flow rate, initial concentration, soil porosity, and current vegetation coverage, uses a water migration model and a plant root absorption model for modeling, dynamically adjusts the predicted operation period by calculating the deviation between the infiltration amount per unit time and the total regional water demand.

5. The ecological restoration method of the arid coal mine spoil site according to claim 4, characterized in that, The intermittent delivery strategy adopts a dynamic flow control algorithm, sets the start-stop ratio in the delivery cycle according to the current system reserve capacity, the environmental evaporation rate, and the soil saturation degree, and the flow ratio is limited by the capacity control threshold and can be automatically adjusted at different operation stages. The operation cycle is divided into an activation period and an intermittent period, the activation period is executed after the system resource management unit allocates the available capacity, and the intermittent period is adjusted according to the environmental temperature change trend to prevent excessive evaporation of water.

6. The ecological restoration method of the arid coal mine spoil site according to claim 5, characterized in that, The operation state recording strategy for pausing the delivery process includes synchronously recording the current operated region, the total amount of medium put in, the average soil humidity change, and the timestamp of the current operation time point, the recording information forms an operation breakpoint state structure, and is written into the system data storage area. When the reserve is recovered, the previous breakpoint information is first read, and the required compensation amount is estimated according to the unfinished operation region, the amount of medium put in, and the soil water shortage degree, and is compared with the medium delivery rate to determine whether it has the execution ability. If the continue execution condition is met, the corresponding delivery pipe segment operation process is directly resumed, otherwise it is delayed to enter the next cycle.

7. The ecological restoration method of the arid coal mine spoil site according to claim 6, characterized in that, The system reserve capacity is obtained by comparing the current regional water demand with the water source supply capacity in real time, and determining whether the delivery operation is allowed according to the reserve capacity threshold, and the delivery control sub-process is only executed under the condition that the system resource scheduling mechanism sends an available signal.

8. The ecological restoration method of the arid coal mine spoil site according to claim 7, characterized in that, The completion standard of the closed-loop recovery control is jointly decided based on the soil humidity recovery rate, the vegetation growth state, and the statistical results of the remaining untreated areas, and the termination threshold is satisfied as the judgment basis; The soil humidity is detected by a multi-point sensing device in real time, and compared with the initial drought state to calculate whether the humidity improvement rate is stable and tends to the expected value; The vegetation growth state is determined by a multispectral sensor whether it reaches the growth standard; The untreated area is compared with the vegetation coverage difference before and after processing according to geographic information data; When the three indicators all reach the preset threshold, it is automatically determined that the ecological restoration is completed, and the operation record is cleared.

9. The ecological restoration method of the arid coal mine spoil site according to claim 8, characterized in that, The ecological restoration method further comprises: The environmental monitoring data is encrypted and transmitted through a key stream, the key stream is obtained through chaotic nested encryption and random distribution of key stream coefficients; The ciphertext monitoring data is transmitted to the central processing system for decryption based on the communication interface; The medium delivery instruction is generated according to the decrypted monitoring data.

Images