Garbage fullness detection and sound-light alarm dispatching method based on pressure sensor

Abstract

The application relates to the technical field of intelligent garbage collection management, and discloses a garbage overflow detection and sound-light alarm scheduling method based on a pressure sensor. The method collects a pressure decay observation sequence after each disposal event, separates a seepage liquid component through a double exponential decay model fitting, extracts a residual pressure component as an effective solid mass pressure contribution value, compares the value with an overflow threshold value after accumulation to drive sound-light alarm; meanwhile, a date attribute template is constructed based on a historical effective solid accumulation rate, an overflow time is predicted, a pre-deployment cleaning and transportation scheduling scheme is generated, the rate template is dynamically corrected according to a real-time water content rate offset during operation, and the template is iteratively updated after the end of each day.

Description

Technical Field

[0001] This invention relates to the field of intelligent waste collection management technology, and more specifically, to a waste overflow detection and audible and visual alarm scheduling method based on pressure sensors. Background Technology

[0002] With the promotion of urban wet waste sorting and collection, wet waste collection bins have been widely deployed in residential and commercial areas. Wet waste has a high and fluctuating moisture content; after being placed in the collection bin, the liquid components continuously drain from the seepage outlet at the bottom of the bin under gravity, causing the pressure sensor readings to naturally decrease when no new waste is added. Current technology typically determines the overflow state directly based on the cumulative pressure value of the pressure sensor and constructs a typical daily filling rate curve template based on the total pressure change rate for collection scheduling. However, this method includes the seeped liquid components in the bin's filling volume. The significant initial pressure increase generated after the high moisture content waste is added is not deducted after the seepage drains, causing the audible and visual alarms to be triggered before the bin volume is full, resulting in premature false alarms. At the waste disposal scheduling level, the template constructed based on the total pressure change rate actually reflects the net rate of change in total pressure, including the leachate discharge effect. When changes in consumption structure on special days such as holidays cause significant differences in moisture content distribution compared to ordinary days, the proportional relationship between the net rate of change in total pressure and the effective solid accumulation rate shifts, resulting in a systematic deviation in the overflow time predicted based on the total pressure template, and a misalignment between the pre-deployed waste disposal scheduling plan and the actual overflow rhythm. Summary of the Invention

[0003] This invention provides a method for detecting and scheduling overflow of waste based on pressure sensors and sound and light alarms, which solves the technical problems in related technologies where wet waste collection bins cannot accurately distinguish the contribution of solid waste and leachate to pressure, and are difficult to achieve accurate overflow detection and forward-looking collection scheduling.

[0004] This invention discloses a method for detecting and scheduling overflowing garbage and sound and light alarms based on pressure sensors, including: when a garbage disposal event is detected to be completed and the disposal port is closed, extracting pressure data segments within a preset observation period after the disposal port is closed, obtaining the pressure decay observation sequence after a single disposal, and obtaining the total pressure increment of this disposal. A double exponential decay model is fitted to the pressure decay observation sequence to obtain the amplitudes of the residual pressure component, the rapid seepage component, and the slow seepage component. The water content estimate is calculated based on the ratio of the sum of the amplitudes of the rapid seepage component and the slow seepage component to the total pressure increment. The residual pressure component is extracted as the effective solid mass pressure contribution value and added to the effective solid cumulative pressure value. The effective solid cumulative pressure value is compared with the solid waste overflow pressure threshold, and overflow determination and alarm are performed. Group historical delivery records by date attribute to generate templates for typical intraday effective solids accumulation rate curves and typical intraday moisture content time period distributions under each date attribute category; Based on the typical intraday effective solids accumulation rate curve template, the effective solids accumulation pressure value is calculated and predicted for each time period starting from the current moment, the predicted overflow time of each device is determined, and spatiotemporal clustering is performed on the devices predicted to overflow on the day and a pre-deployment and cleaning scheduling plan is generated. During daily operation, the real-time moisture content estimate is compared with the typical intraday moisture content distribution template to obtain the moisture content offset. When the moisture content offset exceeds the preset allowable range, the expected effective solid accumulation rate for subsequent periods is corrected based on the ratio of the solid component proportion of the real-time moisture content to the template moisture content, the predicted overflow time is recalculated, and the cleaning and dispatching plan is updated.

[0005] Furthermore, the dual exponential decay model takes the elapsed time after the injection port is closed as the independent variable and the pressure sensor reading as the dependent variable. It is composed of three parts: the residual pressure component, the rapid seepage exponential decay term with the rapid seepage time constant as the decay rate, and the slow seepage exponential decay term with the slow seepage time constant as the decay rate. The rapid leachate decay term corresponds to the process of free water being rapidly discharged from the surface of the waste and large pores under gravity, while the slow leachate decay term corresponds to the process of bound water and liquid gradually seeping out from the interior of the waste and small pores. The five parameters—the residual pressure component, the amplitude of the rapid leachate component, the rapid leachate time constant, the amplitude of the slow leachate component, and the slow leachate time constant—are fitted using the nonlinear least squares method. The fitting objective is to minimize the sum of squared residuals between the model's predicted values ​​and the measured pressure values ​​at each sampling time in the pressure decay observation sequence. The Levenberg-Marquardt algorithm is used to solve the problem, with the median value of each parameter's constraint range serving as the initial iteration starting point.

[0006] Furthermore, when performing the nonlinear least squares fitting, constraints are set for each attenuation parameter: the residual pressure component is constrained to be greater than zero and not exceed the pressure value at the end of the injection; the amplitudes of the rapid seepage component and the slow seepage component are both constrained to be greater than zero and their sum does not exceed the total pressure increment; and the rapid seepage time constant is constrained to be less than the slow seepage time constant.

[0007] Furthermore, the total pressure increment is the difference between the pressure value at the moment the inlet is closed and the baseline pressure value before the inlet is opened; The determination of a delivery event is when the delivery port changes from an open state to a closed state and the pressure value change before and after closing exceeds the delivery identification threshold. When the delivery port is opened or closed but the pressure value change does not exceed the delivery identification threshold, it is determined as an invalid opening or closing event and does not trigger subsequent pressure decay observation.

[0008] Furthermore, the execution of overflow determination and alarm includes: when the effective solid cumulative pressure value reaches the warning ratio of the solid waste overflow pressure threshold, generating a warning signal and driving the audible and visual alarm module to output a warning prompt; When the effective solid cumulative pressure value reaches or exceeds the solid waste overflow pressure threshold, an overflow confirmation signal is generated and the audible and visual alarm module is driven to output an overflow prompt. At the same time, the disposal port is locked to prevent new waste from being disposed of. The overflow signal, effective solid cumulative pressure value, moisture content estimate, equipment number, and timestamp are reported to the management backend. When the moisture content estimate exceeds the preset high moisture content warning value, a high moisture content marker is added to the reported data.

[0009] Furthermore, the generation of typical intraday effective solids accumulation rate curve templates under each date attribute category includes: within each group, the effective solids mass pressure contribution value of each day is accumulated sequentially according to the delivery time to obtain the intraday effective solids accumulation pressure time series of each day; Divide the day into several equal-length periods, and assign the effective solid mass pressure contribution value of each delivery within each day to the corresponding period for accumulation. Take the median of the effective solid cumulative pressure value of each period within the same group to obtain a typical intraday effective solid cumulative pressure curve template. The expected effective solids accumulation rate for each time period is obtained by subtracting the effective solids accumulation pressure values ​​of adjacent time periods in the typical intraday effective solids accumulation pressure curve template, thus forming a typical intraday effective solids accumulation rate curve template. Before taking the median, abnormal daily samples whose effective solid cumulative pressure values ​​at each time point deviate from the group mean by more than a preset standard deviation are removed.

[0010] Furthermore, the spatiotemporal clustering uses the geographical coordinates of the equipment and the predicted overflow time as clustering features. Before performing clustering, the geographical coordinates and the predicted overflow time are normalized respectively. The spatiotemporal clustering is performed using the K-means clustering algorithm. The number of clusters is determined by rounding up the ratio of the total number of predicted overflowing equipment on the day to the upper limit of the number of equipment that a single cleaning vehicle can serve in a single trip. Devices that are geographically adjacent and whose predicted overflow times are similar are grouped into the same cluster. Cleanup vehicles are assigned to each cluster and pre-deployment cleanup routes are planned. When planning the routes, the predicted overflow time of each device is used as a time window constraint, requiring vehicles to arrive at the location of the corresponding device before the predicted overflow time to complete the cleanup.

[0011] Furthermore, the method of correcting the expected effective solid accumulation rate in subsequent periods based on the ratio of the solid component proportion between the real-time moisture content and the template moisture content is as follows: The expected effective solids accumulation rate for subsequent periods in the template is multiplied by a correction factor to obtain the corrected expected effective solids accumulation rate. The correction factor is the ratio of the solids content corresponding to the average real-time moisture content of the current period to the solids content corresponding to the moisture content of the template. The solids content is one minus the corresponding moisture content value. When determining whether the moisture content deviation exceeds the preset allowable range, the average moisture content of multiple consecutive deliveries within the sliding window is compared with the template value of the typical intraday moisture content distribution. When the average moisture content deviation within the sliding window continuously exceeds the preset allowable range, the rate correction and the cleaning and transportation scheduling scheme are updated.

[0012] Furthermore, after the daily scheduling is completed, the sequence of effective solid mass pressure contribution values ​​and the sequence of moisture content estimates for each device on that day are compared with the template of the corresponding date attribute category to calculate the prediction error. When the prediction error exceeds the update threshold, the data of that day is included in the historical dataset of the corresponding date attribute category, and the template calculation is re-executed to obtain the iteratively updated template of the typical intraday effective solid accumulation rate curve and the template of the typical intraday moisture content time period distribution. After incorporating today’s data into the historical dataset, the samples of each day in the historical dataset are assigned weights based on time distance. Samples closer to the current date receive higher weights. Weighted calculations are used when calculating the median effective solid cumulative pressure and the mean moisture content.

[0013] This invention provides a garbage overflow detection and audible-visual alarm scheduling system based on a pressure sensor, comprising: The pressure decay sequence acquisition module is used to extract pressure data segments within a preset observation period after the release event is detected as complete and the release port is closed, to obtain the pressure decay observation sequence and the total pressure increment. The double exponential decay fitting module is used to perform double exponential decay model fitting on the pressure decay observation sequence to obtain the amplitude of the residual pressure component, the amplitude of the rapid seepage component and the amplitude of the slow seepage component, and to calculate the estimated water content. The overflow determination and alarm module is used to extract the residual pressure component as the effective solid mass pressure contribution value and add it to the effective solid cumulative pressure value, and compare the effective solid cumulative pressure value with the solid waste overflow pressure threshold to perform overflow determination and alarm. The template generation module is used to generate templates for typical intraday effective solid accumulation rate curves and typical intraday moisture content time period distributions for each date attribute category by grouping historical delivery records according to date attributes. The predictive scheduling module is used to calculate the predicted effective solids accumulation pressure value on a time-by-time basis based on the typical intraday effective solids accumulation rate curve template, determine the predicted overflow time of each device, perform spatiotemporal clustering on the devices predicted to overflow on the day, and generate a pre-deployment cleaning and transportation scheduling plan. The real-time correction module is used to compare the real-time moisture content estimate with the typical intraday moisture content distribution template during the day's operation to obtain the moisture content offset. When the moisture content offset exceeds the preset allowable range, the expected effective solid accumulation rate for subsequent periods is corrected based on the ratio of the solid component proportion of the real-time moisture content to the template moisture content, the predicted overflow time is recalculated, and the cleaning and transportation scheduling plan is updated.

[0014] This invention separates the seepage component from the total pressure increment by fitting a double exponential decay model to the pressure decay observation sequence after deployment. The accumulated value of the residual pressure component is used as the overflow determination criterion, ensuring that the trigger time of the audible and visual alarm corresponds to the moment the silo volume is actually filled with solids. This reduces the probability of premature false alarms caused by artificially high pressure values ​​due to seepage discharge. The date-specific attribute curve template constructed based on the effective solids accumulation rate is unaffected by the seepage decay effect. Using this template to predict the overflow time and generate a pre-deployment cleaning and transportation schedule reduces the systematic deviation in predicted overflow time caused by differences in moisture content. By introducing a typical intraday moisture content distribution template, the system senses real-time moisture content shifts during daily operation and corrects the expected effective solids accumulation rate for subsequent periods. This allows the predicted overflow time to be dynamically adjusted according to the actual moisture content characteristics of the day, and performs local replanning on associated transportation routes. Continuous iterative updates of the template enable the system to incorporate daily operational data into historical datasets, gradually adapting to long-term business changes and seasonal moisture content trends. Attached Figure Description

[0015] Figure 1 This is a flowchart of the garbage overflow detection and audible-visual alarm scheduling method based on a pressure sensor provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the BIN-A07 pressure decay observation sequence and the fitting curve of the double exponential model provided in the embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the parameter decomposition of the double exponential decay model provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the prediction and calculation of the effective solid cumulative pressure (before correction) of BIN-A07 provided in the embodiment of the present invention; Figure 5 This is a schematic diagram comparing the effective solid accumulation rate before and after moisture content correction (time periods 19-22) provided in an embodiment of the present invention; Figure 6This is a schematic diagram comparing the prediction curves of effective solid cumulative pressure before and after correction provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the effective solid accumulation rate template (time period 16-22) for a typical weekend day provided in an embodiment of the present invention; Figure 8 This is a schematic diagram comparing the typical weekend moisture content template provided in this embodiment of the invention with the actual measured moisture content of the day. Figure 9 This is a schematic diagram comparing various pressure indicators of BIN-A07 under full overflow conditions, provided in an embodiment of the present invention. Detailed Implementation

[0016] Due to its high and fluctuating moisture content, wet waste continuously leaks liquid components from the bottom seepage outlet under gravity after being placed in the collection bin. This causes the pressure sensor readings to naturally decrease when no new waste is added. Current technology directly uses the cumulative pressure value of the pressure sensor to determine the overflow state, including the leaked liquid components in the filling volume. This means that the large initial pressure increase generated when high-moisture-content waste is added is not deducted after the seepage is discharged, triggering audible and visual alarms before the bin volume is full, resulting in premature false alarms due to artificially high pressure values. Furthermore, at the collection and scheduling level, the typical daily filling rate curve template constructed based on the total pressure change rate in current technology actually reflects the net rate of change in total pressure including the seepage discharge effect, rather than the true accumulation rate of effective solids. When changes in consumption patterns on special dates such as holidays lead to significant differences in moisture content distribution compared to ordinary days, the proportional relationship between the net rate of change in total pressure and the effective solid accumulation rate shifts. This results in a systematic deviation in the predicted overflow time based on the total pressure template, with alarms being triggered too early on days with high moisture content and too late on days with low moisture content. Consequently, the pre-deployed waste disposal scheduling plan is out of sync with the actual overflow schedule.

[0017] According to an embodiment of this invention, a method for detecting and scheduling overflowing waste based on a pressure sensor and an audible and visual alarm is provided. It should be understood that the method of this embodiment is executed collaboratively by a local controller deployed in the waste collection bin and a management backend server. A pressure sensor is installed at the bottom of the waste collection bin to continuously collect the pressure value inside the bin. An opening / closing status detection device is installed at the disposal port to monitor the opening and closing status of the disposal port. A leakage outlet is provided at the bottom of the bin to discharge liquid components. An audible and visual alarm module is installed on the outside of the bin to output light and voice prompts. All devices maintain data connection with the management backend through a communication network.

[0018] At least one embodiment of the present invention discloses a method for detecting and scheduling overflowing garbage based on a pressure sensor and an audible and visual alarm, such as... Figure 1 As shown, it includes the following steps: Step 1: Collect pressure decay observation sequences after a single deployment; Pressure sensors continuously collect pressure values ​​inside the chamber while simultaneously monitoring the opening and closing status of the dispensing port. Once a dispensing event is detected as complete and the dispensing port is closed, the pressure value at the moment of port closure is recorded as the dispensing termination pressure value. Pressure data segments within a preset observation period after port closure are extracted to obtain a pressure decay observation sequence after a single dispensing event. Simultaneously, the dispensing termination pressure value is subtracted from the baseline pressure value before the dispensing port opens to obtain the total pressure increment for this dispensing event.

[0019] It should be noted that the aforementioned preset observation duration refers to the length of the time interval during which pressure data is continuously recorded after the discharge port is closed. The value of the preset observation duration should cover the main decay process of liquid components in wet waste discharged from the seepage outlet at the bottom of the silo. For example, it can be set to a fixed value within the range of ten to thirty minutes after the discharge port is closed. In actual deployment, the specific value of the preset observation duration can be calibrated according to the orifice diameter of the seepage outlet in the silo and the seepage characteristics of typical wet waste.

[0020] It should be noted that the aforementioned determination of a delivery event refers to the situation where the delivery port changes from an open state to a closed state, and the pressure change before and after closing exceeds the delivery identification threshold. This delivery identification threshold is a pre-calibrated fixed value used to distinguish between valid delivery and invalid opening / closing actions of the delivery port. The specific value of the delivery identification threshold is determined based on the pressure sensor range and the empty pressure fluctuation range of the silo. When the delivery port opens or closes but the pressure change does not exceed the delivery identification threshold, it is determined to be an invalid opening / closing event, and subsequent pressure decay observation is not triggered.

[0021] Step 2: Perform a double exponential decay model fitting on the pressure decay observation sequence to obtain decay parameters and water content estimates; A double exponential decay model was applied to the pressure decay observation sequence, and the model form is as follows: in, This refers to the elapsed time after the delivery port is closed. For a moment Pressure sensor readings, This represents the residual pressure component after seepage has completed. To determine the magnitude of the rapid exudate, The time constant of the rapid seepage component. The magnitude of the slow seepage. The time constant for the slow seepage component. This represents an exponential function. The five decay parameters are fitted using a nonlinear least squares method to obtain their numerical values. Furthermore, the objective of the nonlinear least squares fitting is to minimize the model's predicted value. The sum of squared residuals between the measured pressure values ​​at each sampling time in the pressure decay observation sequence and the above nonlinear least squares problem is solved using the Levenberg-Marquardt algorithm, with the median value of each parameter constraint range as the initial iteration starting point.

[0022] The estimated moisture content of this product was calculated based on the attenuation parameters. The calculation method is as follows: in, This represents the total pressure increment obtained in step 1 for this application. The magnitude of the rapid seepage component. With slow exudate amplitude The sum represents the pressure contribution corresponding to the liquid components discharged during the seepage process, and the magnitude of the rapid seepage component. With slow exudate amplitude The sum divided by the total pressure increment yields the liquid component percentage, which is used as an estimate of the water content. It should be noted that the above moisture content estimates are... In the calculation, the molecule With denominator Both are in the dimension of pressure, and their dimensions are consistent, so the results are... It is a dimensionless ratio and can be directly used for the rate correction calculation in subsequent step 7.

[0023] It should be noted that the above double-exponential decay model includes two decay terms: a rapid leachate component and a slow leachate component. This is because the discharge process of liquid components in wet waste involves two leachate mechanisms with different rates. The rapid leachate component corresponds to the rapid discharge of free water from the waste surface and macropores under gravity. The time constant of the rapid leachate component... The amount of slow leachate is relatively small. The slow leachate component corresponds to the gradual seepage of bound water and liquid from tiny pores within the waste; the time constant of the slow leachate component is relatively small. Relatively large.

[0024] In this embodiment of the application, in order to improve the fitting stability under the condition of limited observation time, a constraint range is set for each attenuation parameter when performing nonlinear least squares fitting. Specifically, the constraint... Greater than zero and not exceeding the pressure value at the end of the deployment, constrained. and Both are greater than zero and their sum does not exceed the total pressure increment. ,constraint Less than This is to maintain a clear distinction between the physical meanings of the rapid and slow exudate components. By employing these constraints, parameter degradation during the fitting process is reduced, resulting in attenuation parameters with well-defined physical meanings.

[0025] Step 3: Calculate the effective cumulative solid pressure value and perform overflow judgment and alarm; extract the residual pressure component. As the effective solid mass pressure contribution value for this deployment, the effective solid mass pressure contribution value is accumulated into the effective solid cumulative pressure counter to obtain the current effective solid cumulative pressure value in the warehouse. .

[0026] Effective solid cumulative pressure value Compared with the pre-calibrated solid waste overflow pressure threshold Compare the effective solid cumulative pressure values. Reaching the solid waste overflow pressure threshold When the warning ratio is reached, a warning signal is generated and the audible and visual alarm module outputs a flashing yellow light and a warning voice prompt. The aforementioned warning ratio is a pre-set fixed value, such as a solid waste overflow pressure threshold. Eighty percent of the effective solid cumulative pressure value An alarm is triggered when the pressure value exceeds this ratio. (The alarm is triggered when the effective solid cumulative pressure value...) Reaching or exceeding the solid waste overflow pressure threshold When overflow occurs, a full overflow confirmation signal is generated, driving the audible and visual alarm module to output a solid red light and a full overflow voice prompt. Simultaneously, the disposal opening is locked to prevent further waste disposal. The overflow signal and the effective cumulative solid pressure value are then considered. Moisture content estimate The device number and timestamp are reported to the management backend.

[0027] It should be noted that the above-mentioned solid waste overflow pressure threshold It was obtained through pre-calibration. The calibration method involved gradually filling the silo with a known mass of low-moisture solid waste sample, recording the corresponding pressure sensor readings when the silo was full, and using this reading as the solid waste overflow pressure threshold. Solid waste overflow pressure threshold It reflects the pressure level when the volume of the container is actually filled with solid material, rather than the pressure level corresponding to the total mass containing liquid components.

[0028] It should be noted that the aforementioned effective solid cumulative pressure counter is reset to zero after each cleaning operation. The completion of cleaning is determined by the management backend receiving a confirmation signal from the cleaning personnel via the dispatch terminal, or by the pressure value of the storage compartment dropping to within the empty storage baseline for a preset period of time.

[0029] In this embodiment of the application, when the estimated moisture content obtained by fitting is... When the moisture content exceeds the preset high moisture content warning value, a high moisture content marker will be added to the reported data so that the management backend can identify the abnormal moisture content status of the current period in subsequent scheduling calculations. The aforementioned high moisture content warning value is a pre-set fixed threshold. The value of the high moisture content warning value is determined based on the typical wet waste moisture content distribution characteristics of the storage facility's service area. When the estimated moisture content... If the moisture content exceeds the high moisture content warning value, the moisture content of the current application is considered to be significantly high.

[0030] Step 4: Generate a template for the typical intraday effective solids accumulation rate curve by grouping by date attribute; The system extracts long-term, sequential delivery records for each device from the historical database in the management backend. Each record includes a delivery timestamp, an effective solid mass pressure contribution value, an estimated moisture content value, and a corresponding date attribute tag. These date attribute tags identify the type of date to which the delivery record belongs, such as weekday, weekend, or public holiday.

[0031] The deployment records of each device are grouped according to the date attribute label. Within each group, the effective solid mass pressure contribution value of each day is accumulated sequentially according to the deployment time to obtain the intraday effective solid cumulative pressure time series for each day. The intraday effective solid cumulative pressure time series of each day in the same group are aligned according to the intraday time, and the median of the effective solid cumulative pressure value at the corresponding time of each day in the same group is taken to obtain the typical intraday effective solid cumulative pressure curve template under each date attribute category. Differential calculation is performed on the typical intraday effective solid cumulative pressure curve template to obtain the typical intraday effective solid cumulative rate curve template under each date attribute category.

[0032] It should be noted that the above-mentioned alignment by intraday time refers to dividing the day into several equal-length time periods, and accumulating the effective solid mass pressure contribution value of each release within each day into the corresponding time period, thereby mapping the release sequences of different dates onto a unified time period coordinate for comparison and median taking.

[0033] Furthermore, the above differential calculation refers to subtracting the effective solids cumulative pressure values ​​of adjacent time periods in the typical intraday effective solids cumulative pressure curve template. The resulting difference between each time period is the expected effective solids cumulative rate within that time period. The dimensions of the expected effective solids cumulative rate are the same as... Similarly, it reflects the typical increment of the solid waste mass pressure contribution within a unit time period, and is used in step 6 to estimate the effective solid cumulative pressure value for each time period and in step 7 to set the rate correction benchmark.

[0034] In this embodiment of the application, in order to reduce the interference of individual abnormal day data on the template of the effective solid accumulation rate curve of typical day, before taking the median of the effective solid accumulation pressure value at the corresponding time of each day in the same group, abnormal day samples whose effective solid accumulation pressure value at each time deviates from the mean of the group by more than a preset standard deviation multiple are first removed, and then the median of the remaining samples is taken.

[0035] Step 5: Generate a template for typical intraday water content distribution over time periods; For each device, the historical delivery records under each date attribute category are summarized by intraday time period, the estimated moisture content of each delivery is calculated, the average moisture content of each intraday time period is calculated, and the typical intraday moisture content time period distribution template under each date attribute category is obtained.

[0036] It should be noted that the above method of dividing intraday time periods is consistent with the time period division used in step 4 for alignment, so as to ensure that the template of typical intraday effective solid accumulation rate curve and the template of typical intraday water content time period distribution correspond in the time period dimension.

[0037] Step 6: Predict the overflow time of each device based on the template and generate a pre-deployment cleaning and dispatching plan; Before the target scheduling day begins, a typical intraday effective solids accumulation rate curve template matching the date attributes is selected for each device. This is based on the current effective solids accumulation pressure value of each device. Based on the expected solids accumulation rate for each time period in the typical intraday effective solids accumulation rate curve template, the predicted effective solids accumulation pressure value is calculated and extrapolated backwards from the current time period. The predicted effective solids accumulation pressure value is then used to calculate the first time the solid waste overflow pressure threshold is reached. The predicted overflow time of the device is determined by the predicted overflow time. The device overflow time is then sorted by the predicted overflow time to obtain the daily device overflow time prediction schedule.

[0038] Furthermore, the above-mentioned time-period extrapolation refers to using the current effective cumulative solid pressure value. Starting from the current time period, the expected effective solids accumulation rates for each subsequent time period in the typical intraday effective solids accumulation rate curve template are sequentially superimposed onto the effective solids accumulation pressure value. The system obtains the predicted effective cumulative solids pressure value for each future time period, until the predicted effective cumulative solids pressure value first reaches the solid waste overflow pressure threshold. The corresponding time is the predicted overflow time.

[0039] Based on the daily equipment overflow time prediction schedule, and constrained by the working hours of the waste collection vehicles, the equipment predicted to reach overflow time for the day is spatiotemporally clustered. Spatiotemporal clustering uses the equipment's geographical coordinates and predicted overflow time as clustering features to group the equipment. The geographical coordinate dimension ensures that equipment within the same cluster can be continuously collected by a single vehicle spatially, while the predicted overflow time dimension ensures that equipment within the same cluster is suitable for processing in the same collection batch temporally. It should be noted that geographical coordinates and predicted overflow time are features with different dimensions; therefore, both geographical coordinates and predicted overflow times need to be normalized before performing spatiotemporal clustering to eliminate the impact of dimension differences on cluster distance calculation. Equipment with adjacent geographical locations and similar predicted overflow times are grouped into the same cluster. Waste collection vehicles are assigned to each cluster, and pre-deployment collection routes for each vehicle are planned. The daily pre-deployment collection scheduling plan is obtained and distributed to each scheduling terminal.

[0040] Furthermore, the above spatiotemporal clustering is performed using the K-means clustering algorithm. The number of clusters is determined by rounding up the ratio of the predicted total number of overflowing devices to the maximum number of devices that a single waste collection vehicle can serve in a single trip, so that the number of devices corresponding to each cluster does not exceed the single-trip service capacity of a single vehicle.

[0041] In this embodiment of the application, when planning the pre-deployment and cleaning routes of each vehicle, the predicted overflow time of each device is also used as a time window constraint, requiring the vehicle to arrive at the location of the corresponding device before the predicted overflow time to complete the cleaning, so as to reduce the waiting time for cleaning after the warehouse is full.

[0042] Step 7: During the day's operation, the predicted overflow time and removal route are corrected in real time based on the moisture content offset; During the actual operation on the day, each device continuously executes the successive pressure decay fitting process from step 1 to step 3, and reports the effective solid mass pressure contribution value and moisture content estimate obtained from each delivery to the management backend.

[0043] The management backend calculates the deviation between the real-time effective solids cumulative pressure value of each device and the predicted value of the typical intraday effective solids cumulative rate curve template in step 6. At the same time, it compares the estimated moisture content delivered by each device in real time with the typical moisture content in the typical intraday moisture content time period distribution template of the corresponding date attribute category and the current time period in step 5 to obtain the moisture content offset.

[0044] When the moisture content deviation exceeds a preset allowable range, the template value of the typical intraday effective solids accumulation rate curve for subsequent periods is corrected based on the moisture content deviation ratio. Specifically, let the average real-time moisture content for the current period be... The template moisture content in the typical intraday moisture content distribution template corresponding to the date attribute category and time period is . The expected effective solids accumulation rate for a subsequent period in a typical intraday effective solids accumulation rate curve template is: The corrected expected effective solid accumulation rate for that period is... Calculate using the following formula:

[0045] The above formula means: the expected effective solid accumulation rate. At the moisture content of the template The rate of accumulation of solid components per unit time under the given conditions, when the real-time moisture content is... Deviation from template moisture content At that time, the actual proportion of solid components in each delivery was determined by Become Therefore, the expected effective solid accumulation rate will be... Multiplying by the ratio of the two yields the corrected expected effective solid accumulation rate. It should be noted that, and All are dimensionless moisture content ratios defined in step 2. and The dimensions are consistent, and the correction factor is a dimensionless ratio. and With identical dimensions, the dimensions of each term in the above formula match. The aforementioned preset allowable range is a pre-defined threshold range for moisture content deviation. When the absolute value of the moisture content deviation exceeds this threshold, a rate correction is triggered. This threshold is determined based on the historical moisture content fluctuation characteristics of the warehouse's service area. When the real-time moisture content... Moisture content higher than template hour, Less than If the correction factor is less than 1, the expected effective solids accumulation rate for subsequent periods will be lowered; when the real-time moisture content... Lower than the moisture content of the template When the correction factor is greater than 1, the expected effective solids accumulation rate for subsequent periods is adjusted upwards. Based on the corrected expected effective solids accumulation rate, the predicted overflow time of the affected equipment is recalculated, the corresponding audible and visual alarm module warning trigger time is updated synchronously, local replanning is performed on the associated pre-deployed waste removal route segments, and the updated dispatch instructions are sent to the dispatch terminals of the corresponding vehicles.

[0046] Furthermore, the aforementioned recalculation of the predicted overflow time based on the revised expected effective solids accumulation rate refers to using the currently measured effective solids accumulation pressure value of each device. Starting from this point, the revised expected effective solid accumulation rate for subsequent time periods will be used as the starting point. The values ​​are added sequentially until the predicted effective cumulative solids pressure value first reaches the solid waste overflow pressure threshold. The corresponding time is the corrected predicted overflow time. The calculation logic for the corrected predicted overflow time is the same as the time-by-time calculation method in step 6. The difference is that the rate value of subsequent time periods is determined by the expected effective solid accumulation rate. Replace with the corrected expected effective solid accumulation rate .

[0047] In this embodiment, to avoid frequent scheduling changes caused by a single moisture content deviation, when the moisture content deviation exceeds a preset allowable range, the average moisture content of multiple consecutive deliveries within a sliding window is compared with the template value of the typical intraday moisture content distribution, rather than being based solely on the estimated moisture content of a single delivery. Only when the average moisture content deviation within the sliding window continuously exceeds the preset allowable range is subsequent period rate correction and pre-deployed waste disposal route replanning triggered.

[0048] Step 8: Iteratively update the templates for the typical intraday effective solids accumulation rate curve and the typical intraday moisture content distribution over time based on the data of the day; After the daily scheduling is completed, the sequence of effective solid mass pressure contribution values ​​and moisture content estimates for each device are compared with the typical intraday effective solid accumulation rate curve template and typical intraday moisture content time period distribution template for the corresponding date attribute category, and the prediction error is calculated. When the prediction error exceeds the update threshold, the data for the day is included in the historical dataset of the corresponding date attribute category, and the template calculation process in steps 4 and 5 is re-executed to obtain the iteratively updated typical intraday effective solid accumulation rate curve template and typical intraday moisture content time period distribution template, which are used for overflow detection prediction and waste removal scheduling on subsequent scheduling days.

[0049] It should be noted that the aforementioned update threshold refers to a measure of the deviation between the actual data for the day and the predicted value of the typical intraday effective solids accumulation rate curve template. When the deviation exceeds the update threshold, the current typical intraday effective solids accumulation rate curve template is considered insufficiently representative and needs to be updated by incorporating new data. When the deviation does not exceed the update threshold, the existing typical intraday effective solids accumulation rate curve template and typical intraday water content time-period distribution template are maintained unchanged to avoid introducing unnecessary template changes from data within the normal fluctuation range.

[0050] In this embodiment of the application, in order to balance historical stability and response speed to recent changes in the updating of the effective solids accumulation rate curve template and the typical daily water content distribution template, after incorporating the current day's data into the historical dataset, each day's sample in the historical dataset is assigned a weight based on time distance. The closer the sample is to the current date, the higher the weight is. Weighted calculation is used when calculating the median effective solids accumulation pressure value and the mean water content, so that the effective solids accumulation rate curve template and the typical daily water content distribution template can gradually shift towards the characteristics of recent data, while preserving the statistical stability of long-term data.

[0051] This implementation addresses the issue of inflated pressure sensor readings in wet waste collection bins due to leachate discharge. After each disposal event, a double exponential decay model is applied to the pressure decay observation sequence to separate the rapid and slow leachate components from the total pressure increment. The residual pressure component after leachate discharge is extracted as the effective solid mass pressure contribution value. Since the effective solid cumulative pressure value used for audible and visual alarm determination only reflects the pressure level corresponding to the residual solid mass, rather than the total pressure value including the leachate liquid components, leachate discharge is no longer counted in the bin's filling volume in high-moisture-content waste disposal scenarios. This avoids premature false alarms where the bin is not full, and the triggering time of the audible and visual alarm module corresponds to the moment the bin volume is actually filled with solids.

[0052] At the waste collection and scheduling level, since the typical daily effective solids accumulation rate curve template is constructed based on the effective solids accumulation rate rather than the total pressure change rate, the attenuation effect of leachate discharge on the total pressure reading is no longer included in the rate estimation of the typical daily effective solids accumulation rate curve template. Therefore, the filling rate reflected by the typical daily effective solids accumulation rate curve template directly corresponds to the actual accumulation process of solid waste. When changes in consumption patterns on special days such as holidays cause the moisture content distribution to deviate from that of ordinary days, the typical daily effective solids accumulation rate curve template itself is not affected by changes in moisture content. The overflow time predicted based on the typical daily effective solids accumulation rate curve template will not be systematically shifted due to differences in moisture content, and the time arrangement of the pre-deployed waste collection and scheduling plan remains consistent with the actual overflow rhythm.

[0053] The introduction of typical intraday moisture content time-period distribution templates further enables the system to detect deviations between real-time moisture content and historical typical values ​​during daily operation. When the deviation exceeds the allowable range, the expected effective solids accumulation rate for subsequent periods is corrected, allowing the predicted overflow time and the trigger time of the audible and visual alarm module to be dynamically adjusted according to the actual moisture content characteristics of the day. Local replanning of associated pre-deployed collection routes ensures that the scheduling scheme continuously adapts to changes in the actual delivery situation during daily operation. The continuous iterative updates of the typical intraday effective solids accumulation rate curve template and the typical intraday moisture content time-period distribution template allow the system to feed new data acquired daily into the templates, gradually adapting to long-term trend changes brought about by business adjustments and seasonal moisture content variations.

[0054] The following is an example of an application of the present invention, such as Figure 2-9 As shown, the implementation process is as follows: A community's wet waste collection station has deployed a wet waste collection bin numbered BIN-A07. A pressure sensor with a range of 0-200 Pa is installed at the bottom of the bin, with a sampling interval of 5 seconds. A magnetic induction opening / closing detection device is installed at the disposal port. A 12mm diameter leakage outlet is located at the bottom of the bin, and an audible and visual alarm module is installed on the exterior of the bin. The management backend has accumulated 90 days of historical disposal records for this device, covering both weekdays and weekends. The dispatch date is March 15, 20XX (weekend), and the collection vehicle VAN-03 is responsible for dispatching this area on that day.

[0055] At 08:17:43 on March 15, 20XX, the dispensing port opening / closing detection device of BIN-A07 detected that the dispensing port changed from an open state to a closed state. The system read the baseline pressure value before the dispensing port opened as 73.2 Pa and the pressure value at the moment the dispensing port closed as 118.6 Pa. The difference between the two is the total pressure increment for this dispensing operation. Pa exceeds the deployment identification threshold of 8.0 Pa, and is therefore determined to be a valid deployment event. Subsequently, the system continuously captures pressure decay data for 15 minutes (preset observation duration) starting from the moment the deployment port closes, recording a sampling point every 30 seconds, and obtaining a total of 30 sampling points to form a pressure decay observation sequence.

[0056] Table 1. Pressure decay observation sequence (partial sampling points)

[0057] A double exponential decay model was applied to the pressure decay observation sequence formed by the above 30 sampling points. The system used the median value of each parameter constraint range as the initial iteration starting point of the Levenberg-Marquardt algorithm, constraining... Between 0 and 118.6 Pa, constraint and Both are greater than zero and their sum does not exceed 45.4 Pa, constraining... After the fit converges, the following decay parameters are obtained: Table 2. Fitting results of the double exponential decay model

[0058] Based on the fitting results, the estimated moisture content for this application was calculated:

[0059] Estimated moisture content for this application This means that the liquid component accounted for 65.4% of the total mass pressure contribution in this delivery. This value exceeds the system's preset high moisture content warning value of 0.60, and the system adds a high moisture content marker to the reported data.

[0060] The effective solid mass pressure contribution value of this release The Pa (i.e., the difference between the residual pressure component and the baseline pressure before delivery) is accumulated into the effective solids cumulative pressure counter of BIN-A07. Before this delivery, the effective solids cumulative pressure value of BIN-A07 was 62.4 Pa. After accumulation:

[0061] The system's pre-calibrated solid waste overflow pressure threshold Pa, the warning rate is 80%, corresponding to a warning trigger pressure of 96.0 Pa. Currently... Pa, which did not exceed the warning trigger pressure, and the audible and visual alarm module remained in normal operation. The system reported the effective solid cumulative pressure value of 78.1 Pa, the estimated moisture content of 0.654, the high moisture content marker, the equipment number BIN-A07, and the timestamp of March 15, 20XX, at 08:32:18 to the management backend.

[0062] Table 3 Results of the overflow status assessment after this deployment

[0063] The management backend extracts the delivery records for 26 weekend days over the past 90 days for BIN-A07 under the weekend date attribute from the historical database. The day is divided into 48 equal-length time periods (30 minutes each). The effective solid mass pressure contribution value for each weekend day is assigned to each time period and accumulated sequentially to obtain the intraday effective solid cumulative pressure time series for each day. After removing outlier day samples that deviate from the group mean by more than 1.5 standard deviations at any given time, the median of the remaining samples is taken to obtain a typical intraday effective solid cumulative pressure curve template for weekend days. Then, by differencing adjacent time periods, a typical intraday effective solid cumulative rate curve template is obtained.

[0064] Table 4. Template of typical intraday effective solid accumulation rate curves on weekends (partial time periods)

[0065] Synchronously, the historical water content estimates for BIN-A07 under the weekend date attribute are summarized by time period, and the average water content for each time period is calculated to obtain a typical daily water content distribution template for the weekend. The water content template values ​​for each time period use the same 30-minute time period division as in step 4.

[0066] Table 5. Typical Daily Moisture Content Distribution Template for Weekends (Partial Time Periods)

[0067] Before the scheduling began on March 15, 20XX, the management backend selected a typical daily effective solids accumulation rate curve template for BIN-A07 over a weekend. The effective solids accumulation pressure value for BIN-A07 at that time was used as the reference. Starting from Pa (the state at the start time of scheduling at 07:00), the expected effective solid accumulation rate is added sequentially from time period 16 until the predicted value first reaches [the target value]. Pa.

[0068] The calculation process for each time period (starting from time period 16): (End of session 16) (End of session 17) (End of session 18) (End of period 19) (End of period 20) (End of session 21) (End of session 22) Continuing to overlay subsequent time periods, the predicted effective solid cumulative pressure value first reached 120.3 Pa at the end of time period 25 (11:30–12:00), exceeding... Pa, therefore, the predicted overflow time for BIN-A07 is 12:00. The management backend performs the same calculation on the other devices in the area, summarizes and generates a daily device overflow time prediction schedule, and performs spatiotemporal clustering on spatially adjacent devices with similar overflow times. BIN-A07 and three other devices with predicted overflow times between 11:30 and 12:30 are grouped into the same cluster, assigned to VAN-03, and a pre-deployment cleaning route is planned and sent to the dispatch terminal.

[0069] At 09:15 on the same day, the management backend received the estimated moisture content values ​​for BIN-A07 for four consecutive discharges during time period 18 (08:30-09:00): 0.641, 0.658, 0.672, and 0.663. The average moisture content within the sliding window was:

[0070] The moisture content of the template in the corresponding time period 18 The moisture content offset is If the value exceeds the preset allowable limit of 0.08, a rate correction will be triggered.

[0071] Taking the rate correction in time period 19 as an example, the template rate Pa / time period, substituting into the correction formula:

[0072] The same correction was applied sequentially to subsequent time periods, resulting in a general downward revision of the expected effective solids accumulation rate. This was based on the measured effective solids accumulation pressure value of BIN-A07 at that time. Starting from Pa, the calculation is recalculated time-by-time, predicting that the effective solid cumulative pressure value will reach its first peak. Pa's time has been postponed to 13:30. The management backend updates the predicted overflow time of BIN-A07 from 12:00 to 13:30, and simultaneously updates the estimated warning trigger time of the audible and visual alarm module. It performs partial replanning on the pre-deployed cleaning route of VAN-03, moves the cleaning time window of BIN-A07 to the later stage, and sends the updated scheduling instructions to the scheduling terminal of VAN-03.

[0073] Table 6 Comparison of moisture content offset trigger rate before and after correction (partial time periods)

[0074] After the scheduling was completed on March 15, 20XX, the management backend compared the daily effective solid mass pressure contribution value sequence and moisture content estimate sequence of BIN-A07 with the typical intraday template for the weekend to calculate the prediction error. The overall moisture content for the day was higher than normal, and the average prediction error for each time period exceeded the update threshold. Therefore, the system included today's data in the weekend historical dataset and assigned weights based on time distance to the samples in the historical dataset, with samples closer to March 15th having higher weights. Steps 4 and 5 were re-executed to calculate the template, obtaining iteratively updated templates for the typical intraday effective solids accumulation rate curve and the typical intraday moisture content time period distribution for the weekend, for use in subsequent weekend scheduling.

[0075] The data flow throughout the entire implementation process exhibits a clear, step-by-step transmission logic: the pressure decay observation sequence collected in step 1 carries the original pressure time series data into step 2; the fitting output of step 2—the residual pressure component, the amplitude of the rapid seepage component, the amplitude of the slow seepage component, and the estimated water content W—flows to steps 3 (overflow determination), 5 (water content template construction), and 7 (real-time correction), respectively; the effective solid cumulative pressure value output in step 3 serves as both the starting point for time-by-time calculation in step 6 and the starting point for recalculating the predicted overflow time in step 7; the two types of templates constructed in steps 4 and 5 provide a rate prediction benchmark for step 6 and a water content offset comparison benchmark for step 7, respectively; the corrected rate value in step 7 replaces the template rate to re-drive the overflow time calculation, outputting updated scheduling instructions; step 8 feeds back all measured data for the day to the template calculation in steps 4 and 5, forming a closed loop from the day's operational data to the iterative update of the template, ensuring that the prediction benchmark for subsequent scheduling days continues to closely match the actual delivery patterns.

[0076] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A method for detecting and scheduling overflowing garbage based on a pressure sensor and an audible and visual alarm, characterized in that, Includes the following steps: Once a delivery event is detected as complete and the delivery port is closed, extract the pressure data segment within the preset observation period after the delivery port is closed, obtain the pressure decay observation sequence after a single delivery, and obtain the total pressure increment of this delivery. A double exponential decay model is fitted to the pressure decay observation sequence to obtain the amplitudes of the residual pressure component, the rapid seepage component, and the slow seepage component. The water content estimate is calculated based on the ratio of the sum of the amplitudes of the rapid seepage component and the slow seepage component to the total pressure increment. The residual pressure component is extracted as the effective solid mass pressure contribution value and added to the effective solid cumulative pressure value. The effective solid cumulative pressure value is compared with the solid waste overflow pressure threshold, and overflow determination and alarm are performed. Group historical delivery records by date attribute to generate templates for typical intraday effective solids accumulation rate curves and typical intraday moisture content time period distributions under each date attribute category; Based on the typical intraday effective solids accumulation rate curve template, the effective solids accumulation pressure value is calculated and predicted for each time period starting from the current moment, the predicted overflow time of each device is determined, and spatiotemporal clustering is performed on the devices predicted to overflow on the day and a pre-deployment and cleaning scheduling plan is generated. During daily operation, the real-time moisture content estimate is compared with the typical intraday moisture content distribution template to obtain the moisture content offset. When the moisture content offset exceeds the preset allowable range, the expected effective solid accumulation rate for subsequent periods is corrected based on the ratio of the solid component proportion of the real-time moisture content to the template moisture content, the predicted overflow time is recalculated, and the cleaning and dispatching plan is updated.

2. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The double exponential decay model takes the elapsed time after the injection port is closed as the independent variable and the pressure sensor reading as the dependent variable. It consists of three superimposed parts: the residual pressure component, the rapid seepage exponential decay term with the rapid seepage time constant as the decay rate, and the slow seepage exponential decay term with the slow seepage time constant as the decay rate. The rapid leachate decay term corresponds to the process of free water being rapidly discharged from the surface of the waste and large pores under gravity, while the slow leachate decay term corresponds to the process of bound water and liquid gradually seeping out from the interior of the waste and small pores. The five parameters—the residual pressure component, the amplitude of the rapid leachate component, the rapid leachate time constant, the amplitude of the slow leachate component, and the slow leachate time constant—are fitted using the nonlinear least squares method. The fitting objective is to minimize the sum of squared residuals between the model's predicted values ​​and the measured pressure values ​​at each sampling time in the pressure decay observation sequence. The Levenberg-Marquardt algorithm is used to solve the problem, with the median value of each parameter's constraint range serving as the initial iteration starting point.

3. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 2, characterized in that, When performing the nonlinear least squares fitting, constraints are set for each attenuation parameter: the residual pressure component is constrained to be greater than zero and not exceed the pressure value at the end of the injection; the amplitudes of the rapid seepage component and the slow seepage component are both constrained to be greater than zero and their sum does not exceed the total pressure increment; and the rapid seepage time constant is constrained to be less than the slow seepage time constant.

4. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The total pressure increment is the difference between the pressure value at the moment the inlet is closed and the baseline pressure value before the inlet is opened; The determination of a delivery event is when the delivery port changes from an open state to a closed state and the pressure value change before and after closing exceeds the delivery identification threshold. When the delivery port is opened or closed but the pressure value change does not exceed the delivery identification threshold, it is determined as an invalid opening or closing event and does not trigger subsequent pressure decay observation.

5. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The process of determining and alarming overflow includes: when the effective cumulative solid pressure value reaches the warning ratio of the solid waste overflow pressure threshold, generating a warning signal and driving the audible and visual alarm module to output a warning prompt; When the effective solid cumulative pressure value reaches or exceeds the solid waste overflow pressure threshold, an overflow confirmation signal is generated and the audible and visual alarm module is driven to output an overflow prompt. At the same time, the disposal port is locked to prevent new waste from being disposed of. The overflow signal, effective solid cumulative pressure value, moisture content estimate, equipment number, and timestamp are reported to the management backend. When the moisture content estimate exceeds the preset high moisture content warning value, a high moisture content marker is added to the reported data.

6. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The generation of typical intraday effective solids accumulation rate curve templates under each date attribute category includes: within each group, the intraday effective solids mass pressure contribution value of each day is accumulated sequentially according to the delivery time to obtain the intraday effective solids accumulation pressure time series of each day; Divide the day into several equal-length periods, and assign the effective solid mass pressure contribution value of each delivery within each day to the corresponding period for accumulation. Take the median of the effective solid cumulative pressure value of each period within the same group to obtain a typical intraday effective solid cumulative pressure curve template. The expected effective solids accumulation rate for each time period is obtained by subtracting the effective solids accumulation pressure values ​​of adjacent time periods in the typical intraday effective solids accumulation pressure curve template, thus forming a typical intraday effective solids accumulation rate curve template. Before taking the median, abnormal daily samples whose effective solid cumulative pressure values ​​at each time point deviate from the group mean by more than a preset standard deviation are removed.

7. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The spatiotemporal clustering uses the geographical coordinates of the equipment and the predicted overflow time as clustering features. Before performing clustering, the geographical coordinates and the predicted overflow time are normalized respectively. The spatiotemporal clustering is performed using the K-means clustering algorithm. The number of clusters is determined by rounding up the ratio of the total number of predicted overflow equipment on the day to the upper limit of the number of equipment that a single cleaning vehicle can serve in a single trip. Devices that are geographically adjacent and whose predicted overflow times are similar are grouped into the same cluster. Cleanup vehicles are assigned to each cluster and pre-deployment cleanup routes are planned. When planning the routes, the predicted overflow time of each device is used as a time window constraint, requiring vehicles to arrive at the location of the corresponding device before the predicted overflow time to complete the cleanup.

8. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, The method of correcting the expected effective solid accumulation rate in subsequent periods based on the ratio of solid component proportions between real-time moisture content and template moisture content is as follows: The expected effective solids accumulation rate for subsequent periods in the template is multiplied by a correction factor to obtain the corrected expected effective solids accumulation rate. The correction factor is the ratio of the solids content corresponding to the average real-time moisture content of the current period to the solids content corresponding to the moisture content of the template. The solids content is one minus the corresponding moisture content value. When determining whether the moisture content deviation exceeds the preset allowable range, the average moisture content of multiple consecutive deliveries within the sliding window is compared with the template value of the typical intraday moisture content distribution. When the average moisture content deviation within the sliding window continuously exceeds the preset allowable range, the rate correction and the cleaning and transportation scheduling scheme are updated.

9. The method for detecting and scheduling overflowing garbage based on a pressure sensor according to claim 1, characterized in that, After the daily scheduling is completed, the sequence of effective solid mass pressure contribution values ​​and the sequence of moisture content estimates of each device for the day are compared with the template of the corresponding date attribute category to calculate the prediction error. When the prediction error exceeds the update threshold, the data of the day is included in the historical dataset of the corresponding date attribute category, and the template calculation is re-executed to obtain the iteratively updated template of the typical intraday effective solid accumulation rate curve and the template of the typical intraday moisture content time period distribution. After incorporating today’s data into the historical dataset, the samples of each day in the historical dataset are assigned weights based on time distance. Samples closer to the current date receive higher weights. Weighted calculations are used when calculating the median effective solid cumulative pressure and the mean moisture content.

10. A waste overflow detection and audible-visual alarm scheduling system based on a pressure sensor, used to execute the waste overflow detection and audible-visual alarm scheduling method based on a pressure sensor as described in any one of claims 1 to 9, characterized in that, include: The pressure decay sequence acquisition module is used to extract pressure data segments within a preset observation period after the release event is detected as complete and the release port is closed, to obtain the pressure decay observation sequence and the total pressure increment. The double exponential decay fitting module is used to perform double exponential decay model fitting on the pressure decay observation sequence to obtain the amplitude of the residual pressure component, the amplitude of the rapid seepage component and the amplitude of the slow seepage component, and to calculate the estimated water content. The overflow determination and alarm module is used to extract the residual pressure component as the effective solid mass pressure contribution value and add it to the effective solid cumulative pressure value, and compare the effective solid cumulative pressure value with the solid waste overflow pressure threshold to perform overflow determination and alarm. The template generation module is used to generate templates for typical intraday effective solid accumulation rate curves and typical intraday moisture content time period distributions for each date attribute category by grouping historical delivery records according to date attributes. The predictive scheduling module is used to calculate the predicted effective solids accumulation pressure value on a time-by-time basis based on the typical intraday effective solids accumulation rate curve template, determine the predicted overflow time of each device, perform spatiotemporal clustering on the devices predicted to overflow on the day, and generate a pre-deployment cleaning and transportation scheduling plan. The real-time correction module is used to compare the real-time moisture content estimate with the typical intraday moisture content distribution template during the day's operation to obtain the moisture content offset. When the moisture content offset exceeds the preset allowable range, the expected effective solid accumulation rate for subsequent periods is corrected based on the ratio of the solid component proportion of the real-time moisture content to the template moisture content, the predicted overflow time is recalculated, and the cleaning and transportation scheduling plan is updated.

Images